You filed your LLC. Now what?

In this episode of Crab Cakes & Commerce, we break down the 6 critical steps every new business owner needs to take within the first 30 days of filing their LLC in Maryland. From getting your EIN to setting up your business bank accounts, registering with the Comptroller, and securing your registered agent — these are the foundational moves that set your business up for long-term success.

Whether you're a first-time entrepreneur or just getting your paperwork in order, this guide will help you avoid common mistakes and make sure nothing falls through the cracks.

⏱️ Chapters
0:00 — Intro
0:10 — Welcome & Sports Recap (Ravens, Caps, LSU)
4:09 — Personal Updates & Fundraising (Baltimore Running Festival)
6:11 — Today's Topic: First 30 Days After Filing
7:16 — The Discussion Begins
7:54 — Step 1: Getting Your EIN
10:21 — Step 2: Business Bank Accounts
13:50 — Step 3: Register with the Comptroller
15:53 — Step 4: Business Licenses
17:48 — Step 5: Register Trade Names (DBA)
18:55 — Step 6: Registered Agent
20:49 — Wrap-Up & Closing

🔗 Links
• Support Maryland Special Olympics — Mike's fundraising page (https://support.somd.org/fundraiser/7481416)
• Newsletter: mdebaugh.substack.com
• Subscribe for weekly episodes on small business in Maryland

📋 The 6 Steps Covered

Get your EIN (Employer Identification Number) — free and easy through the IRS

Open 3 business bank accounts — profit, operating expenses, and tax

Register with the Maryland Comptroller for sales & use tax

Research and obtain required business licenses (state, county, industry)

Register any trade names / DBA if operating under a different name

Set up a registered agent to protect your personal information
If you found this helpful, drop a comment with what you'd like to hear next — and hit subscribe for more small business content every week.

#CrabcakesAndCommerce #LLC #SmallBusiness #MarylandBusiness #Entrepreneurship #BusinessTips #EIN #RegisteredAgent

Subscribe now

Share

Leave a comment

Let’s invent someone. Call him Ray. He’s assembled from enough real owners that you’ll probably recognize him.

Ray owns a commercial HVAC service company in Anne Arundel County. Eleven trucks, fourteen employees, service contracts with office parks along the Route 100 corridor, and a phone number that’s been the same since 1994. The business throws off real cash. Ray’s customers like him. His techs mostly like him.

Ray is 66. His kids are a nurse in Richmond and a software developer in Denver, and neither one wants the trucks. Ray’s exit plan, as far as anyone can tell, is “someday.”

Across the county, someone else is sitting at a kitchen table with a business plan, a new LLC, and a pitch deck for a commercial HVAC company that doesn’t exist yet. No customers, no trucks, no techs, no track record, and in a few months, no savings.

Both of them want to own a business. Only one of them is starting from zero, and it’s the one who chose to.

Share

What Most People Believe

When you picture becoming a business owner, you probably picture starting something. The origin story. The garage. The first customer. That idea is baked into how we talk about entrepreneurship.

Here’s the belief, stated fairly: starting from scratch is the real way to become an owner. Buying is for private equity firms and people with trust funds.

It’s a reasonable thing to believe. It’s also, for a certain kind of person, the more expensive and riskier path.

What the Numbers Say

Start with the survival numbers, because they’re the whole argument.

The Bureau of Labor Statistics tracks every new business establishment in the country from the year it opens. Of the establishments that opened in the year ending March 2019, about 79% made it through year one, and about 51.5% made it to year five. More recent groups did somewhat better (the 2021 group was at 59% after five years), but the shape is always the same. Roughly one in five is gone in the first year, and something like four or five in ten are gone by year five.

A business that’s been operating for fifteen years has already gotten through that filter. That doesn’t make it safe. It means the question has changed from will anyone pay for this? to will they keep paying for it under a new owner? The second question is a lot easier to answer, because you can look at the books.

Now the supply side. According to the Census Bureau’s Annual Business Survey, 51% of owners of U.S. employer businesses were 55 or older. That figure comes from data collected in 2018, so the group it describes is older now. The businesses are real, the owners are aging, and most of them don’t have a Denver software developer eager to take over.

And the market is there. BizBuySell’s Q2 2026 Insight Report counted 2,117 small businesses sold in the quarter, with a median sale price of $349,250 and a median cash flow of $155,921. Buyers paid an average of about 2.7 times cash flow. Put more plainly, a typical small business sold for a bit under three years of what it pays its owner. In the same report, 45% of owners planning to sell named retirement as the main reason, and only 14% had gotten a professional valuation.

Read that last number again. Almost half the sellers are retiring, and very few have had the business professionally valued. That’s the gap: a lot of sellers who aren’t well prepared, and buyers who show up prepared tend to get better deals.

What It Looks Like With Real Numbers

Let’s build a hypothetical deal from those medians. This is an illustration, not a real transaction. Medians of price and cash flow don’t come from the same business.

• Purchase price: $349,250

• Seller’s discretionary earnings (cash flow to one owner-operator): $155,921

• Equity injection at 10%: about $34,925

• SBA 7(a) loan for the other 90%: about $314,325

• Payment on a 10-year loan at an illustrative 10% rate (the SBA caps variable rates on loans over $250,000 at prime plus 3%): about $4,150 a month, or roughly $49,850 a year

After the loan payment, that leaves around $106,000 a year to pay yourself, pay taxes, and keep a cushion. If you pay yourself $70,000, the business still covers its debt payments about 1.7 times. That clears the 1.25x minimum lenders will be looking for under the new SBA rules (more on those below).

Now picture the startup version of the same year. Most startups don’t pay their founders anything in year one. You’d be paying for the truck, the insurance, the website and the first hire out of savings, while you wait to learn whether anyone will ever pay you.

That’s the core case for buying: the day you close, you own something that already makes money. It’s a very different starting point from hoping it will someday.

The Advantages of Buying

Cash flow on day one.

• Customers, contracts, receivables and a payroll that’s already running. You’re not building revenue from nothing. You’re trying not to lose it.

A track record a lender can underwrite.

• A bank can’t lend on a startup’s projections nearly as comfortably as on a business with three years of tax returns. Under the SBA’s updated rules, first-time acquisitions are underwritten on historical or adjusted trailing earnings, not forecasts. That’s a real advantage for a buyer, because a startup doesn’t have any earnings history to show.

People and systems already in place.

• Trained employees, vendor accounts, pricing that’s been tested, a phone number customers already know. Those take years to build, and at a small business most of them never show up on the balance sheet.

A seller who wants you to succeed.

• A retiring owner who carries part of the price as a seller note has money riding on your success. BizBuySell found that 29% of owners planning to sell expect to offer seller financing. A seller who’s also a creditor has every reason to stick around for the transition and make the introductions.

You can study it before you buy it.

• With a startup you learn how it works after you’ve spent the money. With an acquisition you can read three years of financials, customer concentration, and employee turnover before you sign anything.

The Fine Print

Here’s the part the “buy a boring business and get rich” crowd tends to skip. Buying reduces some risks and brings in new ones.

You can overpay, and it’ll hurt for a decade.

• A startup that fails costs you your savings. An acquisition you overpaid for costs you your savings, and you’re still on the hook for a ten-year loan. The price multiple matters less than whether the earnings are real, and plenty of owners mix personal expenses into the business, pay family members off the books, or have one great year they want you to treat as normal. The seller’s numbers are a claim, not a fact, until your accountant has checked them.

The goodwill may leave with the seller.

• If the customers are loyal to Ray personally, some of them will leave when Ray does. The same goes for the lead tech who’s been waiting fifteen years for a promotion that now goes to you. Customer concentration is the classic risk. If one office park is 40% of revenue, you’re buying one relationship with a business attached.

You’re signing a personal guarantee.

• SBA loans generally require a personal guarantee from significant owners. If the business fails, the debt follows you home. That’s the biggest difference between this and a salaried job, and you should look at it before the adrenaline kicks in.

The rules are about to change.

• The SBA issued SOP 50 10 8.1, which takes effect October 1, 2026 and applies to loans that receive their SBA loan number on or after that date. According to lender and industry summaries, first-time acquisitions will need:

  • a 1.25x debt coverage minimum, up from 1.15x;

  • a seller note on full standby can cover no more than half of the required equity injection, and must stay in place for 36 months;

  • an independent quality-of-earnings review on deals of $3 million or more, excluding real estate.

In practice, a seller note can no longer cover most of your down payment. You’ll need real cash of your own.

Buying well is its own job.

• Deal sourcing, due diligence, negotiation, lender packages, the lease, the transition plan. It’s months of work before you own anything, and a deal can fall apart in week eleven.

The Legal Angle

A few things that determine whether you’re buying a business or buying someone else’s problems:

Asset purchase or stock purchase.

• In an asset purchase you generally pick which assets and liabilities you take on. In a stock or membership-interest purchase you get the whole entity, including its history. Most small deals are asset purchases for that reason. If the SBA is financing a partial buyout, check how the new rules treat your structure before you choose one.

The lease can kill the deal.

• If the business depends on its location, the landlord’s consent to assign the lease is often the real closing condition. Read the lease before you fall in love with the business.

Taxes and liabilities that follow the business.

• Confirm there are no unpaid sales, payroll or withholding taxes, and no pending claims, liens or employment disputes that could land on the new owner. Then put it in writing with representations, warranties and indemnification that actually have money behind them.

A transition and non-compete agreement.

• How long the seller stays, what they’re paid to stay, and a reasonable agreement that they won’t open a competing shop across the street. Enforceability varies with the facts and the state, so have it drafted by someone who does this for a living, in your state.

Who This Is Actually For

Buying isn’t for everyone who’s tired of their job. It’s a strong option for a specific profile:

• You have real capital. Enough for the equity injection, closing costs, working capital, and six to twelve months of personal expenses, without betting your house.

• You have operating experience. You’ve managed people, read a P&L, and dealt with an angry customer. The business already works. Your job is to keep it working, then improve it.

• You have infrastructure. An accountant, a lawyer, a lender relationship, and ideally advisors who’ve seen deals go sideways.

• You care more about owning than inventing. If the dream is a new idea, start something. If the dream is running a good business, you can buy one that already exists.

BizBuySell’s buyer data points the same way: 46% of prospective buyers describe themselves as “corporate refugees”. These are people who’ve spent a career running someone else’s operation and would now like to own one.

What I’d Actually Do

Decide what you’re buying before you look at listings.

• Industry, size range, how far you’ll drive, and what you’re good at. A buyer who says “anything profitable” ends up looking at everything and closing on nothing.

Get pre-qualified before you get excited.

• Talk to an SBA lender now, before the new rules take effect, so you know your real equity requirement and your debt-coverage limit. Your budget is set by the lender’s 1.25x floor, not the asking price. If conventional financing falls short, Maryland’s MSBDFA lists buying an existing business among the uses it finances, with a focus on socially and economically disadvantaged owners.

Look off-market, too.

• Listed businesses are competitive. Retiring owners with no exit plan often aren’t listed at all. The industry association directory, your accountant’s client list, and the owner you’ve bought from for ten years are all places to find deals.

Pay for diligence as if your house depends on it, because it might.

• Get a quality-of-earnings review or at least a forensic accounting review, go through customer concentration, and meet the key employees before closing. Budget for this up front. It’s cheaper than a bad deal.

Put the transition in writing.

• Seller-note terms, how long the seller stays and what they’re paid, customer introductions, and what happens if key customers leave in the first year.

Make the last check the one against your own finances.

• Model the worst realistic year, not the median one. If you could survive the worst year without the guarantee reaching your house, you’re ready to make an offer.

The person at the kitchen table might build something great. Some of them do. But across the county, Ray’s phone number has worked since 1994. His trucks are paid off, his customers pay on net-30, and his biggest problem is that nobody has knocked on his door.

Someone should. The hard part, getting a business to year fifteen, is already done.

Leave a comment

Sources

• U.S. Bureau of Labor Statistics, Business Employment Dynamics: Establishment survival, Table 7 (total private sector)

• U.S. Bureau of Labor Statistics, The Economics Daily: 1-year survival rates for new business establishments by year and location

• U.S. Census Bureau: Business Owners’ Ages: Over Half of U.S. Business Owners Were Age 55 and Over (2019 Annual Business Survey, 2018 data)

• BizBuySell: Insight Report, Q2 2026

• U.S. Small Business Administration: Information Notice, Issuance of SOP 50 10 8.1

• FunderIntel: SBA SOP 50 10 8.1: 7 Things Brokers Need to Know Before October 1

• Jaken Finance Group: SBA SOP 50 10 8.1 Changes for October 2026

• Lendio: Current SBA Loan Interest Rates, September 2026

• Maryland Department of Commerce: Maryland Small Business Development Financing Authority (MSBDFA)

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax or financial advice, and I’m not your lawyer. I’m a guy from Baltimore with a spreadsheet and a law degree, in that order. SBA rules change as of October 1, 2026, and loan rates moved with the Fed’s September hike, so confirm current terms with your lender and have your own attorney and CPA review any deal before you sign.

Coffee’s on. Let’s go through the books.

This was a “read the fine print” kind of week. The SBA handed small businesses more borrowing room than they’ve had in years — attached to more paperwork than they’ve had in years. A fight over the Inner Harbor’s future that’s been simmering since 2022 finally landed in front of a judge. A batch of new federal research quietly confirmed something worth knowing if your bank’s app has gotten a lot slicker while your actual banker keeps changing. And down in Glen Burnie, the clock is running out for businesses still dealing with a condemned office tower.

On the ledger this week: the SBA doubled how much you can borrow across its two flagship loan programs combined, to $10 million, while tightening who qualifies at the same time; a $900 million plan to redevelop Harborplace is now tied up in a lawsuit over the pavilions; businesses displaced by the Empire Towers structural emergency in Glen Burnie have eleven days left to apply for federal disaster loans; new San Francisco Fed research says the more a bank leans on AI, the less of its lending goes to small businesses; Anne Arundel County’s own numbers show a local labor market cooler than the statewide picture suggests; and a new round of state grants opened for anyone selling fresh food in a food desert.

1. The SBA just doubled how much you can borrow — and made it harder to qualify

Effective July 4, 2026, the SBA doubled the combined borrowing limit across its 7(a) and 504 loan programs from $5 million to $10 million per borrower — up to $5 million through each program. The bigger deal is buried in the mechanics: the agency also decoupled the two programs, so a 7(a) balance no longer eats into how much 504 financing you can carry. That matters if you’re financing a building and working capital as separate pieces of the same project, which is exactly what a lot of manufacturing and services expansions look like. Manufacturers get extra room on top of that, since a 504 loan for a separate project doesn’t count against their 7(a) cap.

Before you get excited, read what came with it. Kiplinger’s rundown of 2026 SBA changes — real changes, not the vague “signaled” language in some of the AI-summarized versions floating around — lays out the trade: the SBA eliminated automated credit scoring for small 7(a) loans, meaning more of them now get manually underwritten; collateral requirements expanded on loans under $500,000; ownership eligibility tightened to require full U.S. citizenship or national status; and riskier deals, like acquisitions and startups, now carry a 10% equity injection requirement that didn’t exist before. On the friendlier side, the SBA is waiving upfront guaranty fees on 7(a) loans up to $950,000 for manufacturers through the rest of FY2026.

What I’d do: If you’re financing an acquisition, a building-plus-equipment expansion, or anything that used to bump against the old $5 million ceiling, call your SBA lender this week — the math on pairing 7(a) and 504 just changed in your favor. But go in ready for a real underwriting conversation, not a credit-score rubber stamp: pull two years of financials, confirm every owner’s citizenship documentation is current, and budget for the 10% equity injection if you’re buying rather than building.

2. Harborplace’s $900 million rebuild just landed in court

A group of Baltimore residents and business owners sued the city this week to block demolition of the Harborplace pavilions, arguing the city’s approval process for the redevelopment violated the Baltimore City Charter and zoning requirements. The plan under fire is developer MCB Real Estate’s roughly $900 million project — five new buildings, including two 200,000-square-foot commercial structures, an 8,500-square-foot retail building with a 2,000-seat amphitheater, and two residential towers (32 and 25 stories) holding about 900 units combined, replacing the pavilions on 4.5 acres at the Inner Harbor. David Tufaro, founder of Terra Nova Ventures and one of the plaintiffs, has also filed to have Harborplace designated a historic landmark, telling reporters, “the idea of demolishing them in light of the historical significance is a major reason not to do it.”

Construction was supposed to break ground this fall. It’s worth remembering this deal has already survived one court fight — MCB acquired the site through a 2022 court-approved process — so a lawsuit alone doesn’t mean the project dies. But it does mean the timeline you’ve been planning around just got a lot less certain, and nobody’s said when a judge will rule.

What I’d do: If your business is downtown or Inner Harbor-adjacent and you were counting on redevelopment-driven foot traffic, contractor work, or a fall groundbreaking to plan around, don’t lock in commitments tied to that calendar yet. Watch for a preliminary injunction ruling before you commit capital, staffing, or lease terms to a timeline that assumed construction starts on schedule.

3. Businesses displaced by the Glen Burnie tower emergency have 11 days left for disaster loans

Back in July, Empire Towers — a large office building in Glen Burnie — was condemned after a structural integrity emergency forced an evacuation and displaced the tenants inside, law firms among them. On September 22, Governor Moore announced SBA approval of Economic Injury Disaster Loans for small businesses, agricultural cooperatives, and nonprofits in Anne Arundel County affected by the closure. The terms are real money: up to $2 million per business, up to 30-year repayment, 4% interest for businesses (3.625% for nonprofits), and no payments due for the first 12 months. You don’t need physical damage to your own space to qualify — economic injury from the disruption is enough, and the loans can cover payroll, fixed debts, accounts payable, and ordinary operating expenses.

An SBA Disaster Loan Outreach Center opened September 24 at the Anne Arundel County Office of Emergency Management in Glen Burnie to help with applications.

What I’d do: The application deadline is Friday, October 9, at 5 p.m. — not “sometime this fall.” If you were a tenant, a vendor to a tenant, or a nearby business that lost foot traffic from the closure, you likely qualify even without physical damage of your own. Get to the outreach center in Glen Burnie or apply online before the 9th; a $2 million loan at 4% with a year of no payments is worth the paperwork even if you end up borrowing less than the max.

4. Banks bragging about AI are lending less to small businesses

A September 21 economic letter from the Federal Reserve Bank of San Francisco found something worth sitting with: banks that lean harder into AI put a smaller share of their lending toward small and medium businesses, not a larger one. Across the sample, high-AI-adoption banks directed 12% of their portfolios to SME lending, versus 21% for low-AI banks — and the gap holds even comparing small banks to other small banks (13.3% versus 21.4%). The researchers’ explanation is straightforward: AI is good at processing “hard data” — credit scores, financial statements — and less useful for the relationship-based “soft information” that small-business lending has traditionally leaned on. High-AI banks were also more profitable (0.38 percentage points higher return on assets) but carried more problem loans.

This isn’t a Maryland-specific finding, but it’s not disconnected from you either. Nationally, a U.S. Bank survey published by ABA Banking Journal in July found 75% of small-business owners already use AI themselves — mostly for marketing (56%), data analysis (51%), and automation (44%). So the businesses adopting AI on one side of the counter aren’t necessarily the businesses benefiting from AI adoption on the other side of it.

What I’d do: If your bank’s app has gotten noticeably better while your relationship banker keeps getting reassigned or replaced, that’s not a coincidence, and it’s not necessarily good news for your credit access. Before your next renewal or line-of-credit review, ask directly whether the bank still underwrites relationship-based exceptions to the numbers — the research suggests fewer of them do than five years ago. If your business runs on a banker who knows your story, not just your credit score, a community bank or credit union relationship may be worth more to you than the rate difference.

5. Anne Arundel’s own numbers tell a cooler story than the state’s

The county’s latest “At a Glance” data from the Anne Arundel Economic Development Corporation shows unemployment at 3.6% as of July — down 0.3 points from a year earlier, a fine headline number. Underneath it: total quarterly average employment sits at 287,091 jobs, down 1,886 (-0.7%) from the first quarter of 2025. The sector detail explains where that’s coming from. Construction added 862 jobs and manufacturing added 197 over the year — both growing. Professional and business services lost 1,289 jobs and government lost 1,239, doing most of the damage to the county total.

Real estate and retail tell a mixed story too. August home sales averaged $627,671, up 1.2% year over year, with the median at $507,000, up 1.3% — but only 615 homes sold, down 12.1% from the year before, meaning prices held while volume shrank. Commercial vacancy is genuinely mixed: retail vacancy actually tightened to 3.0%, and flex space improved slightly to 7.6%, while industrial vacancy crept up to 8.7% and office space held flat at 7.4%.

What I’d do: If you’re a professional-services business in Anne Arundel County — law, consulting, back-office, admin — the county’s own numbers back up what you’re likely feeling; you’re not imagining a harder year. If you’re retail-facing, vacancy actually improved, which means a lease renewal conversation might go better than you’d expect. Either way, don’t extrapolate from the low headline unemployment number alone; the sector breakdown underneath it is the part that tells you whether your specific business is in the group that’s growing or the group that’s shrinking.

6. A new round of grant money opened for anyone selling fresh food in a food desert

The state’s NourishMD grant program opened its second application round on September 22, offering $25,000 to $150,000 per project — up to $3.67 million total, split between state and federal rural health funds — to retail businesses that expand access to fresh, healthy food. Eligible operators include grocery stores, corner stores, farmers markets, farm stands, food co-ops, and CSAs, but there’s a real eligibility gate: you need to serve a community with a Fresh Food Gap Score of 65 or higher and dedicate at least 20% of your inventory or sales space to healthy food.

The program runs in two tiers. Tier 1 closes November 2; Tier 2 stays open until February 4, 2027, for anyone who misses the first window. Award notifications aren’t expected until April 2027, so this is patient-capital money, not a bridge loan for next month’s rent.

What I’d do: If you run or are planning a grocery, corner store, or farm stand in an underserved area, check your community’s Fresh Food Gap Score before you do anything else — that’s the gate you have to clear, and DHCD’s program page has the details on how to look it up. Then get your application in for the November 2 tier rather than waiting for February; earlier tiers typically mean less competition for the same pot of money.

The week ahead

• Thursday, October 1 — the Worker Freedom Act (SB 417) takes effect for Maryland employers, restricting mandatory meetings on political or religious matters.

• Friday, October 9, 5 p.m. — deadline to apply for Empire Towers EIDL disaster loans (item 3 above).

• Tuesday, October 20 — Maryland’s September 2026 statewide employment report is released.

• Tuesday–Wednesday, October 27–28 — the Fed’s next FOMC meeting, decision Wednesday afternoon.

• Monday, November 2, 5 p.m. — NourishMD Tier 1 application deadline (item 6 above).

• Tuesday, November 3 — Maryland’s general election.

• Sunday, November 15 — the FAMLI Declaration of Intent window closes for employers weighing a private paid-leave plan.

Two open threads, still open: the Comptroller’s appeal of the digital ad tax ruling, filed September 15, still has no hearing date on the calendar as of this morning. And DHCD’s 4% Small Business Direct Loan round, which closed September 17, finally updated its own status page to “Closed” this week — but there’s still no announcement of who got funded. I’ll keep watching both.

See you next Monday.

Sources

• SBA Doubles Cumulative 7(a)/504 Loan Limit to $10 Million — U.S. Small Business Administration, May 18, 2026 (effective July 4, 2026)

• SBA Loan Rules Just Changed: Here’s What Could Sink Your Application in 2026 — Kiplinger, 2026 (underwriting, collateral, eligibility, equity-injection changes)

• Baltimore group sues to block Harborplace redevelopment demolition of pavilions — CBS News Baltimore, September 2026

• Empire Towers Structural Integrity Incident: Information & Updates — Anne Arundel County Government

• Governor Moore Announces Approval of SBA Economic Injury Disaster Loans for Anne Arundel County Businesses Affected by Empire Towers Structural Emergency — Office of Governor Wes Moore, September 22, 2026

• How AI Adoption Might Affect Bank Lending — Federal Reserve Bank of San Francisco, Economic Letter 2026-25, September 21, 2026

• Bank survey: Small business growth continues amid economic uncertainty — ABA Banking Journal, July 2026 (U.S. Bank survey, national data)

• Anne Arundel County At a Glance — September 2026 — Anne Arundel Economic Development Corporation

• NourishMD Grant Program — Maryland Department of Housing and Community Development

• Maryland Employment and Unemployment Press Release Schedule — Maryland Department of Labor

• Small Business Direct Loans — Maryland DHCD (application status)

• State appeals tax court ruling that struck down digital ad tax — Maryland Matters, September 15, 2026

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice — I’m a guy from Baltimore with a spreadsheet and a bar card. Program details, dates, and rates can change; verify with the agency before you act.

In 1975, a 17-year-old high school senior named Peter Cancro bought the Jersey Shore sub shop where he’d worked since he was 14. His football coach helped him finance it (AP via Yahoo Finance). Fifty years later, Jersey Mike’s has more than 3,100 locations. In January 2025, Blackstone completed its purchase of a majority stake in a deal reportedly valued at $8 billion (NJBIZ).

This isn’t a villain story. Cancro kept a significant stake and still runs the company. The point is simpler: the sign out front no longer tells you who owns the place.

Jersey Mike’s isn’t unusual. One Atlanta private equity firm, Roark Capital, stands behind Subway, Arby’s, Buffalo Wild Wings, Jimmy John’s, Sonic, Dunkin’ and Baskin-Robbins (QSR Magazine). Its brands generated $52.2 billion in system-wide sales in 2024, against McDonald’s $53.3 billion (The Food Institute). That puts a firm most people have never heard of at about 98% of McDonald’s sales.

The same shift is happening, less visibly, at doctors’ offices, vet clinics and newspapers. The name on the door stays local. The owner is a fund with investors elsewhere and a plan to sell within a few years.

The common belief is that this is just business as usual, that a new owner changes the paperwork and nothing else. The data says it depends heavily on the industry and the deal.

What the Numbers Say

Start with scale:

• Physician practices: private equity-owned practice sites grew from 816 in 2012 to 5,779 in 2021, roughly a sevenfold increase (NIHCM).

• Newspapers: private equity’s share of U.S. newspapers rose from about 5% in 2002 to 23% in 2019 (Ewens, Gupta & Howell, NBER).

• Restaurants: private equity invested $94.5 billion in bars and restaurants between 2014 and 2024 (CNBC, citing PitchBook).

Scale alone isn’t the problem. What matters is what happens after the deal closes.

The Playbook, in Plain English

A private equity firm typically buys a business with a small amount of its investors’ money and a large amount of borrowed money. The debt goes on the acquired company’s books, not the firm’s. The company pays it back out of its own cash flow.

Sometimes the firm also sells the company’s buildings and has the company rent them back. This is called a sale-leaseback.

Red Lobster is the case study. In 2014, Golden Gate Capital bought the chain for $2.1 billion and sold its restaurant properties for $1.5 billion (CNBC). According to Restaurant Finance Monitor, the deal priced first-year rent at almost $119 million, rising 2% a year (restfinance.com).

Run those terms forward for ten years:

• Rent in year ten: about $142 million.

• Total rent over the decade: about $1.3 billion.

• Payback: that’s roughly 87% of what the buildings sold for, and Red Lobster still didn’t own a single one of them.

(These are my calculations from the reported lease terms. Actual payments varied as stores closed and leases changed.)

When Red Lobster filed for bankruptcy in 2024, its filings showed rent costs above $190 million the prior year (Restaurant Dive). Golden Gate had sold its stake years earlier.

Red Lobster wasn’t alone. PitchBook found that 10 of the 21 restaurant and bar chains that filed for bankruptcy in 2024, or 48%, were backed by private equity (CNBC). Academic research shows the same pattern over a longer period. A study of 484 large leveraged buyouts found about 20% ended in bankruptcy within ten years, compared with 2% of similar companies, a tenfold difference (Ayash & Rastad, Finance Research Letters).

Buy the Block, Then Set the Price

You might expect the danger to be a big chain undercutting the little guy. The better-documented pattern is quieter. A firm buys one local business, then buys its competitors one at a time. Each purchase is too small to draw much regulatory attention. Eventually one owner holds most of the local market.

The clearest evidence is in health care, where patients can’t comparison-shop in an emergency:

• Three specialties: after private equity bought dermatology, gastroenterology and ophthalmology practices, the amount paid per claim rose 11% and charges rose 20% (Singh et al., JAMA Health Forum).

• Gastroenterology: prices rose $92 per claim, or 28%, after acquisition (Health Affairs).

• Market share: increases were largest where one private equity firm controlled more than 30% of the local market, reaching 18% in gastroenterology and 16% in OB/GYN (Medical Economics, on the UC Berkeley analysis).

That last finding is the whole argument in one number. Prices rise most where competition has been bought out.

The Federal Trade Commission made the same argument in court. It alleged that a private equity firm built a company that bought up nearly every large anesthesia practice in Texas, then raised prices (FTC). Both the firm and the company settled without admitting wrongdoing (FTC; Healthcare Dive). It’s an allegation, not a verdict, but it’s the government’s own description of the playbook.

When Quality Is the Cost

Higher prices are one cost. Worse care is another:

• Nursing homes: private equity ownership raised short-term patient mortality by about 11%, while frontline nursing hours per patient fell (Gupta, Howell, Yannelis & Gupta, Review of Financial Studies). An earlier version of the study estimated roughly 20,150 additional deaths over its twelve-year sample (NBER).

• Hospitals: hospital-acquired conditions such as falls and infections rose 25% after private equity acquisition (NIH summary of Kannan, Bruch & Song, JAMA).

• Newspapers: papers bought by private equity were less likely to close, but they cut reporters and covered less local government (Ewens, Gupta & Howell, NBER).

  

The Other Side of the Ledger

Stopping there would make a tidy argument and an incomplete one. You’ll see the counterexamples in the comments anyway, so here they are first.

The best study of private equity in restaurants found the opposite of what you’d expect. Across 103 chain buyouts in Florida, restaurants got cleaner and safer after the buyout, and menu prices fell (Bernstein & Sheen, Review of Financial Studies).

Consumer products tell a similar story. After buyouts, prices on existing products rose only about 1%. Sales grew 50% faster than at comparable companies, driven by new products rather than price hikes (Fracassi, Previtero & Sheen, Journal of Finance).

The effect on jobs depends on the kind of deal. Across thousands of buyouts, employment fell 13% when private equity bought public companies but rose 13% when it bought privately held ones. Productivity rose 8% on average (Davis et al., Management Science).

So the lesson isn’t that private equity ruins everything it touches. It’s narrower and more useful: risk rises when one owner controls too much of a local market, or when a business carries debt it must pay down fast. Your sub shop probably isn’t the problem. The only vet within 20 miles, now owned by the same firm as the other two, might be.

Where Your Dollar Goes

Even a well-run investor-owned business sends more of your money out of town, through profits, fees and debt payments.

Studies by Civic Economics estimate that independent retailers recirculate about 48% of each dollar locally, compared with under 14% for chains (AMIBA). These studies were often commissioned by local business groups, so treat the exact figures as advocates’ estimates. The direction is hard to argue with.

Here’s what the gap looks like in practice:

• Per $100 spent: about $34 more stays in the community when it goes to an independent.

• Per household: moving $100 a month of routine spending to independents keeps roughly $400 a year circulating locally.

• Across a town: a thousand households doing the same adds up to about $400,000 a year.

(These are my illustrative calculations from the Civic Economics averages, not measured outcomes.)

What About the Legal Side

I can hear the objection: “Mike, if buying up competitors raises prices, isn’t that illegal?”

Sometimes. Federal merger guidelines now say explicitly that a pattern of many small acquisitions can violate antitrust law (Congressional Research Service). But most small deals fall below federal reporting thresholds, and the FTC itself has called roll-ups a route to “stealth consolidation” (FTC). Enforcement tends to arrive after the market has already consolidated.

States are filling some of the gap. Maryland commissioned a study of private equity in physician practices, delivered in January 2025 (Maryland Health Care Commission). But the 2026 bills that would have required notice and review of health care deals, HB 944 and SB 494, never got a committee vote (Maryland General Assembly).

The practical upshot: the law is a backstop. Your spending is the front line.

What I’d Actually Do

Check who owns the place. Read the website’s “About” page and footer for phrases like “a proud partner of” or “part of the ___ family.” Search the business name plus “acquired.” Or ask; independent owners answer that one quickly.

Prioritize where it matters most. Health care, veterinary care and home services are where consolidation shows up in prices. Start there, not with your lunch order.

Move a fixed amount. Pick $100 a month and route it to independents. It’s small enough to sustain, and by the math above it’s about $400 a year that stays on your street.

The Closing Test

Before you spend, ask one question: if this place raised its prices next year, where else could I go?

If the answer is “plenty of places,” the market is working, whoever owns it. If the answer is “nowhere,” competition is already thin, and your spending is one of the few ways to keep a second option alive.

The sign says local. Check who owns it.

What’s one business in your town you only recently learned wasn’t locally owned? Tell me in the comments.

Leave a comment

Sources

• Ayash, B. & Rastad, M. (2021). Leveraged buyouts and financial distress. Finance Research Letters. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3423290

• Bernstein, S. & Sheen, A. (2016). The operational consequences of private equity buyouts: Evidence from the restaurant industry. Review of Financial Studies, 29(9), 2387–2418. https://doi.org/10.1093/rfs/hhw037

• Davis, S. J., Haltiwanger, J., Handley, K., Lipsius, B., Lerner, J. & Miranda, J. (2025). The (heterogeneous) economic effects of private equity buyouts. Management Science, 71(11). https://ideas.repec.org/a/inm/ormnsc/v71y2025i11p9569-9587.html

• Ewens, M., Gupta, A. & Howell, S. T. (2022). Local journalism under private equity ownership. NBER Working Paper 29743. https://www.nber.org/papers/w29743

• Fracassi, C., Previtero, A. & Sheen, A. (2022). Barbarians at the store? Private equity, products, and consumers. Journal of Finance, 77(3), 1439–1488. https://doi.org/10.1111/jofi.13134

• Gupta, A., Howell, S. T., Yannelis, C. & Gupta, A. (2024). Owner incentives and performance in healthcare: Private equity investment in nursing homes. Review of Financial Studies, 37(4), 1029. https://academic.oup.com/rfs/article-abstract/37/4/1029/7441509

• Increases in physician professional fees in private equity–owned gastroenterology practices (2024). Health Affairs. https://doi.org/10.1377/hlthaff.2024.00190

• Kannan, S., Bruch, J. D. & Song, Z. (2023). Changes in hospital adverse events and patient outcomes associated with private equity acquisition. JAMA. Summary: https://www.nih.gov/news-events/nih-research-matters/infections-falls-increased-private-equity-owned-hospitals

• Singh, Y. et al. (2022). Association of private equity acquisition of physician practices with changes in health care spending and utilization. JAMA Health Forum. https://pubmed.ncbi.nlm.nih.gov/36218927/

• Federal Trade Commission, FTC v. U.S. Anesthesia Partners (case page and 2025 settlement release); “Slow the Roll-up” (2024).

• Maryland Health Care Commission (2025). Private Equity Investments in Physician Practices in Maryland; Maryland General Assembly, HB 944 (2026).

• News and industry sources: AP/Yahoo Finance, NJBIZ, QSR Magazine, The Food Institute, CNBC (incl. PitchBook data), Restaurant Dive, Restaurant Finance Monitor, Healthcare Dive, Medical Economics, NIHCM, AMIBA (Civic Economics), Congressional Research Service.

This article is for general information only and is not legal or financial advice. Figures and legislation status are current as of September 2026; check the linked sources before relying on them.

Let’s invent someone. Call him Ray.

Ray runs an HVAC shop out of a warehouse bay off Ritchie Highway — a dozen trucks, second generation, his father’s name still on the door and nobody’s had the heart to take it off. Last spring Ray decided his problem was messaging. He paid a copywriter $4,400 for a new homepage, a rewritten services page, and a six-email follow-up sequence. The copy was good. Ray read the part about “three generations of Anne Arundel County families” out loud to his wife and got a little choked up.

His close rate did not move.

Here’s what Ray never looked at. The intake form on that beautiful new homepage asks eleven questions before it asks for a phone number. His quotes list good, better, best in a plain vertical stack with nothing marked and the cheapest one on top. His maintenance agreements expire silently every March and it is nobody’s job to notice. His scheduling link opens on an empty calendar and asks a homeowner with a dead compressor to invent a date out of thin air.

Ray spent $4,400 on the words. He has never spent ten minutes on the path.

What Most Owners Think “Sales Psychology” Means

Ask an owner what psychological tools big companies use to sell, and you’ll usually get some version of the same answer: a bag of verbal tricks. Urgency lines. Closing techniques. The thing the car dealer does where he goes to “check with his manager.” Most owners assume sales psychology is a more sophisticated kind of talking, and a lot of them find the whole idea a little greasy — which is exactly why they never look into it.

That belief isn’t stupid. It’s also close to backwards.

The research that actually holds up isn’t about what you say. It’s about the structure the customer moves through: what’s already selected, what they see first, what they have to do next, how much friction sits in the path, what number appears beside what other number. Richard Thaler and Cass Sunstein built a whole book on this — Nudge — and their central claim is that somebody already decided all of that for your business. Usually it wasn’t you. It was a software default, or an accident, or whoever set up your booking page in 2019.

They call the person who makes those decisions deliberately a choice architect. You don’t get to opt out of that job. You only get to do it badly.

What follows is the working set, grouped by which part of the buying decision it acts on. A customer deciding whether to hire you is really doing four separate jobs — noticing, comparing, worrying, and following through — and each one has its own tools, its own evidence, and its own way of getting you sued.

I. Tools That Act on Attention

What gets seen, and what happens when nobody sees anything.

1. Defaults — what happens when the customer does nothing

This is the strongest tool in the box and the one almost no small business sets on purpose.

Brigitte Madrian and Dennis Shea studied a company that switched its 401(k) from sign-up to automatic enrollment. Same plan, same match, same HR speech. Participation among new hires went from 37% to about 86% (Madrian & Shea, 2001). Nobody was persuaded of anything; the argument for saving didn’t improve. What changed is that the employee who did nothing was now enrolled instead of not enrolled.

The follow-on finding is the one that should make you sit up. The plan defaulted to 3% of salary in a money market fund. About 61% of automatically enrolled employees stayed at exactly that rate and exactly that fund — a combination only 1% had chosen on their own beforehand. The default didn’t just decide whether. It decided how much, for most people, indefinitely.

Eric Johnson and Daniel Goldstein found the same shape in organ donation: consent ran about 42% under opt-in and 82% under opt-out in their experiment, identical question, opposite default (Johnson & Goldstein, 2003).

How you’d use it: Mark a recommended tier on every proposal. Pre-fill the service plan. Make the maintenance agreement renew with notice instead of expiring in silence. Set the follow-up to happen on a schedule rather than when someone remembers.

The legal line.

• As of June 1, 2026, Maryland’s automatic renewal law is in force — Chapter 204 (SB 49), adding § 14-1328 to the Commercial Law Article (Md. Laws 2025, Ch. 204). If you auto-renew a consumer, you owe clear pre-purchase disclosure of the renewal price and cancellation method, advance notice on longer terms, and a cancellation mechanism at least as easy as sign-up, in the same medium. Violations are unfair or deceptive trade practices under Title 13, enforceable by the Attorney General’s Consumer Protection Division.

2. Friction — the same tool pointed the other way

Sunstein later named the ugly side of this: sludge, friction sitting between a person and something they actually want (Sunstein, 2022).

Here’s the discipline. Every piece of friction in your business is a statement about what you want less of. Walk your own process and read it back as intent. Ray’s eleven-field intake form says, out loud, we would prefer fewer leads. An empty calendar says pick a day, and good luck. “Call the office during business hours” says we’d rather you didn’t.

How you’d use it: Count the steps between interest and contact, then delete half. Ask for the phone number and the problem — you can qualify a human being in ninety seconds, which is what the phone is for.

The legal line.

• Friction on the way out is where this stops being marketing. The FTC’s click-to-cancel rule was vacated by the Eighth Circuit in July 2025 (Latham & Watkins) and the Commission restarted the rulemaking in March 2026 (Covington) — but the rule was never the only exposure. ROSCA and Section 5 enforcement ran straight through the gap. Ask Amazon, which settled for $2.5 billion in September 2025 over Prime enrollment and cancellation design (Alston & Bird).

3. Salience — doing the customer’s math for them

Researchers pulled transaction data from a counter-service restaurant that changed its tipping screen twice over nine months — 5,169 card transactions. Period one: percentages only (5%, 10%, 15%). Period two: percentages with the dollar amount beside them. Period three: back to percentages.

The share of customers who tipped at all went 17.6% → 39.6% → 32.2% (Li & Warren, 2026). Same menu, same prices, same staff. One screen.

Nobody got more generous. People stopped having to do arithmetic under time pressure — and under time pressure, people don’t do arithmetic. They leave.

How you’d use it: Kill every abstraction that makes the customer calculate. “Starting at,” percentage ranges, “plus applicable fees,” financing described in APR without a monthly number. Print the dollar figure.

II. Tools That Act on Comparison

How the customer decides what a thing is worth.

4. Anchoring — the first number sets the scale

Amos Tversky and Daniel Kahneman established it fifty years ago: whatever number lands first drags every later judgment toward it, even when the first number is obviously arbitrary (Tversky & Kahneman, 1974). Nudge puts it in chapter one for a reason.

How you’d use it: Stop leading with your cheapest option. Ray puts good on top of the column and then wonders why he sells so much good. Lead with the full-scope version, then let the smaller packages read as savings against it. Same three options, opposite anchor.

The legal line.

• Anchoring against a fictional former price is a different animal. The FTC’s Guides Against Deceptive Pricing (16 CFR Part 233) require that a “was $X” or “regularly $X” claim reflect a bona fide price you actually offered — not a number invented to make the discount look good. Maryland’s Consumer Protection Act reaches the same conduct.

5. Price presentation — endings, bundling, and what gets listed separately

Eric Anderson and Duncan Simester ran three field experiments with a real retailer’s catalogs. A $9 price ending increased demand in all three — and the effect was strongest on items the retailer hadn’t sold before, where the customer had the least other information to go on (Anderson & Simester, 2003).

That last clause is the useful part. Price cues do the most work where knowledge is thinnest. Which means they work best on exactly the customers least equipped to evaluate you — and that’s worth sitting with for a second before you go reprice your whole service menu.

How you’d use it: Test endings on new offerings, not on the work your repeat customers already know the price of. Present bundles as a single number with the components listed, rather than a stack of line items the customer has to add up.

The legal line.

• Hiding mandatory charges until checkout is now specifically regulated, though the rule’s scope is narrow: the FTC’s Rule on Unfair or Deceptive Fees (16 CFR Part 464) took effect May 12, 2025 and covers live-event ticketing and short-term lodging. If you’re neither, you’re still squarely inside Section 5 and the Maryland CPA. The safe rule hasn’t changed in fifty years: the number you advertise should be the number the customer pays.

6. Payment framing — how it feels to hand over the money

Drazen Prelec and Duncan Simester ran auctions for Boston sports tickets where bidders were told they’d pay by cash or by credit card. Credit-card bidders bid roughly twice as much (Prelec & Simester, 2001).

The mechanism is the pain of paying — how vividly the customer experiences the money leaving. Cash hurts most. Card hurts less. Financed, auto-drafted, or bundled into a monthly plan hurts least.

How you’d use it: If your price is a real obstacle, the lever may be the payment structure rather than the number. A monthly maintenance plan and an annual invoice for the same total are, psychologically, two different prices. Legitimately so — you’re also smoothing the customer’s cash flow.

III. Tools That Act on Risk

What the customer is afraid of, and who they check with.

7. Loss framing — the same fact, pointed at what they’d lose

Kahneman and Tversky’s prospect theory established that losses land harder than equivalent gains (Kahneman & Tversky, 1979). Nudge covers it as loss aversion, and it’s the closest thing in this list to a genuine language tool — the only one where the words themselves are the lever.

How you’d use it: “You’ll save $40 a month on your bill” and “you’re losing $40 a month right now” describe the same duct work. The second one gets a call back. Be careful with it: it’s the tool on this list most easily turned into fear-mongering, and homeowners have long memories about which contractor scared them.

The legal line.

• Loss framing built on a fabricated loss — inflated damage claims, invented code violations, the classic “your heat exchanger is cracked” when it isn’t — isn’t a framing choice. In Maryland that’s deceptive trade practice territory, and for a licensed trade it’s a board complaint waiting to happen.

8. Social proof — and the honest caveat about it

Noah Goldstein, Robert Cialdini, and Vladas Griskevicius tested hotel towel-reuse cards. The standard environmental appeal got 35.1% reuse. A descriptive-norm card — “the majority of guests reuse their towels” — got 44.1%. And the provincial version, naming the guest’s own room, got 49.3% (Goldstein, Cialdini & Griskevicius, 2008). The closer the reference group sits to the customer’s actual situation, the more it moves.

Now the part most people quoting that study leave out. A German replication, in two studies with larger samples, found no advantage for the norm messages at all — and in one condition the “guests in this room” version did worse than the plain environmental appeal (Bohner & Schlüter, 2014).

So: use the mechanism, but use it the way the theory suggests rather than the way the headline number promises. Specificity of reference group is the transferable idea. “Four of your neighbors on Bayside Drive had us out this spring” is doing something that “trusted by thousands” cannot.

The legal line.

• This is the tool with the sharpest teeth, and it caught a lot of small businesses off guard. The FTC’s Rule on the Use of Consumer Reviews and Testimonials (16 CFR Part 465) took effect October 21, 2024 and applies to businesses of every size. It bans fake and AI-generated reviews, undisclosed insider reviews, conditioning an incentive on the review being positive, and using threats to suppress negative reviews (FTC guidance). Knowing violations carry civil penalties — currently up to $53,088 per violation, adjusted for inflation (FTC, 2025). Per review. Asking your happy customers to leave honest reviews is fine, and it’s the single highest-return thing on this entire list. Asking your nephew is now expensive.

9. Scarcity and urgency — most popular, weakest hand

The classic study had subjects rate cookies from a jar holding either ten or two; the scarce cookies rated higher (Worchel, Lee & Adewole, 1975). Every countdown timer on the internet descends from that jar.

I’d put scarcity last on purpose. It has the widest adoption, the thinnest field evidence relative to its popularity, and by far the most consumer skepticism attached to it — your customers have been trained by a decade of fake countdown clocks. Real scarcity works because it’s real: you have four install slots before the heat breaks, and saying so is information.

The legal line.

• Manufactured urgency is straightforwardly deceptive — a timer that resets, “only 2 left” that’s never true, a “this week only” price that runs all year. Section 5 and the Maryland CPA both reach it, and it’s among the easiest deceptions for a regulator to document, because the evidence is a screenshot.

IV. Tools That Act on Commitment

What keeps someone moving once they’ve started.

10. Endowed progress — start them partway

Joseph Nunes and Xavier Drèze gave car wash customers one of two loyalty cards: an eight-stamp card, or a ten-stamp card with two stamps already filled in. Identical work required. Over nine months, 19% completed the eight-stamp card and 34% completed the pre-stamped ten (Nunes & Drèze, 2006).

People finish things they’ve started. They don’t start things.

How you’d use it: Structure any multi-step process so the customer arrives already underway. Pre-fill what you know on the intake form. Send the proposal with the scope section completed rather than blank. Frame onboarding as step two of four, not step one of three.

11. Reciprocity — the small unrequested gift

Servers who delivered candy with the check got larger tips, and the researchers found the effect tracked reciprocity rather than mood or sugar — how the gift was given mattered more than how much candy it was (Strohmetz, Rind, Fisher & Lynn, 2002).

How you’d use it: The unrequested, un-invoiced small thing. Tightening the loose fitting you noticed on an unrelated call. The filter you left behind. It has to be genuinely free and genuinely unasked-for — reciprocity collapses the moment it reads as an opening move.

The legal line.

• If you advertise the gift as “free,” the FTC’s guidance on free offers (16 CFR Part 251) applies: the regular price of the thing you must buy to get it can’t be inflated to cover it.

12. The small first commitment

Jonathan Freedman and Scott Fraser’s foot-in-the-door experiments showed that people who agreed to a trivial request were markedly more likely to agree to a larger one later (Freedman & Fraser, 1966).

How you’d use it: This is what a $79 diagnostic visit is actually for, and why “free estimate” sometimes converts worse than a cheap paid one. A small yes is a different kind of commitment than no yes at all.

The legal line.

• A small first commitment must not quietly become a large one. This is the same negative-option exposure as tool #1: if the trial converts to a paid plan, Maryland § 14-1328 now governs how you disclose it and how easily the customer gets out.

What About the Legal Side

Three honest problems, and you should have all three before you spend a Saturday on this.

The evidence base is softer than the conference talks.

• In 2022 a team reanalyzed the largest meta-analysis of nudging and found severe publication bias; the headline effect dropped from d = 0.43 to d = 0.04 after correction, and the authors concluded no evidence remained that nudges work as general-purpose tools (Maier et al., 2022). That finding is contested — the original authors pushed back hard — but you should know it exists before someone sells you a “behavioral audit.”

The failures cluster, and the pattern is useful.

• The tools that collapse are the clever ones. The decoy tier — the deliberately bad third option meant to make your middle package shine — largely evaporates once people see real products instead of spec tables (Frederick, Lee & Baskin, 2014). “Too many choices paralyzes buyers” is mostly folklore; a meta-analysis of choice overload found a mean effect near zero (Scheibehenne, Greifeneder & Todd, 2010). The social-proof study above didn’t replicate in Germany. What survives best is section I: defaults, friction, and telling people what a number means — tested against payroll systems and card readers, not undergraduates rating hypothetical hotels.

And there’s the question of whether you should.

• Every tool above works identically whether you’re pointing it at something good for the customer or something bad. The test I use is simple: a nudge is a design choice you’d be comfortable explaining to the customer out loud. Sludge is one that only works because they didn’t notice. Marking your recommended package passes. Burying the cancel link fails. Same psychology, and only one of them is still legal in Maryland.

What I’d Actually Do

Write down what your business does when the customer does nothing.

• Not what it should do — what it does. Unanswered quotes, silent renewals, abandoned carts, the estimate you sent in July. That list is your choice architecture, and it was written by nobody. This takes an afternoon.

Fix section I before you touch anything else.

• One default set on purpose, half the fields deleted, dollar figures replacing percentages. That’s the whole evidence-backed core, it costs nothing, and it doesn’t require you to be clever.

Ask your actual customers for reviews, and specify the reference group when you use them.

• Highest return on this list, and the compliance rules are easy to follow as long as every review came from a real customer who wasn’t paid for a positive one.

Audit your urgency claims and delete any that aren’t true.

• If the timer resets, it’s not a marketing tactic — it’s a documented deception with your own screenshot as the exhibit.

Make the exit exactly as easy as the entrance.

• Same medium, same clicks, no retention maze. Maryland law if you auto-renew anybody, and the cheapest test of whether everything else you’ve built is honest: if a design only survives because leaving is hard, it was never persuasion.

Ray’s copywriter did good work, by the way. That’s the part that stings — the words on that homepage are better than the words on mine.

But words are what a customer reads once, on a good day, with time to spare. The structure is what your business does at 9:15 on a Tuesday night when a homeowner in Pasadena is standing in a cold house looking at an empty calendar, deciding whether you’re worth the effort.

The path is the thing. It’s almost always the thing.

Share

Leave a comment

Sources

• Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press.

• Madrian, B. C., & Shea, D. F. (2001). The Power of Suggestion: Inertia in 401(k) Participation and Savings Behavior. The Quarterly Journal of Economics, 116(4), 1149–1187. https://doi.org/10.1162/003355301753265543

• Johnson, E. J., & Goldstein, D. (2003). Do Defaults Save Lives? Science, 302(5649), 1338–1339. https://doi.org/10.1126/science.1091721

• Sunstein, C. R. (2022). Sludge Audits. Behavioural Public Policy, 6(4), 654–673. https://doi.org/10.1017/bpp.2019.32

• Li, Q., & Warren, N. B. (2026). When percentages meet amounts: Evidence from a natural experiment on digital tipping prompts. International Journal of Hospitality Management, 137, 104680. https://doi.org/10.1016/j.ijhm.2026.104680

• Tversky, A., & Kahneman, D. (1974). Judgment under Uncertainty: Heuristics and Biases. Science, 185(4157), 1124–1131. https://doi.org/10.1126/science.185.4157.1124

• Anderson, E. T., & Simester, D. I. (2003). Effects of $9 Price Endings on Retail Sales: Evidence from Field Experiments. Quantitative Marketing and Economics, 1(1), 93–110. https://doi.org/10.1023/A:1023581927405

• Prelec, D., & Simester, D. (2001). Always Leave Home Without It: A Further Investigation of the Credit-Card Effect on Willingness to Pay. Marketing Letters, 12(1), 5–12. https://doi.org/10.1023/A:1008196717017

• Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47(2), 263–291. https://doi.org/10.2307/1914185

• Goldstein, N. J., Cialdini, R. B., & Griskevicius, V. (2008). A Room with a Viewpoint: Using Social Norms to Motivate Environmental Conservation in Hotels. Journal of Consumer Research, 35(3), 472–482. https://doi.org/10.1086/586910

• Bohner, G., & Schlüter, L. E. (2014). A Room with a Viewpoint Revisited: Descriptive Norms and Hotel Guests’ Towel Reuse Behavior. PLOS ONE, 9(8), e104086. https://doi.org/10.1371/journal.pone.0104086

• Worchel, S., Lee, J., & Adewole, A. (1975). Effects of supply and demand on ratings of object value. Journal of Personality and Social Psychology, 32(5), 906–914. https://doi.org/10.1037/0022-3514.32.5.906

• Nunes, J. C., & Drèze, X. (2006). The Endowed Progress Effect: How Artificial Advancement Increases Effort. Journal of Consumer Research, 32(4), 504–512. https://doi.org/10.1086/500480

• Strohmetz, D. B., Rind, B., Fisher, R., & Lynn, M. (2002). Sweetening the Till: The Use of Candy to Increase Restaurant Tipping. Journal of Applied Social Psychology, 32(2), 300–309. https://doi.org/10.1111/j.1559-1816.2002.tb00216.x

• Freedman, J. L., & Fraser, S. C. (1966). Compliance without pressure: The foot-in-the-door technique. Journal of Personality and Social Psychology, 4(2), 195–202. https://doi.org/10.1037/h0023552

• Maier, M., Bartoš, F., Stanley, T. D., Shanks, D. R., Harris, A. J. L., & Wagenmakers, E.-J. (2022). No evidence for nudging after adjusting for publication bias. PNAS, 119(31), e2200300119. https://doi.org/10.1073/pnas.2200300119

• Frederick, S., Lee, L., & Baskin, E. (2014). The Limits of Attraction. Journal of Marketing Research, 51(4), 487–507. https://doi.org/10.1509/jmr.12.0061

• Scheibehenne, B., Greifeneder, R., & Todd, P. M. (2010). Can There Ever Be Too Many Options? A Meta-Analytic Review of Choice Overload. Journal of Consumer Research, 37(3), 409–425. https://doi.org/10.1086/651235

• Maryland Laws 2025, Chapter 204 (Senate Bill 49) — Consumer Protection: Automatic Renewals, adding Md. Code, Com. Law § 14-1328 (effective June 1, 2026)

• 16 CFR Part 465 — Rule on the Use of Consumer Reviews and Testimonials and FTC business guidance

• 16 CFR Part 464 — Rule on Unfair or Deceptive Fees (effective May 12, 2025)

• 16 CFR Part 233 — Guides Against Deceptive Pricing and 16 CFR Part 251 — Guide Concerning Use of the Word “Free”

• FTC Publishes Inflation-Adjusted Civil Penalty Amounts for 2025

• Eighth Circuit Vacates FTC’s Click-to-Cancel Rule — Latham & Watkins

• FTC Launches New Rulemaking on the Negative Option Rule — Covington & Burling

• FTC Reaches Record $2.5 Billion Settlement with Amazon over Prime Subscription Practices — Alston & Bird

Ray is invented — assembled out of enough real intake forms and unmarked proposals that somebody reading this is going to think I’ve been in their shop.

Nothing above is legal advice and I’m not your lawyer. I’m a guy from Baltimore with a spreadsheet and a bar card. The automatic renewal and consumer review rules in particular are new enough, and the federal negative-option rulemaking unsettled enough, that you should read the current text yourself — or have someone read it for you — before you decide your sign-up flow or your review process is fine. Civil penalty amounts adjust for inflation; verify the current figure before you quote it. Crabcakes & Commerce is written for Maryland small business owners.

Leave a comment

You've already signed more contracts this week than you think. The coffee. The app you clicked "I agree" on. The job you took over the phone for two grand. In this episode of The Discussion, I break down what actually makes a contract binding (offer, acceptance, consideration, and a meeting of the minds), which deals legally have to be in writing, and the handful of clauses that do most of the work in a small business contract: scope, payment, change orders, and how you get out.

FYI - Here is the link to the scaling business in Maryland webinar I discussed:

Coffee’s on. Let’s go through the books.

Quiet-ish week on the surface, busy week underneath. The Fed did something it hasn’t done since 2023. The state answered a question I left hanging for you last Monday. The jobs report landed with a headline number and a footnote that don’t quite agree with each other. And while all that was happening, a Baltimore poultry company quietly opened an 80,000-square-foot plant in Hanover like it was no big deal — which, compared to everything else this week, it almost was.

On the ledger this week: the Fed raised rates a quarter point Wednesday, its first hike in three years; the Comptroller appealed the digital ad tax ruling I told you about two weeks ago, freezing the refunds; Maryland’s August jobs report says unemployment fell for the third straight month while the state also lost 8,200 jobs; Holly Poultry opened a new Anne Arundel County facility and 140 jobs with it; the DHCD 4% loan round closed Thursday and nobody’s said what happened yet; and if you run a pass-through entity, the Comptroller changed how your remaining 2026 estimated payments get calculated — back in April, but I’d bet most of you never heard about it.

1. The Fed just raised rates for the first time in three years

I told you two weeks ago the market had swung to pricing a hike instead of a cut. It happened. The Federal Open Market Committee voted 12–0 on Wednesday to raise the target range for the federal funds rate a quarter point, to 3.75%–4.00% — the first increase since 2023. The Committee’s own language was blunt: inflation “remains elevated,” and the move is meant to “support a timelier return” to the 2% target. Economic activity is still described as expanding at a solid pace, with job gains keeping pace with the workforce.

The part that matters more than the quarter point itself: 16 of 18 committee participants penciled in at least one more hike this year in their updated projections, with four expecting two. This isn’t a one-and-done correction. The committee is signaling a cycle, not an event.

Most small-business borrowing in this state is priced off prime, and prime moves the day the Fed does. If you’re carrying a $500,000 variable-rate note, this hike alone is roughly $1,250 a year — manageable on its own, less manageable if October or December bring another one on top of it.

What I’d do: If you’ve been putting off locking a variable-rate note into something fixed because “the Fed might still cut,” that argument is dead. Call your lender this week, not next month — the next FOMC meeting is October 27–28, and the dot plot says another hike is more likely than not. If you’re drawing on a line of credit for fall inventory or year-end payroll, draw it now; the rate on money you already borrowed doesn’t move retroactively.

2. Maryland keeps fighting for a tax a court already struck down

Remember the cliffhanger from two weeks ago? On August 14, the Maryland Tax Court struck down the state’s digital advertising tax, ruling for Apple, Google, and Peacock TV and ordering refunds. I told you the appeal deadline landed September 14 and that nobody had confirmed a filing. Now we know: Comptroller Brooke Lierman filed an appeal in the Circuit Court for Anne Arundel County on September 15, asking a judge to review the tax court’s ruling. “I fully support the Attorney General of Maryland’s appeal of these decisions,” Lierman said in a statement, adding she remains “hopeful that, upon review, Maryland courts will find this law constitutional.”

No hearing date has been set. Until one is, the refunds stay frozen — and the state has collected roughly half a billion dollars over the tax’s five years in effect, about 43% of what it would have raised running at the full $250 million-a-year pace the original estimate assumed. The Baltimore Sun reports tax-law analysts questioning whether it’s worth the state’s resources to keep defending a law three separate tax court rulings have now called unconstitutional. That’s a fair question, and I don’t know the answer. What I do know is the litigation isn’t over, and neither is the uncertainty for anyone whose business model touches digital ad spend.

You still don’t pay this tax directly — it hits companies with $100 million or more in global revenue. But it funds the Blueprint for Maryland’s Future, and a half-billion-dollar hole in that funding doesn’t stay abstract for long if the state loses on appeal and has to find the money somewhere else.

What I’d do: Nothing changes today. But don’t treat this as resolved — budget as if the tax survives, and if you buy digital advertising through a platform that passes this cost through, keep an eye on your invoices for any pass-through-tax line item that shifts once a circuit court ruling actually lands.

3. The jobs report gave you a good headline and a worse footnote

Maryland’s unemployment rate fell to 4.1% in August, the third straight monthly decline, released Friday, September 18. That’s the number that made the headline. Here’s the number that didn’t: the state lost 8,200 total nonfarm jobs between July and August, driven almost entirely by private-sector losses, with construction alone shedding 2,900 jobs over the two-month window.

Read past the headline and the picture gets murkier, not clearer. July’s initial estimate of 11,700 jobs added got revised down to 10,200, which means the two-month combined gain is only about 2,000 jobs — a 0.1% growth rate, matching the national pace. State labor officials are telling reporters to look at the two-month window instead of any single month because the monthly survey is volatile. That’s not spin, that’s just true of how this data gets collected — but it also means neither the “unemployment keeps falling” headline nor the “state lost 8,200 jobs” headline tells you the whole story on its own. Both are accurate. Neither is complete.

The sector detail is worth your attention regardless of which top-line number you trust: accommodation and food services led hiring with 4,000 jobs added over the two months, followed by professional and technical services at 2,800. Construction, private education, health care and social assistance, and retail all lost ground.

What I’d do: If you’re in construction, retail, or health care and you’ve felt hiring get harder to justify the last two months, the data backs up what you’re feeling — you’re not imagining it. If you’re in food service or professional services, the same data says you’re swimming with the current, not against it. Either way, trust your own bookings ledger over any single month’s headline number; this is exactly the kind of report that gets summarized into a single misleading sentence by anyone who didn’t read past the first paragraph.

4. A Baltimore poultry company just bet 140 jobs on Hanover

Buried under bigger headlines this week: Holly Poultry, the Baltimore-based, family-owned poultry distributor, opened an 80,000-square-foot manufacturing and production facility in Hanover, in Anne Arundel County near BWI, and plans to create 140 full-time jobs there over the next four years. The company has spent three decades supplying restaurants and wholesalers across the Mid-Atlantic; the new facility lets it push further into grocery retail under its own consumer brands, One Fine Chicken and Easy Street, sold through Giant, Harris Teeter, Safeway, and others.

The state and county put together up to $875,000 in assistance for the project: a $750,000 conditional loan through Advantage Maryland, a $50,000 workforce-training grant, and a $75,000 conditional loan from the Anne Arundel County Economic Development Corporation. That’s real, structured public money attached to a real building with a real address, not a press-release promise.

I’m putting this next to item 3 on purpose. The same week the state’s jobs data shows construction and manufacturing-adjacent sectors losing ground on paper, a manufacturer just spent real capital betting the other way in Anne Arundel County specifically. One data point doesn’t overrule a trend, but it’s a useful reminder that statewide averages hide a lot of local variation.

What I’d do: If you supply food-service, grocery, or logistics operations anywhere near BWI or the Hanover corridor, a new 80,000-square-foot facility with fulfillment operations is a potential customer or a potential competitor for warehouse labor — worth a phone call either way, before your competitors make it first.

5. The cheapest loan money in the state closed its doors, and nobody’s said what happened

I flagged this deadline for you twice. DHCD’s Small Business Direct Loan round — up to $2 million, 4% fixed, terms to 30 years — closed at 11:59 p.m. Thursday, September 17. As of this morning, the agency’s own page still lists the application status as “Open,” even though the stated closing date has passed — that’s very likely just a page that hasn’t been updated rather than an actual extension, but I checked, and there’s no announcement of results, an extension, or a new round.

If you applied, there’s nothing to do but wait. If you missed it, the preference lanes haven’t gone anywhere and neither has the pattern: DHCD runs these competitive rounds periodically, prioritizing projects that rehab vacant or blighted property, create housing, expand fresh-food access, or preserve child care capacity, and the agency’s business development team takes consultations year-round regardless of whether a round is open.

What I’d do: If you missed this window, book an initial consultation with DHCD’s business development team now rather than waiting for the next round announcement — getting on their radar before a deadline is posted puts you ahead of the scramble. And if you’re one of the applicants waiting on results, sign up for DHCD’s email updates so you’re not refreshing the website.

6. A quiet one: your PTE estimated payments changed back in April, and you may not have noticed

This one didn’t happen this week — it happened in April, and I only found it because it was sitting in the middle of an otherwise recycled research brief with no date attached. But it’s real, it’s useful, and I’d bet most of you running a pass-through entity haven’t seen it: the Comptroller issued guidance, effective April 13, 2026, on how tax year 2026 PTE estimated payments get calculated, after the legislature postponed a broader change to how PTE taxable income is calculated until tax year 2027.

The short version: if your electing PTE made its first-quarter 2026 payment assuming the newer, broader calculation method, you likely overpaid, and you’re allowed to adjust your remaining quarterly payments to account for it. The Comptroller is also disregarding any entity-level tax election made with that first-quarter payment and will instead honor whatever election comes with your next filing.

Q3’s payment deadline was September 15 — already behind us as of this writing. The next one is January 15, 2027.

What I’d do: If you or your accountant filed a Q1 2026 PTE payment before mid-April without accounting for this, pull that filing and check whether you overpaid. If so, you can true it up in your January payment rather than waiting for a refund. This is a five-minute conversation with your CPA that could matter more than anything else in this section.

The week ahead

• Tuesday, October 20 — Maryland’s September 2026 statewide employment report is released.

• Tuesday–Wednesday, October 27–28 — the Fed’s next FOMC meeting, with a decision Wednesday afternoon. The dot plot says don’t be surprised by another hike.

• Thursday, October 1 — the Worker Freedom Act (SB 417) takes effect for Maryland employers, restricting mandatory meetings on political or religious matters.

• Tuesday, November 3 — Maryland’s general election, including the governor’s race and statewide offices.

• Sunday, November 15 — the FAMLI Declaration of Intent window closes for employers weighing a private paid-leave plan.

• Friday, January 15, 2027 — Q4 PTE estimated tax payments due, per item 6 above.

See you next Monday.

Share

Leave a comment

Sources

• Federal Reserve issues FOMC statement — Federal Reserve Board, September 16, 2026 (rate decision, vote, target range)

• Fed rate decision September 2026: Rates rise to 3.75%-4% — CNBC, September 16, 2026 (dot plot, projections)

• Next FOMC Meeting: October 28, 2026 — Finance Calendar (next meeting date)

• Digital ad sales tax struck down by court — Maryland Matters, August 14, 2026

• State appeals tax court ruling that struck down digital ad tax — Maryland Matters, September 15, 2026 (appeal filing, Comptroller statement, revenue figures)

• State fight for digital ad tax is not over — yet — Baltimore Sun, September 17, 2026

• Eye on the Economy: A Look at Maryland and the Regional Economy in August 2026 — RESI of Towson University, September 18, 2026 (unemployment rate, job losses, Fed context)

• New estimates suggest Maryland added 2,000 jobs across July and August amid monthly fluctuations — Nottingham MD, September 18, 2026 (July revision, two-month combined figures, sector detail)

• Holly Poultry expands Maryland manufacturing efforts — business.maryland.gov, September 15, 2026

• Holly Poultry expands in Maryland, plans to add 140 jobs — Nottingham MD, September 18, 2026 (incentive package breakdown)

• Small Business Direct Loans — Maryland DHCD (live application status, deadline, terms, eligibility)

• Changes to Tax Year 2026 Pass-Through Entity Estimated Payments — Maryland Comptroller Tax Alert, effective April 13, 2026

• Maryland Employment and Unemployment Press Release Schedule — Maryland Department of Labor

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice — I’m a guy from Baltimore with a spreadsheet and a bar card. Program details, dates, and rates can change; verify with the agency before you act.

Coffee’s on. Let’s go through the books.

Quiet week for new headlines, loud week for old ones coming due. Nothing brand-new blew up Annapolis this week — instead, three storylines I’ve been tracking for you all hit their deadlines at once, a poll landed with a “Maryland” label that doesn’t quite earn it, and the state quietly shipped a genuinely useful free tool while nobody was looking.

On the ledger this week: the 30-day clock to appeal August’s digital ad tax ruling runs out today, and nobody’s confirmed what happened; the Fed meets Tuesday and Wednesday with hot inflation data behind it and rate-hike odds climbing; DHCD’s 4% fixed-rate small business loan round has three days left; a “Maryland” small-business optimism report that hit inboxes this month is actually the national number with a local quote stapled on; and the state relaunched its free site-selection and funding-search tool for anyone scouting a location.

1. The digital ad tax appeal deadline is today. As of this morning, nobody’s saying what happened.

Quick recap for anyone just joining: on August 14, the Maryland Tax Court struck down the state’s first-in-the-nation digital advertising tax, ruling in favor of Apple, Google, and Peacock TV and ordering refunds with interest. The court found the tax — which runs 2.5% to 10% depending on a company’s global revenue, and only applies to businesses over $100 million in worldwide revenue — violated the federal Internet Tax Freedom Act, the dormant Commerce Clause, and due process. Comptroller Brooke Lierman said she disagreed and would work with the Attorney General to defend it.

Here’s the part that lands today. Under Maryland law, a party has 30 days from the Tax Court’s order to file a petition for judicial review in circuit court — that’s Md. Code Ann., Tax-Gen. § 13-532. Thirty days from August 14 lands on Sunday, September 13, which means the practical deadline is today, Monday the 14th. I looked for confirmation that the Comptroller’s office actually filed before sending this out this morning, and I couldn’t find one — every story I can locate is from mid-August, reporting the intent to appeal, not a filing. If an appeal lands today, the refunds — roughly $535.5 million collected through July — stay frozen while it plays out. If it doesn’t, that money starts moving, and Annapolis has an even bigger hole to fill heading into 2027.

You don’t pay this tax; it’s aimed at the Apples and Googles of the world. But the state’s tax appetite doesn’t stay confined to the companies it’s aimed at, and you’ve already seen the sequel: the 3% tax on IT and data services that arrived last year to help plug a budget gap. That fight isn’t dead either — a sitting state senator published an op-ed in the Baltimore Sun on September 8 calling for its repeal, on top of the Chamber’s repeal push I told you about two weeks ago. Nothing has changed procedurally — no new vote, no new collections data — but the political pressure is building in public, not just in committee rooms.

What I’d do: Nothing today requires action from most of you. But if you buy IT services, keep confirming that 3% shows up as its own line item on invoices, and don’t assume either tax fight is settled — budget 2027 as if both stay exactly as they are, and treat any repeal as a pleasant surprise rather than a plan.

2. The Fed meets this week, and the data got hotter, not cooler

Last week I told you markets were pricing roughly 65% odds of a quarter-point rate increase at this week’s meeting. The two inflation reports that landed since then didn’t calm anyone down. August producer prices rose 0.4% for the month and 5.4% over the past year, with core prices (excluding food, energy, and trade services) up 0.3% for the month and 4.7% year over year — diesel fuel alone jumped 24.1%. The next day, August consumer prices rose 0.4% for the month, 3.4% over the year, with core CPI up 0.3% for the month and 2.4% year over year, driven partly by a 3.9% jump in gasoline.

That’s not the data set that talks a central bank into cutting. Rate-futures trackers moved accordingly, though — and I’ll be straight with you — they don’t agree by how much. CBS News reported CME FedWatch odds near 90% after the CPI print; a same-week Yahoo Finance piece put the number at 61% following the PPI report; an independent tracker had it at 83%. I couldn’t get a clean, single number straight off CME’s own site before deadline. Take the specific percentage with a grain of salt; take the direction — up, and meaningfully — as the real signal. The Federal Open Market Committee meets Tuesday and Wednesday, September 15–16, with the decision due Wednesday afternoon.

What I’d do: Same advice as last week, with more urgency. If you’re carrying a variable-rate note and there’s a fixed-rate alternative on the table, this is the week to take it seriously rather than let it sit in your inbox. If you were planning to draw on a line of credit for fall inventory or payroll, do it before Wednesday’s announcement, not after — you can’t undo a rate that’s already moved.

3. Three days left on the cheapest fixed-rate money in the state

I flagged this loan round two weeks ago and reminded you again last week; this is the last call. DHCD’s Small Business Direct Loan round — up to $2 million, a 4% fixed rate, terms up to 30 years — closes at 11:59 p.m. Thursday, September 17. I checked the agency’s own page again this morning: application status still reads “Open,” opened August 17, closes September 17. No extension, no early closure, no announced results yet.

The preference lanes haven’t changed: projects that rehab vacant or blighted property, create housing, expand fresh-food access, preserve or expand child care, or show clear community need get priority, and you need to be in a Sustainable Community or Priority Funding Area. Personal guarantees and collateral are required; working capital is eligible, a purely residential project is not.

Go back and read item 2. A 4% fixed note, in a week the Fed may raise rates, is not a form you skip because it looks like paperwork.

What I’d do: If you’re anywhere near a preference lane, this is the week — not “sometime this month.” Get your last two years of financials and a use-of-funds page together today, and if you haven’t already talked to DHCD’s business development team, call now. Three business days is enough time to apply. It is not enough time to also gather documents from scratch.

4. A “Maryland” optimism survey that isn’t really about Maryland

A few of you sent me a headline this month about Maryland small-business optimism cooling in August. I went and found the actual report, and here’s the honest version: it’s the national NFIB Small Business Optimism Index, published September 8, with a quote from Maryland’s NFIB state director bolted onto a syndicated version of the same release two days later. The Maryland-labeled page says so itself, if you read past the headline: “state-specific data is unavailable.” There is no separate Maryland index.

That doesn’t make the underlying number wrong, just mislabeled. The national index fell 1.1 points in August to 98.7, coming off July’s 99.8 — still above the 52-year average of 98.0. Labor quality remained the top problem owners named, at 23%, though that’s down from 27% in July. Inflation ticked up two points to 16%, tying with taxes for the second-most-cited concern. Maryland’s state director’s quote — “fewer sales and rising costs” — tracks the national mood; it just isn’t its own data point.

What I’d do: Read past the headline on any “state” data point before you make a decision based on it — this is exactly the kind of thing an AI-generated summary will flatten without telling you. And treat your own bookings ledger, not a national mood index, as your actual early-warning system. If labor quality is still your top headache, you’re not alone, but you’re solving your problem, not the national average’s.

5. The state built a free tool for anyone scouting a location or hunting for capital

Buried under the bigger stories this month: on September 3, Governor Moore relaunched the Maryland Community Business Compass as the expanded Maryland Community Compass, a free state platform at compass.maryland.gov. The original version launched in December 2025 and has already drawn more than 15,000 users and helped direct nearly $14 million to high-need areas, according to the governor’s office.

The expanded version adds a neighborhood-mapping tool for spotting development activity and permitted uses by area, a parcel-level “site insights” tool that tells you what’s actually buildable on a specific address before you spend money finding out the hard way, and a funding directory that aggregates more than 2,600 grant and loan programs from over 100 sources. It’s a genuinely practical stack for anyone doing what I talked about in past editions — scouting a storefront, checking zoning before signing a lease, or hunting for a capital-stack match on a project.

What I’d do: If you’re anywhere in the process of picking a location — new lease, expansion, second site — put in the address before you call a broker, not after. It won’t replace due diligence, but it’ll tell you in five minutes what used to take a week of phone calls to three different county offices.

The week ahead

• Today, Monday, September 14 — the 30-day window to appeal the digital ad tax ruling runs out. Watch for confirmation either way.

• Tuesday–Wednesday, September 15–16 — Federal Open Market Committee meeting. Decision Wednesday afternoon.

• Thursday, September 17, 11:59 p.m. — DHCD’s Small Business Direct Loan round closes.

• Friday, September 18 — Maryland’s August statewide employment report.

• Thursday, October 1 — the Worker Freedom Act (SB 417) takes effect for Maryland employers.

• Sunday, November 15 — the FAMLI Declaration of Intent window closes for employers weighing a private paid-leave plan.

See you next Monday.

Sources

• Digital ad sales tax struck down by court — Maryland Matters, August 14, 2026

• Maryland Tax Court Invalidates Landmark Digital Advertising Tax — Forvis Mazars, August 25, 2026 (tax rate structure, 30-day appeal window under Md. Code Ann., Tax-Gen. § 13-532)

• Maryland IT Tax Nets 77% Less Than Projected in First Year — Baltimore Sun via GovTech, September 1, 2026

• J.B. Jennings: Maryland must repeal its tech tax — Baltimore Sun guest commentary, September 8, 2026

• Producer Price Index News Release — August 2026 — U.S. Bureau of Labor Statistics

• Consumer Price Index News Release — August 2026 — U.S. Bureau of Labor Statistics

• Fed rate hike in September “all but guaranteed,” economists say — CBS News

• Fed Rate Hike Odds Rise to 61% After PPI Comes in Hotter Than Expected — Yahoo Finance

• Meeting calendars and information — Federal Reserve Board (September 15–16, 2026 FOMC meeting dates)

• Small Business Direct Loans — Maryland DHCD (live application status, deadline)

• New NFIB Survey: Main Street Optimism Cools in August but Holds Above Long-Term Average — NFIB, September 8, 2026 (national index)

• Main Street Optimism Cools in August but Holds Above Long-Term Average — NFIB Maryland state-news page, September 10, 2026 (same national data, local quote, “state-specific data is unavailable”)

• Governor Moore Announces Expanded Maryland Community Compass Platform to Help Community Developers Revitalize their Neighborhoods — Office of Governor Wes Moore, September 3, 2026

• MD Community Compass Platform expands, relaunches — Maryland Daily Record, September 4, 2026

• Maryland Community Compass — Maryland State Innovation Team

• Maryland Employment and Unemployment Press Release Schedule — Maryland Department of Labor

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice — I’m a guy from Baltimore with a spreadsheet and a bar card. Program details, dates, and rates can change; verify with the agency before you act.

Today my firm, Crimson Business Consulting, is publishing its first annual Consumer Outlook Analysis Report. It pulls together the official statistics, Federal Reserve surveys, credit-bureau data, corporate earnings disclosures, and market forecasts released through mid-September, and it adds a Maryland section on top of the national picture. Every figure in it is sourced and dated.

I’ll be direct about the conclusion, because the data is direct about it. The American consumer is weaker than the headline numbers suggest, the strength that does exist is concentrated in the top fifth of households and dependent on the stock market, and most of what’s scheduled for 2027 makes it worse. In no way has the current administration made any strides to show they care about these issues and if anything, believe the opposite is happening. This is not a “cautiously optimistic” report. Here are the findings.

1. Sales are up. Purchases are down.

Retail sales are 5% higher than a year ago, and that number gets repeated as proof the consumer is fine. It isn’t proof of anything except inflation.

Consumer prices are up 3.4% over the past year. Gasoline is up 27.4%. Once you take prices out, real retail growth is somewhere around 1 to 1.5%, and the actual unit data is worse than that. Circana’s back-to-school numbers showed general merchandise dollars down 1% and units down 2%, with discretionary units down nearly 4%. Restaurants are selling more dollars to fewer people: same-store sales up 1.8%, traffic down 1.9%. July retail sales fell 0.6% when economists expected a gain, the first monthly decline since last October.

The clearest read comes from small business owners themselves. In the NFIB’s August survey, a net negative 9% reported higher sales over the past three months. That’s the weakest reading since November. Meanwhile 31% of them are still raising prices. Higher prices on fewer units is the whole story of 2026, and it means any business measuring itself by revenue is overstating its own health.

2. Wages are losing to prices, and savings are nearly gone

Nominal wages are growing about 3.1%. Prices are growing 3.4%. Real average hourly earnings are down 0.3% from a year ago, per Friday’s BLS release, and the Employment Cost Index has real private-sector wages down 0.4%. The one group still ahead of inflation is people who switched jobs. Workers over 55 are furthest behind.

Households have kept spending anyway by saving less. The personal saving rate was 3.0% in July and touched 2.6% in June. Readings that low were last seen in the 2022 inflation spike and, before that, 2007. Real consumer spending in July grew by less than a tenth of a percent, and spending on goods fell outright.

The Fed’s own household survey found that only 63% of adults could cover a $400 emergency in cash. Bankrate found 29% now carry more credit card debt than emergency savings. Vanguard reported a record 6% of 401(k) participants took hardship withdrawals last year, with foreclosure or eviction the leading reason.

3. The good quarter was a tax refund

Second-quarter consumer spending grew at a 3.4% annual rate after growing just 0.5% in the first quarter. That looked like a recovery. It was a refund check.

The 2025 tax law made several deductions retroactive without updating withholding tables, so the entire tax cut arrived as spring refunds. The IRS paid out $296 billion, up 17%, with the average refund up 11% to $3,275. Goldman Sachs called the second quarter “the temporary byproduct of the tax refund surge” and expects real spending to slow to 1 to 1.5% in the second half. Withholding was fixed for 2026, so next spring’s refunds go back to normal. The boost was one-time, and it’s already been spent.

4. Sentiment is at the lowest level ever recorded

On Friday the University of Michigan’s preliminary September consumer sentiment index came in at 47.8. The previous record low, in a survey that goes back to the 1950s, was 50.0 in June 2022, when gas was five dollars and inflation was nine percent. The expectations component is 45.8. Year-ahead inflation expectations jumped from 4.0% to 4.6% in a single month.

The Conference Board’s expectations index has been below 80, the level it associates with a recession within a year, for nineteen consecutive months. The New York Fed’s survey found 38.6% of households saying their finances are worse than a year ago and 32.6% expecting further decline.

Sentiment has been a poor predictor of spending for three years, and the report says so. But it is a good description of the bottom half of households, who don’t own stock and are paying $4.22 for gas. And that leads to the most important finding.

5. The “resilient” consumer is a stock portfolio

If the median household is squeezed, who is driving the spending growth? The Fed’s July minutes answer in plain language: stock gains supported spending “particularly among higher-income households,” while “low- and moderate-income households were under increasing strains.”

Moody’s estimates the top 20% of households now account for about 60% of all consumer spending and hold about 90% of equities. The bottom half of the country holds 2.5% of household wealth, a share that hasn’t moved in five quarters. The S&P 500 is up about 20% in a year and the average Fidelity 401(k) hit a record $155,800. That money is real and it is being spent.

But it means the strong half of the consumer economy rests on equity prices, which the Fed’s own staff flagged in July as a downside risk. And even the upper half is behaving defensively. Dollar General reports trade-down from households earning over $100,000 has gone from “sporadic” to everyday. Walmart’s share gains came “particularly among higher-income households seeking value.” PwC’s holiday survey has gift spending down 2% and travel down 24%, with trade-downs “at every income tier.” When the people with money start shopping at the dollar store, that is not resilience.

6. Debt isn’t collapsing, but the cracks are where you’d expect

To be fair to the other side: household debt is $18.8 trillion, yet the New York Fed says delinquency rates “have held steady over the past two years,” card charge-offs at banks fell this year, and every major card issuer beat credit expectations in July. Debt relative to income is near its lowest since the 1990s. This is a squeeze, not a crash, and the report is careful about that distinction.

But the stress is concentrated exactly where a squeeze shows up first. Credit card rates average 22%. Subprime auto delinquency hit a multi-decade high in January. FHA mortgage delinquency is 11.8%, up more than a point in a year. More than 17% of student loan borrowers are 90 days past due now that reporting has resumed, and newly defaulted borrowers saw credit scores drop 91 points. Among buy-now-pay-later users, 47% paid late in the past year and 29% are using it for groceries. The average new-car payment is a record $777, and one buyer in five pays over $1,000 a month.

7. Most of 2027’s damage is already scheduled

This is the finding I’d most want a business owner to absorb. A large share of next year’s consumer squeeze is not a forecast. It has a date.

This Wednesday, September 16, the Fed meets, and markets put the odds of a rate hike around 70%. Three regional Fed presidents already dissented in July in favor of one. The Fed’s own projections have rates still at 3.6% at the end of 2027. Any plan that assumes cheaper money next year is unsupported.

On October 1, states begin paying 75% of SNAP administrative costs. On December 11, the federal funding resolution expires, with only three of twelve spending bills through the House and none through the Senate. On January 1, ACA premiums rise again; they already rose 58% this year when the enhanced subsidies expired and 3 million people dropped coverage, and insurers have proposed a median 14% more for 2027. In the first quarter, Medicaid work requirements and twice-yearly eligibility checks take effect, and the refund season normalizes. AARP estimates about 3 million people could lose SNAP as the cost shifts land through October 2027.

The prediction markets have read the same calendar. Polymarket puts the odds of a recession by the end of this year at 8%. By the end of 2027: 32%. Kalshi had 2027 at 41% back in May. Nobody can time a recession, but the market’s view that next year is four times riskier than this one is not a guess about the cycle. It’s arithmetic on the calendar above.

And the energy shock that’s driving the wage squeeze shows no sign of ending. Crude went back above $100 last week after U.S. strikes on Iranian-linked tankers. The government’s forecast has gasoline falling to $3.35 next year, but that forecast assumes Middle East output recovers by spring. Nothing in the news supports that assumption, and the report does not rely on it.

8. Maryland has the same problems plus a federal one

Everything above applies in every state. Maryland adds a layer.

We lost 24,900 federal jobs in 2025, the most of any state, roughly 18% of the federal workforce here. The state has added 29,800 jobs so far in 2026, and that recovery is real, but employment is still about 1% below a year ago. Maryland’s unemployment rate is now 4.2%, above the national 4.1%, up from 3.6% in early 2025. State wages grew 3.0% last year against 4.2% nationally; Montgomery County’s grew 1.6%. Federal contractor layoffs continue, with Amentum’s Hanover cuts effective at the end of October, and the December 11 funding deadline lands directly on this state’s largest employer.

Maryland households also carry more. Third-highest credit card balance per user in the country at about $9,000. First or second in student loan balances at about $45,500. Auto insurance 21% above the national average. About 290,000 BGE customers behind on their bills as of March, with BGE asking for another $8 a month starting in January. The seventh-worst foreclosure rate in the country.

The state’s own finances show the two-tier consumer clearly. Income tax receipts came in 9% higher last fiscal year, driven by a 20% jump in capital gains and dividends. Sales tax, which measures whether ordinary households are buying ordinary things, came in 1% over estimate. Comptroller Lierman said in December that high-income activity “can mask the struggles of those on the downward layer of the K.” Ocean City’s May hotel revenue fell 24%, one Boardwalk owner reported June sales down 28%, and statewide casino revenue fell 6% in August. And the state faces structural deficits of $2.6 to $3.4 billion a year starting in fiscal 2028, so there is little room to cushion any of this.

What the report recommends

The planning section of the report is written for owner-operated businesses, and it comes down to a handful of things.

Measure in units, not dollars. Put transaction counts next to revenue for the last two years. Build 2027 on flat units and 3 to 4% price. If the plan needs unit growth, name where it comes from.

Know which half of your customers you serve. Lower- and middle-income customers will show steady dollars, falling units, and high sensitivity to the first-quarter 2027 cliffs; give them a low-price way in. Upper-income customers are fine until the stock market isn’t; don’t add fixed capacity on the strength of their 2026.

Protect margin before chasing volume. Producer prices are up 5.4% and tariffs are passing through. Raise prices in small, explained steps rather than one large one later.

Hold cash and assume financing stays expensive. Model floating-rate debt at prime plus one. Keep six months of operating reserve. Pay down anything at 22%.

Retain rather than hire. Wage pressure in Maryland is strong and hiring intentions are negative. Keeping experienced staff costs less than replacing them in a low-hire market.

Plan for December 11 and the first half of 2027. If revenue depends on federal workers, contractors, or customers on Medicaid or SNAP, model a two-month disruption from mid-December and a step down in demand through the first half of next year. Defer fixed-cost commitments until the first-quarter data is in.

Treat the spring as a one-off. April and May were a refund. Do not use them as a run rate.

None of this is a prediction of collapse. Delinquencies are flat, layoffs are low, and the banks are fine. What the data describes is a consumer who keeps spending in dollars while buying less, financed by savings that are almost gone and by a tax refund that won’t repeat, with the strong half of the economy leaning on a stock market the Fed itself calls a risk and a calendar of scheduled cuts landing in the first three quarters of 2027. That is the outlook. Plan for it.

See the full report at crimsonbc.com.

Sources

• University of Michigan Surveys of Consumers — Preliminary September 2026

• Conference Board — Consumer Confidence, August 2026

• BLS — Consumer Price Index, August 2026

• BLS — Real Earnings, August 2026

• BLS — Employment Cost Index, June 2026

• BLS — Producer Price Index, August 2026

• BEA — Personal Income and Outlays, July 2026

• Census Bureau — Advance Monthly Retail Sales, July 2026

• Circana — Back-to-School Signals a New Reality

• Black Box Intelligence — Monthly Restaurant Update

• NFIB — Small Business Economic Trends, August 2026

• Federal Reserve — Economic Well-Being of U.S. Households in 2025

• Bankrate — 2026 Emergency Savings Report

• Vanguard — How America Saves 2026

• IRS — Filing Season Statistics, Week Ending April 17, 2026

• Yahoo Finance — Goldman Sachs Warns of Consumer Spending Slowdown

• Yahoo Finance — NY Fed Survey of Consumer Expectations, August 2026

• Federal Reserve — FOMC Minutes, July 28–29, 2026

• Fortune — Richest Consumers Powering U.S. Economy (Moody’s/Zandi)

• Federal Reserve — Distributional Financial Accounts, Bottom 50% Share

• Fidelity — Q2 2026 Retirement Analysis

• Dollar General — Q2 Earnings Call Transcript

• Walmart — Q2 FY27 Earnings

• PwC — Holiday Outlook 2026

• New York Fed — Household Debt and Credit, Q2 2026

• Federal Reserve — Consumer Credit G.19

• Mortgage Bankers Association — National Delinquency Survey, Q2 2026

• New York Fed — Federal Student Loan Defaults Return After Pandemic Pause

• Chain Store Age / LendingTree — BNPL Usage Survey 2026

• Edmunds — Q2 2026 New-Vehicle Financing

• Yahoo Finance — Fed Rate Hike Odds Hit 70%

• Federal Reserve — Summary of Economic Projections, June 2026

• CRFB — Appropriations Watch FY2027

• KFF — 2026 ACA Marketplace Premiums and Enrollment

• KFF — 2027 Preliminary Rate Filings

• AARP — SNAP and Medicaid Changes Timeline

• Polymarket — Recession Predictions

• Yahoo Finance — Kalshi 2027 Recession Odds

• NPR — Crude Oil Tops $100 a Barrel

• EIA — Short-Term Energy Outlook, September 2026

• Maryland Matters — Maryland Lost 25,000 Federal Jobs in 2025

• Maryland Department of Labor — July 2026 Labor Review

• Maryland Department of Labor — 2026 WARN Log

• BLS — County Employment and Wages in Maryland, Q4 2025

• WalletHub — Credit Card Debt by State

• EducationData.org — Student Loan Debt by State

• MoneyGeek — Average Cost of Car Insurance in Maryland

• Maryland Matters — BGE Rate Increase Protest

• Maryland PSC — BGE Seeks $156 Million Electric Rate Increase

• ATTOM — Foreclosure Rates by State, July 2026

• Southern Maryland Chronicle — Maryland FY2026 Closeout Report

• Maryland Comptroller — Board of Revenue Estimates, December 2025

• Maryland DLS — Effect of the 2026 Legislative Program on the Financial Condition of the State

• FOX45 / Baltimore Sun — Ocean City Tourism, July 2026

• Maryland Lottery and Gaming — August 2026 Casino Revenue

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or investment advice. All figures are as of the second week of September 2026 and most update monthly; the Fed meets Wednesday, August retail sales land the same day, and the Maryland Board of Revenue Estimates meets September 24. Verify with the source before acting. The full Consumer Outlook Analysis Report is available from Crimson Business Consulting.

Most small business owners can tell you what they spent on marketing last month. Far fewer can tell you whether it worked. In this episode of Crabcakes & Commerce, I break down the two numbers that answer that question: Customer Acquisition Cost (CAC), what it costs you to land a new customer, and Lifetime Value (LTV), what that customer is worth to you over the time they stick around.

In The Discussion, I walk through what each number is, how to calculate it with the figures you already have, and how to read them side by side to see if your marketing is paying for itself or just spending your cash. Then I run a real example [a local (type of business) spending $X to bring in customers who buy Y over Z months] to show how the math plays out and what to do when the numbers don’t look good.

If you’ve ever wondered whether that ad campaign was worth it, this one’s for you.

Share

Leave a comment

There are two numbers that decide whether a price makes you money or quietly loses it. Most people know one of them. Almost nobody tracks both. This is what they are and how they work together.

This is education, not financial advice. Your costs are yours — run the math with your own numbers.

Share

Let’s invent someone. Call him Lou.

Lou runs a bakery on a corner in Highlandtown. Twelve years in. He gets there at 4:30 in the morning and he knows, to the tray, how much he’s going to make, because his grandmother taught him that a baker who runs out is a baker who loses a customer. So he doesn’t run out.

Which means that at 6:40 every evening, twenty minutes before he locks the door, Lou is standing in front of a rack. Two dozen rolls. Half a sheet of the sour cherry thing that sells like crazy on Saturdays and sits on a Tuesday. A loaf and a half of rye. He has done the math on that rack more than once, and the math says it’s somewhere around forty dollars of ingredients and labor, every night, into a bag, into the dumpster, for which he also pays.

Lou isn’t real. He’s assembled from enough real people that if you own a food business in this city, you’ve stood in front of his rack.

Now here’s the part that matters. Twice a year somebody from a school, a church, or a Little League walks into Lou’s shop with a clipboard and asks him to “give back.” And Lou — who gives away nothing and throws out forty dollars a night — says he’d love to, but money’s tight.

He’s telling the truth. And he’s got it backwards.

Here’s What Most People Believe

Most of the advice a small business owner gets about community involvement assumes there’s a checkbook involved. Sponsor the team. Buy the table at the gala. Match employee donations. Write the check to the food bank at the end of the year, if the year went well.

None of that is wrong. It’s just built for a business that has a budget line for it, and most of the businesses I work with don’t. So the advice gets filed under “someday,” which is where it dies.

The better list — the one I actually want to hand you — starts from a different question. Not what can I give? but what am I already losing, and who could use it?

Because every business is leaking four things all the time: inventory it will throw away, hours when the lights are on and nobody’s buying, space that sits empty, and expertise that only ever gets used for paying clients. Each of those is a loss you’re already taking. Converting a loss you’ve already booked into something a neighbor can use isn’t charity. It’s just a better use of a write-off.

Here’s the company that made me see it.

The Surprise Bag, Because I’m a Numbers Guy

Too Good To Go is an app. A bakery, a grocer, a restaurant, a cafe lists a “Surprise Bag” of whatever it didn’t sell that day at 50 to 75 percent off what it would have charged. Customers reserve it in the app, show up at closing time, and take home a mystery. The business gets a monthly payout for food it was about to pay someone to haul away. The company says it has more than 180,000 partner businesses and 120 million registered users worldwide, and INSEAD wrote it up as a business-school case for exactly the reason I’m about to give you.

Notice what’s missing from that description. Nobody is being noble. The business isn’t donating anything. The customer isn’t doing anyone a favor. The landfill is the only party that gets less.

And now look at the two numbers the company puts on its own partner page, because they’re the real story. Too Good To Go reports that 41 percent of customers bought something else while picking up their bag, and 61 percent said they only visited the business because of it.

Those are company-reported figures, so treat them the way you’d treat any number a company puts on its own sales page. But even discounted, they say something the “meals saved” counter on the home page doesn’t: this is a customer-acquisition channel wearing a waste-reduction costume. Six in ten of the people walking in at closing time had never been in before. Four in ten spent money on something that wasn’t on sale.

Lou doesn’t have a marketing budget. Lou has a rack. It turns out the rack is the marketing budget.

Now, I can hear you: “Mike, why would I hand an app a piece of every bag? I’ll put a chalkboard on the sidewalk that says ‘everything in the case is half off after 7.’”

Do that. Seriously — that’s Too Good To Go with no middleman, and I’ll tell you to do it below. But understand what the chalkboard can’t do. The chalkboard only reaches people who were already walking past your door. The app reaches people who were on their couch looking for a cheap dinner and had never heard of you. The whole value of that 61 percent number is that those were new feet. Which one you want depends on whether your problem is the rack or the foot traffic. For a lot of businesses, it’s both.

And that’s just Tier 1. The same logic runs through every item that follows.

The Same Logic, Without the App

You don’t need to be a bakery for this to apply. Look at what’s been built on top of the principle that the thing you’d throw out has a buyer at the right price: Flashfood does it for groceries near their sell-by date, out of a dedicated fridge in the store. Misfits Market and Hungry Harvest — Hungry Harvest is a Mid-Atlantic operation, so a neighbor — sell the ugly produce. Martie sells overstocked shelf-stable goods at a steep discount.

None of those are things a single storefront can join. That’s not why I’m listing them. I’m listing them because a whole category of companies has been built on buying what other businesses were going to landfill. Somebody has already found the buyer for the thing you’re throwing out.

So translate it. A florist’s Friday bouquets. A hardware store’s open-box returns. A boutique’s end-of-season rack. The printer’s misprints. A contractor’s leftover materials after a job. Every one of those is a Surprise Bag with a different label, and none of them need an app. They need a shelf, a sign, and a price.

The Two Fears That Stop People

If the sale doesn’t happen — the rack’s still there at 7 — then the next best thing is to give it away. And here is where most owners freeze, for two reasons, both of which have been answered for years.

“What if somebody gets sick?” This is the one I hear most, and it’s the one Congress dealt with in 1996. The Bill Emerson Good Samaritan Food Donation Act gives good-faith food donors to nonprofits limited protection from liability. It covers restaurants, grocers, caterers, farmers, hotels, and more. The fear that stops most owners from donating has been legally addressed for three decades. I’m not saying it’s a blank check — you still have to act in good faith and hand off food that’s actually fit to eat. I’m saying the thing you’re worried about was the whole point of the statute.

“I can’t afford to give it away.” Two answers. The first is that you were about to pay to throw it out, so “afford” is doing a lot of work in that sentence. The second is the tax code, and this one surprises people.

Under Section 170(e)(3), a business that donates food to a qualified nonprofit can take an enhanced deduction — not just what the food cost you, but your cost plus half the difference between cost and fair market value, capped at twice your cost. For years that was C-corporations only. The PATH Act of 2015 made it permanent and opened it up to S-corps, LLCs, partnerships, and sole proprietors. Most small owners assume a donated loaf is worth its flour and labor on their return. It can be worth more than that.

Confirm the math with your accountant — there are documentation requirements, and I’m not your CPA. But go in knowing the enhanced deduction exists, because a lot of accountants who don’t work with food businesses have never had a reason to look it up.

“I don’t have time to drive it anywhere.” Fine, a third fear, and this one’s the easiest. MealConnect, run by Feeding America, is a free logistics layer for exactly this problem. You post what you have with a photo and a pickup window. A vetted local nonprofit accepts it and sends its own people to pick it up. No minimums, no fee to the donor — cases, banana boxes, prepared food, refrigerated or frozen. It has a built-in tax-benefit calculator. It has moved more than five billion pounds of food.

You post. They drive. That’s the whole workflow.

The Mechanics That Run Themselves

Everything above is about inventory. Now the other three leaks — hours, space, expertise — and the beautiful thing about the mechanics in this section is that most of them are funded by someone other than you.

Suspended coffee. A customer buys two coffees and drinks one. The barista logs the other as “suspended,” and anyone who needs it can claim it, no questions, no story required. It works for any menu item — a sandwich, a slice, a bowl of soup. The Boston Globe wrote up a cafe that runs it as a wall of sticky notes. The business spends nothing. It’s customer-funded, and the owner sets the rules. And it does two things at once: it’s a dignity mechanism, because the person receiving orders like everyone else, and it’s a loyalty mechanism, because the person giving now has a reason to feel good about your shop specifically.

Round up at the register. A Saint Louis University study found that asking customers to round up to the nearest dollar pulls in roughly 21 percent more than asking for a whole-dollar donation, because losing 40 cents doesn’t register as a loss. It works best when the cause matches what you sell and the customer already likes you. One thing to be clear-eyed about, and I’ll come back to it in the fine print: the customer is the donor here, not you. You’re the plumbing. A small business that names a specific local cause and posts the running total on the wall gets the goodwill; a business that’s vague about where the money goes gets the eye-roll people reserve for the big chains.

Lend the room. Empty dining room on a Monday? Back office after 6? Nonprofits, tutoring groups, neighborhood associations, and every volunteer board in the city are chronically short on free meeting space. Repair Cafés and tool libraries often start in exactly this kind of borrowed room. Your cost is unlocking the door and turning on the lights. Your dead hours are somebody else’s only hours.

Host the fridge. A community fridge is a public refrigerator, usually outside a business or a church, stocked by neighbors and nearby shops. The fridge is typically donated; what you contribute is an outlet, a wall, and some electricity. Call your county health department before you plug one in — the rules vary by jurisdiction and I’m not going to pretend I know yours.

Lend the skill. This is the one with the best return and the least uptake. The bookkeeper does a food pantry’s books one Saturday a quarter. The restaurant teaches the shelter kitchen how to batch-cook. The marketing shop builds the Little League’s website in an afternoon. The U.S. Chamber reports that 90 percent of companies doing skills-based volunteering saw lower turnover, and cites an NFIB survey in which 82 percent of small owners said community involvement improves the local business climate and about 70 percent said it matters for employee morale. Two hours of your actual expertise is worth more to a nonprofit than a $200 check almost every time — and it costs you two hours you’d have spent on something else.

Give the byproduct to the neighbor who wants it. Coffee grounds to the community garden. Sawdust to the animal shelter. Pallets and boxes to whoever’s moving this weekend. The U.S. Chamber’s piece on monetizing waste makes the right first move: do a waste audit. Literally look in the dumpster for a week. You can’t redirect what you haven’t counted.

Give the product, not the cash. This is SCORE’s number one on its list of ways to give back, and it should be. A $60 dinner certificate costs a restaurant something like $20 in food. The nonprofit values it at $60. The couple who wins it at the silent auction values it at $60. And they show up, order a bottle of wine, and maybe come back. That’s the cheapest marketing most owners never buy.

And two more from the same list that cost nothing if you were already spending: buy from the small business down the street instead of the chain — catering, printing, holiday gifts, services — because it keeps the dollar in the zip code at zero incremental cost. And let the team pick the cause for whatever small pool you do set aside. SCORE files that under marketing; the turnover numbers above say it’s really an HR line.

The Maryland Minute

If you’re going to do the donation route, you’ve got partners here. The Maryland Food Bank and Food Rescue Baltimore are the obvious first calls for anyone with surplus food in this region.

If you’re in the city, the Baltimore Food Matters program — a partnership between NRDC and the Office of Sustainability — has a directory of food-scrap haulers, offers composting training, and has made grants of up to $10,000 to local partners running waste-reduction projects. The city’s stated target is to cut food waste in half by 2030. Grant programs change and the details on that page are the ones to check.

And if you farm — this one’s farms only, not restaurants or retailers — the Maryland Farm Food Donation Tax Credit is worth up to $5,000 a year at 100 percent of wholesale value for donating to certified food banks. More on that below, because it’s a useful lesson in what a policy can’t do.

The Fine Print, Because I’m Also a Lawyer

Let me argue the other side, because a piece like this is worthless if I don’t.

The 41 and 61 percent are the company’s numbers. Too Good To Go has every reason to put its best figures on its partner page, and I haven’t seen them independently verified. The reason I’m comfortable using them is that the direction is obviously right — a person who reserved a bag from their couch was not otherwise walking in — even if the magnitude is softer than the sales page says. Before you list, ask a local partner business what their pickup crowd actually looks like. Ask what the app keeps per bag, too. I don’t have a number I trust on that, so I’m not printing one.

The round-up isn’t your deduction. When a customer rounds up, the customer is the donor. You’re passing the money through, and you don’t get to deduct it. Anybody who tells you otherwise is about to get you a letter. The benefit to you is the goodwill and the loyalty, which is real but is not a line on your return.

The enhanced deduction has paperwork. Section 170(e)(3) is real and it’s better than most owners think, but it comes with documentation rules, and the value depends on your entity type and your numbers. Everything I said above is the argument for a conversation with your accountant, not a substitute for one.

Policy doesn’t fix logistics. The Maryland farm tax credit is a good idea that critics say mostly rewards farms that were already donating and does nothing about the actual bottleneck, which is getting food from the field to the food bank. That’s not a knock on the credit. It’s the general lesson: the constraint on donation is almost never the incentive. It’s the truck. That’s why MealConnect, which is only a truck, matters more than any deduction.

Some of this needs a phone call to the county. The fridge especially. Health-code treatment of community fridges varies, and “a guy on Substack said it was fine” is not a defense I’d want to run.

And a chalkboard has a ceiling. The do-it-yourself closing-time deal is great and free. It also trains your regulars to wait until 7. Watch your full-price sales for a month after you start it, and if the rack is emptying because the 5 o’clock crowd became the 7 o’clock crowd, you haven’t reduced waste — you’ve cut your price.

That’s the honest case against. None of it changes what I’d tell Lou.

What I’d Actually Do

Look in the dumpster for a week. Before you pick a program, count what you’re losing. Inventory, yes, but also hours when nobody’s in, rooms nobody’s in, and skills nobody’s paying you for. Write it down. You’ll be surprised which column is biggest, and it’s usually not the one you expected.

Pick the loss, not the cause. Most people start by choosing a charity and then figuring out what to give it. Reverse that. Start with the biggest thing you’re already losing, and then find the neighbor who can use it. The match will be obvious once you’re looking at it from the right side.

Make two phone calls. One to your accountant, to ask specifically about the Section 170(e)(3) enhanced deduction for your entity type. One to your county health department, if anything you’re planning involves food outside your four walls. Both calls take fifteen minutes and both remove the excuses that have been stopping you.

Sell the surplus first, give it second. Chalkboard at 7, or the app, or both. What doesn’t sell gets posted on MealConnect on your way out the door. That’s the whole nightly routine: sell, post, lock up.

Put the customer-funded stuff on the counter. A suspended-coffee board, a round-up ask for one named local cause, a running total on the wall. None of it costs you a dime and all of it makes your regulars feel like your shop is theirs. That’s the loyalty you were going to try to buy with a punch card.

Give two hours, not two hundred dollars. Once a quarter, put your actual professional skill to work for one organization down the street. It’s the highest-return thing on this list and the one almost nobody does.

Lou’s still standing in front of the rack at 6:40. Same two dozen rolls, same half-sheet of sour cherry.

The difference is that now three of those rolls are in a bag somebody reserved from their couch in Canton, and they’re going to walk in for the first time to pick it up. The rye’s going to Food Rescue Baltimore because he posted it on his phone while the ovens cooled. The clipboard people got a $40 gift certificate for the raffle, which cost him twelve dollars in flour and put a new couple at the counter on a Saturday.

He didn’t write a check. He just stopped paying to throw away the thing his neighbors wanted.

Around here, that’s the closest thing to free money there is.

Business Built Together,

Mike DeBaugh Jr. Crimson Business Consulting, Baltimore, MD

Leave a comment

Share

Sources

Too Good To Go and the “sell the surplus” model

• Too Good To Go, Surplus Food Marketplace (partner page) — 41% / 61% customer figures; 180,000+ partners; 120M+ users. Company-reported.

• Too Good To Go, How the app works — Surprise Bag discount range and pickup mechanics.

• INSEAD, Too Good To Go: A Surprise Bag Creates a Win for Business and Environment (case study).

• BagRescue, 8 apps like Too Good To Go (US) — Flashfood and other alternatives.

• one5c, 4 apps that fight food waste — Misfits Market, Hungry Harvest, Martie.

Liability and tax treatment of food donation

• USDA, Federal incentives for businesses to donate food — IRC §170(e)(3) enhanced deduction; PATH Act of 2015 expansion.

• USDA, Donating — Food Loss and Waste.

• Feeding America, Bill Emerson Good Samaritan Food Donation Act.

• National Restaurant Association, “Good Samaritan” Act eases ability to donate healthy, uneaten food.

• MealConnect, How it works.

• Feeding America, MealConnect: 5 billion pounds rescued.

Maryland and Baltimore

• Baltimore Office of Sustainability, Baltimore Food Matters Program — hauler directory, composting training, grants up to $10,000, 2030 target.

• Food Rescue Baltimore.

• Maryland Food Bank.

• Maryland Department of Agriculture, Maryland Food Donation Program — farm food donation tax credit.

• The Fulcrum, Maryland Farm Food Donation Tax Credit analysis.

Customer-funded and no-cost mechanics

• Suspended Coffees, How it works.

• The Boston Globe, Through pay-it-forward wall, Ula Cafe serves up coffee with a side of kindness (Sept. 8, 2021).

• Marketplace, More businesses are asking us to round up for charity (Mar. 28, 2022) — Saint Louis University research (Katie Kelting), ~21% lift.

• Patriot Software, 6 facts about checkout donations — customer, not business, is the donor.

• Wikipedia, Community fridge.

• Wikipedia, Repair Café.

• The Tool Library (Buffalo, NY), FAQ.

Skills, byproducts, and buying local

• U.S. Chamber of Commerce CO—, Skills-based volunteering benefits — 90% turnover figure; NFIB survey (82% / ~70%).

• U.S. Chamber of Commerce CO—, How to cut and monetize business waste.

• SCORE, 10 ways small businesses can give back to the community.

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice, and I’m not your lawyer or your accountant — I’m a guy from Baltimore with a spreadsheet and a law degree, in that order. The tax figures come from the USDA’s summary of federal law and the specifics depend on your entity and your books; the Baltimore grant program and the app’s terms both change. Verify with the source, and with your CPA, before you act on any of it.

Subscribe now

Coffee’s on. Let’s go through the books.

It’s Labor Day, which means most of you are reading this on the porch and the state’s offices are closed. Good. Because the week that starts tomorrow has more dated deadlines packed into it than any week since spring, and I’d rather you see them now than on Wednesday.

On the ledger this week: Canada’s counter-tariffs on $27.6 billion of U.S. goods take effect at 12:01 tomorrow morning; the state quietly flipped the switch on FAMLI employer registration; Maryland’s 3% tech tax brought in less than a quarter of what Annapolis promised, and the invoices have started to sting; the Fed meets next week with markets calling a rate hike a better-than-even bet; a new “captive audience” law lands October 1 with a $25,000 penalty attached; and there are ten days left on DHCD’s 4% fixed-rate loan round.

1. Canada shoots back at 12:01 a.m. tomorrow

Last week I told you it was coming. Here’s what’s in it. Canada’s counter-tariffs on U.S. goods take effect at 12:01 a.m. on Tuesday, September 8, covering $27.6 billion in imports from the U.S. at rates of 15%, 25%, or 50% depending on the product. Ottawa is matching, dollar for dollar, the 50% U.S. tariff on Canadian goods that took effect August 22. The targeted sectors are steel, dairy, appliances, agricultural equipment, pulp and paper, and electronics — and the list runs to 629 tariff lines, from corrugated boxes and envelopes to bread dough and toilet paper. Goods already in transit to Canada when the clock strikes are exempt; goods that leave your dock Tuesday are not.

Why this matters more here than the national coverage suggests: Canada is Maryland’s single largest export customer, at $2.2 billion in 2025 — 13% of everything the state ships abroad. The retaliation list was built to hit swing-state exports, not ours specifically, but it’s broad enough that a Baltimore County fabricator or an Eastern Shore food processor with a Canadian account should assume they’re on it until they’ve checked. And the other direction has been live for two weeks: the U.S. tariffs on Canadian goods carry no exemption for goods that qualify under USMCA, so Canadian-sourced inputs you’ve been buying since August 22 already cost more, whether or not your supplier has said so yet.

One caveat I’ll state plainly: the U.S. tariffs were imposed under a section of the 1930 Tariff Act that has never been used before, and trade lawyers expect it to be challenged in court. So this could get unwound. It could also escalate. Neither outcome is something you can put on a calendar.

What I’d do: If you sell into Canada, pull your HS codes and check them against Ottawa’s list today — it’s organized by tariff code, so it’s a thirty-minute job, not a project. If you buy Canadian inputs, call the supplier this week and ask for the landed-cost math in writing; the “in transit” exemption means some shipments dodge it and some don’t, and you want to know which of yours are which. And on both sides, this is the moment to read the price-escalation and change-in-law clauses in your contracts — the ones you skipped when you signed — because that language decides who eats a 25% swing.

2. FAMLI registration is live. This is the part you actually have to do.

Last Monday I flagged the September 1 opening of the Declaration of Intent window for employers considering a private paid-leave plan. The bigger news came the same day, with less fanfare: employer registration for FAMLI is now open, and the state is telling employers to register “as soon as possible.” That applies to every employer with at least one Maryland-based employee. No headcount floor, no exemption for the one-person LLC with a part-time helper.

The state’s own summary of the money is worth reading twice. Contributions are split between employer and employee; you can withhold up to half from your people via payroll. If you have fewer than 15 employees, you qualify for the small-employer discount, which exempts you from the employer’s share entirely — the state says more than 80% of Maryland employers will qualify. You still register, you still withhold the employee half, and you still file. The dates that follow are firm: withholding starts with the first pay period in January 2027, the first contribution payment is due April 30, 2027, and benefits — up to 12 weeks of job-protected leave at up to $1,000 a week — begin in January 2028.

The honest read: for most shops under 15 people, this is a payroll-deduction line and a quarterly filing, not a new cost. The risk isn’t the money. It’s finding out about the deduction from an employee who noticed it before you explained it.

What I’d do: Register at paidleave.maryland.gov this month, while it’s a ten-minute chore instead of a December scramble. Then write the one-paragraph employee notice now — what FAMLI is, what comes out of the check starting in January, what it buys them in 2028 — and put it in the drawer. If your payroll provider handles filings, sign the power of attorney early; the state’s been clear that the provider can’t register you, only you can.

3. The tech tax missed its number by $387 million. Your invoices didn’t.

The Comptroller’s Office has now put a preliminary first-year figure on Maryland’s 3% sales tax on IT and data services, the one that took effect July 1, 2025 to help close a $3.3 billion budget hole. The estimate at passage was about $500 million for fiscal 2026. Actual collections: $112.8 million, from roughly 2,900 companies — a shortfall of about 77%, per the Baltimore Sun’s reporting last week, with the most recent quarter still being processed. The office says the original forecast was made “before complete data were available,” which is a polite way of saying nobody knew how many transactions would actually land inside the NAICS codes the law taxes.

To be fair to the forecasters, this isn’t a September surprise. Maryland Matters reported back in March that first-quarter collections had the tax on pace for roughly $50 million — about a tenth of the estimate — so the full year actually came in better than the early read. What’s new is the full-year number and the owner-level detail: a Columbia managed-services firm says the tax costs it an additional $15,000 a month and has pushed it to raise prices and trim services, including cybersecurity; a Frederick computer shop says an employee spends about a day a week hand-building invoices because its billing software still can’t handle the tax. The Maryland Chamber is calling for repeal. A Hagerstown IT provider says some clients have moved billing addresses out of state — which, if it’s happening at scale, would explain a chunk of the shortfall better than any forecasting error. Nobody has the data to say whether it is.

Here’s the part I’d underline. A tax that raises $113 million instead of $500 million doesn’t get repealed in a budget year; it gets replaced or widened. Add the digital ad tax refunds now on the table after last month’s ruling, and Annapolis walks into the 2027 session looking for something north of half a billion dollars. The CNBC ranking that put Maryland’s economy 49th of 50 this summer — the figure the Salisbury chamber was citing on the Shore this week — is the argument against. The budget math is the argument for. I don’t know which wins.

What I’d do: If you buy IT services — and you do, even if it’s just the person who “does the computers” — pull a year of those invoices and confirm the 3% is broken out as a line item, not buried in the rate. It’s deductible either way, but you can’t manage what you can’t see. If you sell them, and your customers use the service in multiple states, the multiple-points-of-use certificate is the mechanism that keeps you from charging Maryland tax on Pennsylvania work; if you’re not using it, you’re overcharging and your customers will eventually figure that out. And budget 2027 assuming the tax stays. If it goes, that’s found money.

4. The Fed meets next week, and the market is betting on a hike

For two years the question was when the Fed would cut. It isn’t anymore. After Friday’s jobs report — 162,000 jobs added in August, unemployment steady at 4.1%, roughly triple what forecasters expected — rate-futures traders put the odds of a quarter-point increase at the September 15–16 meeting at about 65%, up from roughly 50% the day before. The odds have swung from 35% to 64% and back over the past three weeks on the strength of individual speeches, so treat that figure as a mood, not a forecast. The target range sits at 3.50–3.75%. Inflation on the Fed’s preferred gauge is running around 3.7%, and three officials dissented in favor of a hike at the July meeting. The two data points that decide it — producer prices on Thursday, September 10 and consumer prices on Friday, September 11 — land before the committee sits down.

One wrinkle worth knowing: early write-ups of Friday’s report circulated with several different headline numbers, and at least one is still wrong. The BLS figure is 162,000, and June and July were revised up by a combined 55,000. That is not a labor market asking for cheaper money. Maryland’s own July numbers, 11,700 jobs added and unemployment at 4.2%, say the same thing locally.

Why you care: most small-business credit in this state is priced off prime, and prime moves the day the Fed does. A quarter point on a $500,000 variable-rate SBA 7(a) is about $1,250 a year — not ruinous, but the direction is the story. If the market is right, the cheapest money of this cycle is behind you.

What I’d do: Look at every variable-rate note you carry this week and ask a simple question: is there a fixed-rate alternative available right now that I’d regret not taking? Item 6 below is one. And if you were planning to lean on a line of credit for fall inventory, draw it before the 16th, not after — the rate on an existing draw is what it is; the rate on a new one may not be.

5. October 1: the “captive audience” law, and the poster you don’t have yet

Among the dozens of laws that take effect on October 1, one is aimed squarely at how you talk to your employees. SB 417, the Worker Freedom Act, prohibits employers from taking adverse action against an employee or applicant who declines to attend a meeting about the employer’s views on religious or political matters — and “political matters” expressly includes union organizing. It takes effect Thursday, October 1. Violations can draw civil penalties up to $25,000, plus back pay, reinstatement, and attorneys’ fees. Voluntary meetings are fine. Required training on legal compliance — harassment, safety, discrimination — is expressly exempt.

There’s a second obligation with a lag: you’ll have to post a notice of employee rights under the law and hand it to new hires, but that requirement kicks in only once the Commissioner of Labor and Industry publishes a model notice, which is due by November 1. So for October you’re bound by the law but can’t yet post the poster. Welcome to Maryland.

Two honest caveats. First, most owners reading this have never held a mandatory meeting to share their opinion on a ballot question, and the law changes nothing for you. Second, similar laws in California and New York have been temporarily blocked by federal courts on preemption grounds, and Maryland’s version will likely face the same challenge. I wouldn’t plan around it being struck down; I would expect the enforcement picture to be muddy for a while.

What I’d do: Add one sentence to the handbook before October 1: attendance at any meeting concerning the company’s views on political or religious matters is voluntary, and no one will be penalized for skipping it. Then actually mean it. If you have an all-hands where you talk about a referendum, a candidate, or a union drive, say out loud at the start that people are free to leave. And put “Maryland Worker Freedom Act poster” on the November checklist so it’s not the December surprise.

6. Ten days left on the cheapest fixed-rate money in the state

I mentioned this round last week; the calendar has moved. DHCD’s competitive round of Small Business Direct Loans — up to $2 million at a 4% fixed rate, terms up to 30 years — closes at 11:59 p.m. on Thursday, September 17. The application status on the agency page is “Open” as of this morning, and the round is worth $14 million statewide, which means it will be competitive and the preference categories will do the sorting: projects that rehab vacant or blighted property, create housing, expand fresh-food access, preserve or expand child care, or show clear community need. You need to be in a Sustainable Community or Priority Funding Area, and personal guarantees and collateral are required. Working capital is an eligible use; a pure residential project is not.

Read item 4 again and then read this one. A 4% fixed note in a market where the Fed may be raising is not a program to skim past because the form looks like work.

What I’d do: If you’re anywhere near one of the preference lanes, book the initial consultation with DHCD’s business development team this week — it’s a form, not a pitch, and it gets you a human before the deadline. Have your last two years of financials, a use-of-funds page, and a paragraph on community impact ready. If you’re a child-care operator or you’re putting a vacant storefront back into use, you are exactly who this round was written for.

Share

The week ahead

• Tuesday, September 8, 12:01 a.m. — Canada’s counter-tariffs take effect.

• Thursday, September 10 — August producer price index. Friday, September 11 — August consumer price index. The last inflation readings before the Fed meets.

• Sunday, September 13 — the 30-day window to appeal the August 14 digital ad tax ruling runs out; falling on a weekend, expect any filing by Monday the 14th. An appeal freezes the refunds.

• Tuesday–Wednesday, September 15–16 — Federal Open Market Committee meeting. Decision Wednesday afternoon.

• Thursday, September 17, 11:59 p.m. — DHCD Small Business Direct Loan round closes.

• Friday, September 18 — Maryland’s August statewide employment report.

• Thursday, October 1 — Worker Freedom Act takes effect. Sunday, November 15 — FAMLI Declaration of Intent window closes for employers pursuing a private plan.

See you next Monday.

Sources

• List of products from the United States subject to counter-tariffs effective September 8, 2026 — Department of Finance Canada (effective time, $27.6 billion scope, rates, sectors, in-transit exemption)

• Canada’s New Retaliatory Tariffs on US Goods: What Canadian Businesses Need to Know — Fasken, August 27, 2026 (August 22 U.S. tariffs, no USMCA exemption, Section 338, contract clauses)

• Maryland’s Top 10 Exports — World’s Top Exports (Canada as top export destination, $2.2 billion, 13% of 2025 exports; based on U.S. Census USA Trade Online data)

• Governor Moore Announces Employer Registration for Paid Family and Medical Leave Insurance Is Live — Maryland Department of Labor, September 1, 2026

• For employers — Maryland FAMLI — Maryland Department of Labor (registration requirement, TPA limits)

• Maryland IT Tax Nets 77% Less Than Projected in First Year — Baltimore Sun via GovTech, September 1, 2026 ($112.8 million, ~2,900 companies, owner quotes, Chamber repeal call)

• Initial returns on IT and data sales tax falling short of estimates — Maryland Matters, March 11, 2026

• Maryland releases emergency rules and guidance for sales tax on technology services — PwC (multiple-points-of-use certificates; links to Comptroller Technical Bulletins 54 and 56)

• DHCD Announces Small Business Lending Program, Competitive Round for Low-Cost Loans — Maryland DHCD, July 8, 2026 ($14 million round)

• Maryland is No. 36 on Top States for Business 2026 — CNBC, July 9, 2026 (economy category rank of 49)

• Chamber Chat: Maryland Ranking Low for Small Business — WMDT 47ABC, September 4, 2026

• The Employment Situation — August 2026 — U.S. Bureau of Labor Statistics, September 4, 2026 (162,000 jobs, 4.1%, revisions)

• Rate-Futures Traders Add to Bets on September Fed Hike After Bigger-Than-Expected August Jobs Gain — Roic News, September 4, 2026 (hike odds, target range, July dissents, PPI/CPI dates)

• Maryland employment rose by 11,700 jobs in July — Maryland Department of Labor, August 21, 2026

• Maryland enacts new labor-friendly laws — Brightmine (SB 417 effective October 1, 2026; poster due by November 1; CA/NY injunctions)

• Forthcoming Maryland Laws Target Employer Practices and Labor Rights — Ogletree Deakins, April 14, 2026 (SB 417 penalties, exemptions, notice requirement)

• Small Business Direct Loans — Maryland DHCD (application window, terms, eligibility, preferences)

• Digital ad sales tax struck down by court — Maryland Matters, August 14, 2026

• Maryland Employment and Unemployment Press Release Schedule — Maryland Department of Labor

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice — I’m a guy from Baltimore with a spreadsheet and a bar card. Program details, dates, and rates can change; verify with the agency before you act.

Every owner wants the same thing from this question: a yes or a no. Preferably a yes. What federal law offers instead is six factors, no scoring system, no passing grade, and a cheerful reminder that not one of them decides anything on its own. It’s the legal equivalent of asking your doctor whether you’re fine and being handed a list of six things to consider.

So this week I sat down and talked through all six — what each one is actually asking, where owners talk themselves into the wrong answer, and why the agreement you both signed carries considerably less weight than you were hoping.

Fair warning: nobody finishes this test feeling great about it. But there’s a meaningful difference between not knowing where you stand and knowing exactly where you stand and not loving it. Only one of those is fixable.

The six factors:

1. Opportunity for profit or loss depending on managerial skill

2. Investments by the worker and the potential employer

3. Degree of permanence of the work relationship

4. Nature and degree of control

5. Extent to which the work is integral to the business

6. Skill and initiative

Leave a comment

Let’s invent someone. Call him Dave.

Fourteen years running a mechanical contracting outfit out of Essex. Nine trucks, eleven guys, a yard off Eastern Boulevard with a fence he keeps meaning to fix.

He learned the trade from a guy who learned it from a guy, and for about nine of those fourteen years the phone rang enough that he never had to think very hard about anything.

Then it got tight.

So Dave did what you’re supposed to do. He read the book. He went to the two-day thing at the hotel by the airport. He came home with a strategy, and it was a good one:

• Stop chasing one-off emergency calls and build a book of commercial preventive-maintenance contracts. Recurring revenue. Predictable schedule. Higher margin. Every consultant in America would have signed off on it.

That was two years ago. Revenue is flat. He has six maintenance contracts and four of them came from guys he already knew.

Dave isn’t real. He’s assembled from enough real people that if you own a service business in this state, you have an idea of who this could be.

Here’s what I want to argue: there is nothing wrong with Dave’s strategy.

It’s a good strategy. It’s such a good strategy that I’d bet real money a few hundred shops within an hour of his yard are chasing the same thing this quarter. That’s a guess, and I’ll flag it as one. What follows isn’t.

Here’s What Most Owners Believe

The belief goes like this: if the numbers are flat, I need a better idea.

A sharper niche. A different customer. A repositioning. Some angle the other guys haven’t found. It’s a completely reasonable belief and I want to state it fairly, because it comes from a real place — the successful companies you read about all have a Story, and the Story is always about an insight. Somebody saw the thing nobody else saw.

So when your own numbers go sideways, the natural conclusion is that you’re missing an insight.

I don’t think you are. I think the insight is the cheap part.

The Numbers, Because I’m a Numbers Guy

Start with how crowded the room actually is.

Per the Census Bureau’s 2023 data — combining businesses with employees and businesses without, which are counted in two separate datasets — Maryland has:

• 5,548 plumbing, heating and air-conditioning contractors

• 9,627 landscaping businesses

• 19,189 businesses whose job is cleaning things

• 18,138 residential construction businesses

• 13,441 restaurants and other eating places

• 15,266 beauty salons

Total, all industries:

• 742,017 Maryland businesses.

  • Of those, 599,050 have no employees at all.

Narrow it to Dave’s actual competitive geography — the Baltimore–Columbia–Towson metro — and that same category holds 2,188 businesses:

• 982 with employees.

• 1,206 without.

One honest note before that number does any work: the Census bucket bundles residential plumbers in with mechanical contractors, so 2,188 trucks are not all circling the same rooftop unit. But a good few hundred of them are close enough to Dave’s work that they’d take his maintenance contract if he offered it Tuesday.

Now the survival side. BLS tracks private-sector establishments from the year they open, March to March. Here’s where the Maryland cohorts stood in March 2025 — and read the first column, because each row is a different birth year, not one group followed for a decade:

That 50.1% belongs to the businesses born in the twelve months ending March 2020, which means their fifth birthday and a global pandemic arrived together. Weight it accordingly.

And now I can hear you: “Mike, is Maryland just a bad place to run a business?”

Not really — and I’d rather lose the cheap version of my argument than keep it. Maryland runs one to three points under the national rate in the early years and lands essentially dead even by year ten (34.6% versus 34.7%). If you wanted me to tell you the state is the problem, the data won’t cooperate.

Which is exactly why the number matters. Half of everything that opened five years ago is gone — here, and everywhere. (That’s the all-private-sector number; survival varies a lot by sector, and the table I’m citing doesn’t break it out, so don’t let me imply it does.) The graveyard is not sorted by state. I don’t think it’s sorted by idea either.

So what is it sorted by?

The Arrow Points at the Business, Not the Business You’re In

Strategy academics have spent forty years arguing about a single question: how much of a company’s profitability comes from what industry it’s in versus which specific company it is?

Richard Schmalensee ran it in 1985 and concluded industry was the thing that mattered. Six years later Richard Rumelt re-ran it on the same FTC data, corrected a weakness in Schmalensee’s method, and got the opposite answer — in his own words:

“the most important sources of economic rents are business-specific; industry membership is a much less important source.”

His decomposition put business-unit effects at 46.4% of the variance in profitability against industry effects of 8.3%. (His second sample, which gets quoted less often — including by me, until about an hour ago — says 44.2% and 4.0%. Same direction, weaker numbers, and you should have both.)

Anita McGahan and Michael Porter re-ran it six years later with different data and got the same shape: industry 19%, business-specific 32%.

Different methods, different decades, same direction. Which business you are matters more than which business you’re in.

Two honest problems with that, and I’d rather raise them than have you raise them.

1. First, both studies look at large public corporations, not eleven-truck shops in Essex.

2. Second — and this one genuinely cuts against me — McGahan and Porter found that industry effects are larger in services, retail, and lodging than in manufacturing.

Most of you are in services. Take the number as a direction, not a measurement.

But there’s a body of research aimed directly at businesses your size, and it’s better than I expected.

The Part That Actually Has a Number On It

David McKenzie and Christopher Woodruff surveyed 20,400 firms across seven countries and scored them on twenty-six mundane practices — marketing, stock-keeping, record-keeping, financial planning. Not vision. Not positioning. Whether you write things down.

The median firm in that sample had zero employees. Ninety-seven percent had nine or fewer. These are Dave-sized businesses, or smaller.

A one-standard-deviation improvement in that practices score was associated with 22–35% higher labor productivity — that’s the spread between their loosest and tightest specifications, and I’m handing you both ends because picking the flattering one is how people lie with regressions — plus roughly 15% higher sales growth over five and a half years and a small but real survival bump (2.2 percentage points).

Correlational, and the authors say so plainly — well-run firms and profitable firms may both be caused by something else, like the owner. Fine. So here’s the causal version.

Miriam Bruhn, Dean Karlan and Antoinette Schoar ran an actual randomized trial:

• 432 small and medium enterprises in Mexico, of which 150 were randomly offered a year of management consulting — and only 80 of those 150 took it.

• Five years later, measured across the whole offered group, employment and total wage bill were up about 50%. The practices that moved were marketing, financial accounting, and long-term planning.

And at national scale, the U.S. Census surveyed roughly 35,000 manufacturing plants twice and found that management practices account for more than 20% of the variation in productivity — about the same as R&D, and more than IT or workforce skill. A one-standard-deviation better management score was associated with a 3.1 percentage point lower chance of the plant dying, against a mean death rate of 11.8%. That’s a 26% reduction in the odds of going out of business.

Now the single most useful finding in any of this.

When Nicholas Bloom and colleagues ran a management RCT on Indian textile plants — productivity up 17% in the first year — they asked the obvious question afterward: if these practices work this well and cost this little, why wasn’t anybody already doing them?

The answer wasn’t ignorance. Of all the practices these plants had not adopted, only about a quarter went unadopted because nobody knew the practice existed. Roughly 60% went unadopted because the owner knew about it and had decided it wouldn’t work for him. (In the paper’s own accounting, across every practice measured: about 15% failed on lack of information, about 44% on incorrect information.)

Read that again, because it’s the whole article.

The information was already available.

What was scarce was the willingness to do it.

The Cheapest Thing in Your Business Is the Phone, and You Are Probably Losing on It

Let me get concrete, because “execute better” is useless advice unless it has a number attached.

In 2011, researchers audited 2,241 U.S. companies by submitting a web lead to each one and timing the response. 37% responded within an hour. Twenty-three percent never responded at all. Among those that did respond within thirty days, the average was 42 hours.

The same paper reports a second and separate dataset — 1.25 million leads across 42 companies — in which firms that made contact within an hour were nearly seven times as likely to reach an actual decision maker as firms that waited even one hour longer, and more than sixty times as likely as firms that waited a day. Two different studies. They almost always get quoted as one, and I’d rather split them than inherit somebody else’s sloppiness.

That study is fifteen years old. So here’s a 2026 one, published in Management Science, built on 11.6 million real conversations on a freelance marketplace — which is about as close as academic research gets to a customer asking a contractor for a quote. A one-hour delay in replying was associated with roughly a 46% reduction in the odds of getting hired. A full day, roughly 90%.

The researchers also found the boundary condition, and it’s the important part: the advantage disappears when the fast reply is obviously automated. Speed works because it signals a real human is paying attention. An autoresponder buys you nothing.

“Mike, my guys are on roofs. In August. In Baltimore.”

I know. That’s the point. So is everybody else with a truck. Nobody is going to out-idea anybody in this market. Somebody is going to out-answer everybody — and in the one controlled audit anybody has actually run, roughly one in four companies never answered at all. That audit was 2011. Whether it’s better or worse now, I don’t know, and neither does anyone selling you a statistic about it.

One more, because it sets the clock. The JPMorgan Chase Institute looked at cash flows for 597,000 small businesses and found the median holds 27 days of cash buffer. By industry: construction 20 days, repair and maintenance 18, restaurants 16. (That data is from 2015, so treat the levels as dated and the shape as durable.)

Your strategy has quarters to be wrong. Your cash has about twenty days.

The Fine Print, Because I’m Also a Lawyer

Here’s where I argue against myself, and this section is genuinely the reason I wanted to write this piece.

The classic textbook version of this argument — the one taught in strategy courses, including the chapter that sent me down this hole — rests on two showcase examples. Both of them have since been dismantled. Neither fact is in the textbook.

Example one: Herman Miller’s SQA unit. In 1995 the office-furniture giant built a lean operating unit called SQA — simple, quick, affordable — to serve small businesses. New supply chain, new IT system, everybody from sales to the loading dock drilled on speed and accuracy. The numbers are famous: order-to-delivery cut from the industry’s six-to-eight weeks down to under two, on-time and complete shipments running at 99.7% — at a company where a decade earlier nearly a third of all orders had arrived late or incomplete — and sales growing 25% a year. Textbook proof that execution beats idea.

What the textbook doesn’t mention is that on March 15, 2002, Herman Miller’s board approved the “phase-out of the SQA business model.” Its own filings that year: net sales down 34.3%, a net loss of $56.0 million, headcount down 29.1%, $81.6 million in special charges. Management’s own summary: “We eliminated the RED initiative and SQA brand, focusing our investments on the Herman Miller brand.”

The 2005 book describes SQA in the present tense. It had been folded into the parent in 1998 and formally killed in 2002.

Two more honest wrinkles. The most-cited source for those SQA numbers is a 2004 trade-press retrospective co-written by Herman Miller’s own president of North America and a consultant selling the framework — company figures, published after the unit was already dead, not independent measurement. And a 1997 Forbes piece put the industry standard at five weeks, not six-to-eight, and the growth at 30% against a company average of 16%. The story is real. The precision is softer than it’s usually presented.

Example two: Southwest Airlines. Michael Porter’s 1996 diagram of Southwest’s “activity system” is probably the single most-reproduced image in strategy teaching. No seat assignments. No baggage transfers. No connections with other airlines. Limited passenger service. Fifteen-minute gate turns. Rivals copied the low fares and failed, Porter argued, because they couldn’t copy the interlocking system.

As of this year, that diagram is a list of things Southwest used to do.

After Elliott Investment Management took a $1.9 billion stake in June 2024 and reached a settlement that October seating five of its designees on the board, effective November 1: checked-bag fees took effect May 28, 2025 ($35 first bag, $45 second). Assigned seating began January 27, 2026, ending fifty-four years of open boarding. The airline now has nine partner airlines.

And — this is the part that complicates everyone’s tidy story, mine included — it made money. Record FY2025 revenue of $28.1 billion. First-quarter 2026 operating margin up 8.1 points year over year, with about 60% of customers paying up from the base product versus 20% the year before.

Though not cleanly: in July 2026 the company cut its full-year adjusted EPS guidance to $3.25–$4.25, from a prior “at least $4.00”; second-quarter operating margin was still only 3.4% GAAP (6.7% adjusted, both up year over year); and there’s a documented trail of customer backlash and post-launch fixes.

So what do I actually take from two dead examples?

Not that execution doesn’t matter — the research above is stronger than either anecdote. What I take is a correction to how the idea is usually sold. Alignment gets taught as a monument: build the interlocking system, and it protects you. Herman Miller and Southwest both say it’s a practice. Herman Miller’s own filings note that the SQA operating disciplines “were incorporated into other areas of the company” even as the brand was killed — the way of working outlived the thing it was housed in. And Wiggins and Ruefli, studying advantage across decades of public companies, found that sustained advantage is increasingly not one advantage held a long time but “a sequence of advantages” strung together over time.

Stringing them together is an execution skill. It’s just not a permanent one.

Three more things that could be wrong here:

The management research I leaned on has real limits. McKenzie and Woodruff is correlational and set in developing countries. The World Management Survey — the most cited body of work in this field — doesn’t survey firms under 50 employees at all, which means the most famous evidence on this question structurally excludes most of my readers.

Nobody has cleanly separated “failed because badly run” from “failed because wrong market.” I looked. The failure data doesn’t decompose that way, and anybody who tells you 70% of businesses fail for one reason or the other is selling something.

And the line everybody quotes on this topic — Jeffrey Pfeffer’s “It is more important to manage your business right than to be in the right business” — I could not find in any accessible source. It’s repeated in textbooks and attributed to Pfeffer, and it may well sit on a page nobody’s digitized. But I’m not going to print a quote I can’t verify and pretend I checked. Rumelt’s peer-reviewed version says the same thing and I’ve read it.

What I’d Actually Do

Time your own call-back this week, and don’t warn anybody. Have your spouse, your brother-in-law, anyone whose number isn’t in your CRM, submit your web form at 2 p.m. on a Tuesday and call your main line at 4. Write down the minutes. That’s it — that’s the whole diagnostic, and it costs you nothing. In a controlled audit of 2,241 companies, 23% never responded at all. If your number is worse than an hour, you have found the highest-return project in your business and it isn’t a strategy project.

Write down the five things that have to go right every single time. This is the SQA lesson with the branding stripped off: they didn’t win because “simple, quick, affordable” was a clever positioning. They won for a few years because they picked a small number of operational promises and built the whole shop around hitting them. For Dave that’s probably: answer within an hour, show up in the window, quote in writing within 24 hours, bill within 48, follow up at 30 days. Five sentences. Then measure whether you hit them, which is the part everyone skips.

Count your buffer days before you count anything else. Total your average monthly outflows, divide your operating cash by the daily figure, and look at the number. If you’re at the construction median you have about twenty days. Every strategic decision you make should be sized against that number, not against the plan.

Pay for the thing your strategy requires, or admit the strategy isn’t real. If your plan depends on senior techs who can talk to a facilities manager without you in the truck, and a senior tech in your shop makes what a two-year tech makes, you don’t have that plan — you have a document. This is the single most common misalignment in the whole literature: the company announces one thing and the payroll does another. Your payroll is your actual strategy. It’s the only version your people can read.

Then, and only then, go back to the idea. Dave’s maintenance-contract plan is good. It’ll work — after the callbacks are fast, the promises are written down, and the buffer is real. Run it in that order. Reversed, you get what he got: a strategy that was right the whole time, sitting in a folder, while the phone rang.

Dave doesn’t need a better idea.

There are 5,548 companies in this state doing his kind of work, better than two thousand of them within an hour of his yard, and not one of them has anything he can’t get. The trade is a commodity. The equipment is a commodity. The plan he paid a hotel ballroom for is, if we’re honest, a commodity too.

What none of them can copy is what happens at his shop at 7:40 on a Tuesday morning.

The strategy was always the cheap part. Everybody keeps buying it anyway.

Leave a comment

Subscribe now

Sources

Business density and survival

• U.S. Census Bureau, County Business Patterns 2023 (establishments with employees) and Nonemployer Statistics 2023 (businesses without employees). Maryland and Baltimore–Columbia–Towson MSA figures; the plumbing/heating/AC counts are NAICS 23822, which bundles residential plumbers with mechanical contractors. The two datasets cover mutually exclusive universes, so the combined totals above are my own addition. CBP counts establishments (locations), not firms, so a company with three shops counts three times — meaning my competitor counts run a little generous. 2023 is the most recent year available for both.

• U.S. Bureau of Labor Statistics, Business Employment Dynamics, Table 7, “Survival of private sector establishments by opening year,” data through March 2025 — Maryland · U.S. total private · program page. Establishments, not firms; excludes Maryland’s ~599,000 businesses with no employees; “survival” does not distinguish failure from sale, relocation, or retirement.

• JPMorgan Chase Institute, Cash is King: Flows, Balances, and Buffer Days (September 2016) — 597,000 small businesses, 470M+ transactions, 2015 data. Chase-customer accounts only; excludes credit lines and outside reserves.

Industry vs. firm effects

• Schmalensee, Richard. 1985. “Do Markets Differ Much?” American Economic Review 75(3): 341–351 — the study Rumelt corrected.

• Rumelt, Richard P. 1991. “How Much Does Industry Matter?” Strategic Management Journal 12(3): 167–185. DOI: 10.1002/smj.4250120302. Sample A: industry 8.3%, business-unit 46.4%. Sample B: 4.0% and 44.2%.

• McGahan, Anita M., and Michael E. Porter. 1997. “How Much Does Industry Matter, Really?” Strategic Management Journal 18(S1): 15–30. DOI: 10.1002/(SICI)1097-0266(199707)18:1+<15::AID-SMJ916>3.0.CO;2-1 — Wiley landing page. Note their finding that industry effects are larger in services, retail, and lodging.

• Wiggins, Robert R., and Timothy W. Ruefli. 2005. “Schumpeter’s Ghost: Is Hypercompetition Making the Best of Times Shorter?” Strategic Management Journal 26(10): 887–911. DOI: 10.1002/smj.492

Management practices and firm performance

• McKenzie, David, and Christopher Woodruff. 2017. “Business Practices in Small Firms in Developing Countries.” Management Science 63(9): 2967–2981. DOI: 10.1287/mnsc.2016.2492 · free PDF. 20,400 firms; median firm has zero employees. Correlational. The 35% productivity figure is their least-controlled specification and 22–23% their most-controlled; both appear in the article above.

• Bruhn, Miriam, Dean Karlan, and Antoinette Schoar. 2018. “The Impact of Consulting Services on Small and Medium Enterprises: Evidence from a Randomized Trial in Mexico.” Journal of Political Economy 126(2): 635–687. DOI: 10.1086/696154. 432 firms; 150 randomly offered treatment, 80 took it up. The ~50% employment effect is an intention-to-treat estimate across the full offered group. Causal.

• Bloom, Nicholas, Erik Brynjolfsson, Lucia Foster, Ron Jarmin, Megha Patnaik, Itay Saporta-Eksten, and John Van Reenen. 2019. “What Drives Differences in Management Practices?” American Economic Review 109(5): 1648–1683. DOI: 10.1257/aer.20170491 · free PDF. ~35,000 plants per wave, U.S. Census MOPS. Correlational.

• Bloom, Nicholas, Benn Eifert, Aprajit Mahajan, David McKenzie, and John Roberts. 2013. “Does Management Matter? Evidence from India.” The Quarterly Journal of Economics 128(1): 1–51. DOI: 10.1093/qje/qjs044 · NBER working paper 16658. RCT, causal — but only 28 plants across 17 firms, averaging 270 employees, so take it as a mechanism, not a magnitude. The headline figure also moved across drafts: the 2010 preliminary version says about 10%, the NBER working paper 11%, the published QJE version 17% in the first year. I’ve used the published version and told you the others exist.

• Scur, Daniela, Raffaella Sadun, John Van Reenen, Renata Lemos, and Nicholas Bloom. 2021. “The World Management Survey at 18: Lessons and the Way Forward.” IZA Discussion Paper 14146 — source for the survey’s 50-employee minimum firm size.

Response time

• Oldroyd, James B., Kristina McElheran, and David Elkington. 2011. “The Short Life of Online Sales Leads.” Harvard Business Review 89(3): 28. HBR · full text PDF. 2,241 companies audited. Not peer-reviewed; observational.

• Hart, Einav, Eric M. VanEpps, Ovul Sezer, and On Amir. 2026. “Speed Is a Signal: When Faster Replies Increase Hiring Likelihood.” Management Science. DOI: 10.1287/mnsc.2024.06185. ~11.6 million marketplace conversations plus experiments with 3,600+ participants. The headline delay magnitudes are from the observational data; the experiments establish the mechanism.

Herman Miller / SQA

• Herman Miller, Inc., FY2002 Form 10-K (SEC) — board approval of the SQA phase-out on March 15, 2002; FY2002 net sales, net loss, headcount, and special charges; the note that SQA’s operating disciplines were incorporated elsewhere in the company. See also FY2003 and FY2004 10-Ks.

• Walker, Brian, and David Bovet. “Value-Chain Report: Herman Miller Reinvents for Success.” IndustryWeek, December 21, 2004 — most-cited source of the two-week cycle, 99.7% on-time, and 25% growth figures, and of the “almost one-third” late-or-incomplete baseline. Co-written by Herman Miller’s then-president of North America.

• “A Touch of Schizophrenia.” Forbes, July 7, 1997 — the competing figures (five-week industry standard; 30% SQA growth vs. 16% company-wide).

• Harvard Business Essentials, Strategy: Create and Implement the Best Strategy for Your Business (Harvard Business School Press, 2005), ch. 5, “From Strategy to Implementation: Seeking Alignment” — the chapter that prompted this piece.

Southwest Airlines

• Porter, Michael E. “What Is Strategy?” Harvard Business Review 74(6), November–December 1996: 61–78 — the original activity-system diagram.

• Southwest Airlines, July 25, 2024 announcement of assigned and premium seating (SEC) · March 11, 2025 announcement of bag fees and Basic fares (SEC — this filing gives the May 28, 2025 effective date but not the dollar amounts; the $35/$45 figures were announced separately and are reported here) · October 24, 2024 agreement with Elliott Investment Management (SEC).

• Southwest Airlines quarterly results: Q4/FY2025 · Q1 2026 · Q2 2026 (SEC).

• CNBC, “Southwest ends open seating after 54 years. Here’s what the last flight was like”, January 28, 2026 · “Elliott takes $1.9 billion stake”, June 10, 2024.

• Customer Experience Dive, “Southwest continues ‘refining’ assigned seating”, March 3, 2026.

A note on one thing I could not source: the sentence “It is more important to manage your business right than to be in the right business,” widely attributed to Stanford’s Jeffrey Pfeffer and quoted in the Harvard chapter above, returns no exact match in any source I could reach. Pfeffer makes the argument at length in Competitive Advantage Through People (Harvard Business School Press, 1994) and The Human Equation (1998), and the quote may sit on a page that isn’t digitized — but I couldn’t verify it, so I haven’t used it as a quotation.

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice, and I’m not your lawyer — I’m a guy from Baltimore with a spreadsheet and a law degree, in that order. The Census business counts are 2023 vintage and the BLS survival data runs through March 2025; both will move. The cash-buffer figures are from 2015 and should be treated as shape, not level. Southwest’s results are current through its second-quarter 2026 release and will look different by the time you read this. Verify anything you’re about to make a payroll decision over.

Football coaches don’t panic when the other team shows them something new — they adjust. Small business owners can do the same thing. In this Chalk Talk, I break down how to read a competitor’s strategy, figure out what they’re actually good at, and counter it without blowing up your own game plan.

From Crabcakes & Commerce — practical business breakdowns for Maryland small business owners.

In this first Chalk Talk, I break down how to read a competitor's strategy the way you'd read an offense: what they're actually good at, where they're overcommitted, and what they've given up to be good at that thing. Every strategy has a tradeoff baked in. Find the tradeoff, and you've found the counter.

Coffee’s on. Let’s go through the books.

Quiet week on the surface — no tariff walls went up, no grant deadlines detonated. But the quiet weeks are when the paperwork deadlines sneak up on you, and there’s one opening tomorrow morning that most owners I talk to don’t know exists yet.

On the ledger this week: the window to opt your business out of remitting a year of paid-leave contributions to the state opens tomorrow, September 1; a Maryland court threw out the digital ad tax and put roughly half a billion dollars of collected revenue in play; the Sail250 receipts came in at $114.5 million and I have thoughts about who’s counting; the Richmond Fed says Mid-Atlantic factories are treading water; the state stood up a $3 million fund to grow its own nurses; and small-business optimism ticked up in July while uncertainty did too, which sounds contradictory until you run a business.

1. FAMLI’s first real deadline opens tomorrow. Most owners don’t know it’s there.

Maryland’s paid family and medical leave program — FAMLI — got a round of local coverage this week, mostly framed as a new deduction coming out of Marylanders’ paychecks starting in 2027. That’s true, but it buries the part that matters to you this week.

First, the shape of the thing, straight from the state. Every employer with at least one Maryland employee is required to register — no exceptions under state law. Contributions and quarterly wage reports start flowing in 2027, with the first quarterly filings due in April 2027. Benefits start January 2028: up to 12 weeks of paid, job-protected leave at up to $1,000 a week. If you have fewer than 15 employees, you’re a “small employer” — you carry all the same registration and reporting duties, but you only remit 50% of the total contribution rate, and you can withhold that amount from employee pay.

Now the deadline. If you’d rather run FAMLI through a private insurance plan than the state fund — and your broker will absolutely be calling you about this — there’s a specific mechanism for the 2027 “seeding period.” Employers who intend to apply for a private plan must file a Declaration of Intent between September 1 and November 15, 2026. That window opens tomorrow. Be clear-eyed about what it buys you: an accepted Declaration doesn’t make the contributions disappear. You still collect them starting January 2027 — you just hold them in an escrow account instead of sending them to the state, and where that money ultimately goes depends on whether your private plan is approved and stays in place. Leave the private plan early in the first years and you owe the state back contributions with interest. It’s a cash-position play, not a discount.

The process runs through the FAMLI portal, requires a consultation with a licensed insurance agent and their attestation, and only an Authorized Officer of the business can submit it — your payroll company can’t register you or file the Declaration, though once you sign a power of attorney it can handle quarterly reports and payments. And a Declaration of Intent is not the plan approval itself; that’s a separate application due October 1, 2027.

One honest caveat: for most very small businesses, the state plan will likely be the simpler and cheaper path, and doing nothing about private plans is a perfectly defensible choice. The Declaration of Intent matters if you’re the kind of shop — steadier payroll, existing disability coverage, a benefits broker you trust — where a private plan might pencil out. What isn’t defensible is not registering at all, or finding out about contribution withholding from your employees instead of telling them first.

What I’d do: Register at paidleave.maryland.gov this fall regardless of which path you take — it’s required either way. If you have any interest in a private plan, get the broker conversation on the calendar in September, not the week of November 15 — which falls on a Sunday, incidentally, and I wouldn’t test how the portal feels about that; the consultation and attestation take time you won’t have at the deadline. And if you run payroll in-house, put “FAMLI line item” on your January 2027 checklist now, because that deduction will generate exactly one predictable employee question and you want the answer written down before it’s asked.

2. The digital ad tax is dead — pending appeal — and half a billion dollars is in play

On August 14, the Maryland Tax Court struck down the state’s first-in-the-nation digital advertising tax and ordered refunds to Apple, Google, and Peacock TV, plus interest. The court’s reasoning cut deep: the tax violates the federal Internet Tax Freedom Act, the dormant Commerce Clause, and due process — it taxed digital ads while leaving equivalent non-digital ads alone, and pegged rates to companies’ global revenue rather than their Maryland business.

You don’t pay this tax. It only ever applied to companies with $100 million-plus in global revenue. So why is it in your newsletter? Because of where the money was going and what happens now that it might go back. The tax was earmarked for education funding under the Blueprint, and the state has collected about $535.5 million under it through July, per the Comptroller’s office. If the ruling stands, Annapolis has to fill that hole — and recent history says that when Annapolis fills holes, the shovel lands somewhere in the business community. The 3% tech services tax that arrived this July is what hole-filling looks like.

Comptroller Brooke Lierman said she respectfully but strongly disagrees and will work with the Attorney General to defend the law, and the state has 30 days from the ruling to appeal — a window that runs out in mid-September. An appeal would put the refunds on hold, and this thing has already been up and down the federal courts once. Nobody should treat August 14 as the last word.

What I’d do: Nothing operational — this one’s a watching brief. But when the 2027 legislative session starts gaming out revenue replacements, the proposals will surface in November and December committee chatter, not in January. If your trade association has a legislative committee, this is the fall to actually read its emails.

3. Sail250’s receipts: $114.5 million, 608,000 visits, and a grain of salt

The economic impact report for June’s Sail250 Maryland & Airshow Baltimore landed earlier this month, and the headline numbers are big: an estimated $114.5 million in economic impact — $83.4 million direct, $31.1 million indirect — from 608,148 visits across seven venues over the week of June 24–30. Food and beverage took the biggest slice at $44.5 million total impact; hotels took $27.2 million. Against $1.94 million in event expenditures, the report claims a 49-to-1 return on every public and private dollar invested.

Now the grain of salt, because I’m a numbers guy and this is what we’re for. The report was prepared by the Living Classrooms Foundation — the same organization that produced the event — with support from the Baltimore Development Corporation and others. Organizer-commissioned impact studies are a genre, and the genre runs generous: “indirect impact” multipliers are squishy, and a “visit” is not a wallet. A 49:1 ROI would make Sail250 one of the best investments in the history of money. And here’s a data point that argues for holding the figures loosely: the Banner reported the day before the event that downtown hotels weren’t half full for the week, with organizers expecting 200,000 to 250,000 attendees. Two months later the report says 608,148 visits and $27.2 million in hotel impact. Somewhere between the empty rooms in the preview and the full ones in the report is the truth, and I can’t tell you exactly where.

Here’s what I don’t hold loosely: the harbor was genuinely packed, 33 ships and an airshow drew a real crowd whatever the multiplier, and Baltimore has now shown it can land and run a large-scale event without incident. More will follow, and they’ll be announced with lead time.

What I’d do: If you’re in hospitality, retail, or services anywhere near the water, treat this report as a forecast, not a history. The next big event will get announced months out. Have a one-page playbook ready — extended hours, event-priced menu or package, temp staffing contact, sidewalk signage — so you’re executing in week one instead of improvising. The businesses that captured Sail250 money were the ones ready before the ships showed up.

4. Mid-Atlantic factories are treading water

The Richmond Fed’s August survey of Fifth District manufacturers — that’s us, plus DC, Virginia, most of West Virginia, and the Carolinas — came out Tuesday, and the short version is: flat. The composite index eased to 4 in August from 5 in July, which in this survey’s language means activity is technically growing but barely. Underneath, the same mixed picture that’s been running all year: shipments holding up, expectations softer, and price pressures that haven’t fully let go.

A one-point move in a diffusion index is noise, not news, and I won’t pretend otherwise. The reason it makes the ledger is the trend it confirms: five months of readings stuck in a narrow band around zero — from a dip to −2 in April to a high of 6 in July. Mid-Atlantic manufacturing isn’t contracting, but it has no momentum — and for the supply chains that run through it, that means vendors who are cautious, slow to add capacity, and quicker than usual to pass through cost increases while volume is soft.

What I’d do: If you make things or buy from people who do, this is the environment for locking in terms rather than assuming next quarter looks like last quarter. Flat markets are when suppliers get flexible on price to keep volume — and when the weaker ones quietly get shaky. If a supplier’s lead times are stretching while everyone else’s are steady, that’s not a demand signal, that’s a them signal. Have a second source scoped for anything you can’t operate without.

5. The state is growing its own nurses, $3 million at a time

Maryland and the nonprofit Social Finance launched the Maryland Pay It Forward Fund this month — a $3 million pot, seeded with $2 million from the state’s Talent Innovation Fund plus $1 million in philanthropic money from Blue Meridian Partners, that provides zero-interest loans to students in associate-degree nursing programs, starting at the Community College of Baltimore County with more colleges intended to follow. The mechanics are unusually practical: up to $5,000 to cover the tuition gap after financial aid, plus up to $500 a month — capped at $12,000 over two years — for the stuff that actually derails degrees: transportation, child care, rent. About 80 nursing students in the first year, more if the pilot holds.

Eighty nurses is not a workforce policy; it’s a pilot. But the design is worth noticing because of why it exists: the state cites thousands of open nursing positions, and this program bets that a meaningful share of the shortage isn’t about nursing-school seats — it’s about students dropping out over a $400 car repair. If you run a business anywhere near healthcare — staffing, billing, transport, food service into facilities, child care near a hospital — the state just told you where it thinks the bottleneck is.

What I’d do: If you employ anyone working toward a credential — CNA to RN, apprentice to journeyman — steal the design. A small employer-funded completion stipend tied to staying enrolled costs a fraction of turnover and recruiting, and the state just published the template for what the support should cover. And if you’re in child care specifically: a cohort of nursing students getting $500 a month partly earmarked for child care is a demand signal being wired directly into your market.

6. Optimism up, certainty down — July’s small-business mood, a month late

The NFIB’s July numbers are the freshest small-business sentiment we have, and they’re worth a minute even though it’s Labor Day weekend and the data is a month old — I’d rather tell you that plainly than pass July off as breaking news. A housekeeping note first: this morning’s research brief handed me a set of figures that turned out to be from July 2025. The internet is full of last year wearing this year’s clothes. Here’s the current release.

The national Small Business Optimism Index rose 2.4 points in July to 99.8, just above its 52-year average of 98. The Uncertainty Index rose two points to 91. And the number I’d circle: 27% of owners named labor quality their single most important problem — up eight points in one month, and the top problem on the list. There’s no fresh Maryland-specific cut of this data yet; when NFIB’s state office publishes one, I’ll bring it here.

Optimism rising while uncertainty rises too isn’t a contradiction — it’s what it feels like when your own shop is doing fine and everything around it keeps changing. Most owners I know are living in exactly that gap.

What I’d do: The labor-quality number is the actionable one, and it rhymes with item 5: if more than a quarter of your competitors say they can’t find good people, the ones who develop people win by default. That’s train-and-retain, not post-and-pray. As for the optimism index itself — it’s a fine mood ring, but your own bookings ledger is a better forecast than anyone’s survey, including this one.

The week ahead

• Tuesday, September 1 — FAMLI Declaration of Intent window opens for employers considering a private plan. It runs through November 15.

• Wednesday, September 2 — Maryland’s July substate employment data: county, city, and MSA. This is when Baltimore and Anne Arundel become real numbers instead of vibes.

• Thursday, September 3 — Monthly Labor Review, the state’s narrative writeup of the same data.

• Monday, September 7 — Labor Day.

• Tuesday, September 8 — Canada’s retaliatory tariffs on steel, dairy, appliances, agricultural equipment, pulp and paper, and electronics take effect, as announced August 22.

• Mid-September — the Comptroller’s 30-day window to appeal the August 14 digital ad tax ruling runs out. An appeal freezes the refunds.

• Thursday, September 17, 11:59 p.m. — DHCD’s $14 million Small Business Direct Loan round closes. Seventeen days from today.

• Friday, September 18 — Maryland’s August statewide employment report.

See you next Monday.

Sources

• For employers — Maryland FAMLI — Maryland Department of Labor (registration requirement, small-employer rules, quarterly reporting, TPA limits)

• Private plans — Maryland FAMLI — Maryland Department of Labor (Declaration of Intent window, September 1–November 15, 2026)

• Paid family and medical leave is coming to Maryland — Maryland FAMLI (January 2028 benefits, 12 weeks, $1,000/week)

• FAMLI tax to begin deduction from Marylanders’ paychecks in 2027 — The Baltimore Sun, August 27, 2026

• Maryland tax court voids digital ad tax, orders refunds to Apple, Google and Peacock TV — Associated Press, August 14, 2026

• Digital ad sales tax struck down by court — Maryland Matters, August 14, 2026

• Maryland Digital Ad Tax Rulings Shift Debate to What Comes Next — Bloomberg Law ($535.5 million collected through July 2026)

• Maryland Tax Court Invalidates Landmark Digital Advertising Tax — Forvis Mazars (Comptroller response, appeal timeline)

• SAIL250 Maryland & Airshow Baltimore drew over 600K people — Baltimore Fishbowl (impact report figures, Living Classrooms authorship)

• ‘Operation Sail’ brought millions to Baltimore. Sail250 hopes to repeat. — The Baltimore Banner

• Manufacturing Survey — Federal Reserve Bank of Richmond (August 2026 composite index, released August 25)

• State of Maryland and Social Finance Announce Maryland Pay It Forward Fund — Office of Governor Wes Moore

• State of Maryland and Social Finance Announce Maryland Pay It Forward Fund to Strengthen Maryland’s Nursing Workforce — Maryland Department of Labor (fund structure and amounts)

• Maryland, Social Finance launch $3 million Pay It Forward fund for nursing degrees — Fox45 Baltimore (first-year cohort size)

• NEW NFIB SURVEY: Small Business Optimism Continues to Rise — NFIB, August 11, 2026 (July 2026 national index, uncertainty, labor quality)

• Small Business Direct Loans — Maryland DHCD (September 17 deadline)

• Maryland Employment and Unemployment Press Release Schedule — Maryland Department of Labor (September release dates)

• As Canada readies retaliatory tariffs, Mark Carney says his nation is ‘at war’ with U.S. — NPR, August 22, 2026 (September 8 retaliation)

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice — I’m a guy from Baltimore with a spreadsheet and a bar card. Program details, dates, and rates can change; verify with the agency before you act.

"Scaling" is the word every business owner uses and nobody defines. It lives on LinkedIn somewhere between a sunset photo and a quote from a guy who has never once made payroll. This week I try to pin it down: what scaling actually means, how it's different from just getting bigger and more tired, and why the guy running six lawn crews at a healthy margin is doing great — even though the internet keeps telling him he's behind.

On this episode of C&C, were talking all things fun with growing your business and how that gameplan changes when you are chasing “growth” or “scaling”. I do my best to make it fun. Results may vary.

Share

Subscribe now

Let’s invent someone. Call her Angie.

She runs a commercial HVAC and mechanical contracting company out of a shop off Washington Boulevard in Halethorpe. Twenty-two employees. Her father started it in 1987 with a van and a garage in Arbutus, and the original hand-lettered sign is still on her office wall because her mother won’t let her throw it out.

About a third of her work is federal buildings. The rest is schools, medical offices, and a few property managers who call her at two in the morning in January.

Angie isn’t real. She’s assembled from enough real people that if you’re reading this, you’ve probably know an Angie.

Two years ago she decided to buy the building next door — the one she’s been renting pieces of since 2019. She did it the right way, too. Cleaned up three years of financials. Paid down the cards. Got her debt-service coverage to where a lender actually smiles. And then, because every single person she asked told her the same thing, she waited.

“Rates are coming down. Wait for the Fed to cut.”

She got the refreshed quote this month. The building didn’t change. Her credit didn’t change. Her business got stronger. And the payment came back higher than the number she’d been carrying around since spring — a couple hundred dollars a month higher, call it $2,400 a year, on a deal she has been patient about for two years.

Nothing she did caused that. And nothing she could have done would have prevented it.

The floor moved under her.

That floor has a name. It’s the bond market, and for three weeks it has been the biggest story in American business — one almost nobody running a small business here is reading, because it arrives in somebody else’s vocabulary. Yields. Term premium. The long end.

It is not somebody else’s problem. On Tuesday, August 18, the national debt crossed $40 trillion for the first time — $40,047,425,768,420.22, if you want the change. The government is currently paying $3 billion dollars a day on debt and the only higher payment on the books is Social Security. The same day, the 30-year Treasury touched 5.34%, a 19-year high.

That is not an abstraction. It’s the raw material Angie’s next loan is made out of. Yours too.

Here’s What Most Owners Believe

Angie’s advisors weren’t being stupid. They were repeating the thing almost everybody believes: the Fed raised rates, so my loan got expensive, and when the Fed cuts, it’ll get cheap again. So I’m waiting.

Half of that is correct. The wrong half is the expensive half.

The Fed sets one rate — the overnight rate banks charge each other. Right now the target is 3.50%–3.75%, and prime — the number your line of credit and most SBA 7(a) loans are pegged to — sits three points on top of it, at 6.75%. That part, the Fed really does control.

But the Fed does not set the rate on a 10-year loan, or a 30-year mortgage, or the SBA note on Angie’s building. Those prices are set by millions of strangers deciding what they’ll pay to lend money to the United States government. That’s the bond market. And it can raise your cost of money while the Fed sits perfectly still — which is exactly what it has been doing all month.

Waiting on the Fed is not a plan. It’s a hope, and it’s pointed at the wrong building.

What a Bond Actually Is, I Promise This Takes Four Minutes

Somewhere in someone’s house, there is a savings bond a great-aunt gave someone at a graduation. It says $50 on the front. She paid $25 for it.

That’s the whole thing, right there.

A bond is a loan, written down, that can be resold. She lent the government $25. The government wrote her an IOU that said we’ll pay you back $50 on a certain date. The gap between the $25 and the $50 is the interest. That’s it. That’s a bond.

Now scale it up. The federal government spends more than it collects in taxes. To cover the gap, it sells IOUs — Treasury bills, notes, and bonds — to anyone who’ll buy them: pension funds, insurance companies, Japan, your own bank, hedge funds in the Cayman Islands. Some IOUs pay back in two years. Some in thirty.

Here’s the part that trips everybody up, so stay with me for one paragraph.

When bonds get cheaper, interest rates go up.

Those are the same event described from two sides.

Say you lend a neighbor $1,000 and he hands you a paper promising $50 a year for ten years, then your $1,000 back. Fine. Now say you need cash and want to sell that paper to somebody else. But in the meantime, brand-new papers just like it are paying $70 a year. Nobody is going to give you a full $1,000 for a paper that only pays $50. They’ll offer you $850.

Same paper. Same $50 a year. Lower price. And because the buyer now pays $850 to get $50 a year, their return — the yield — is higher than yours was.

Price down, yield up. Always. It’s a seesaw with a plank you can’t bend.

So when you read “the 30-year Treasury hit 5.34%,” what actually happened is that the price of government IOUs fell, because more people wanted to sell them than buy them. The yield is just the scoreboard.

Two things make people demand a higher yield.

1. They think inflation will eat the money. If you’re getting paid back in dollars that buy less, you want more dollars.

2. They think the deal has gotten worse. Maybe there’s too much paper for sale. Maybe thirty years just feels like a long time to be locked in. Economists call that second one the term premium — the extra you charge for tying your money up for a long stretch when you can’t see the far end of it.

And one last piece, the one that connects all of this to you.

The rate on U.S. government debt is the floor price of money in America. It’s what a lender earns for taking essentially no risk at all. Every other loan in the country is priced as “that number, plus however risky you are.” Your business is riskier than the U.S. Treasury. So your rate is that number, plus a spread.

Raise the floor, and everything standing on it comes up with it. Including you.

The Numbers, Because I’m a Numbers Guy

Here’s where the floor is as I write this.

That 30-year at 5.34% is the highest since 2007. The 10-year — the one that matters most for real estate — hit 4.75%, a 19-month high. Prime is 6.75%. And interest on the federal debt now runs about $1.0 trillion a year, roughly 3.3% of the entire economy — more than we spend on defense.

Now, I can hear you: “Mike, that’s just inflation. Prices spiked, rates followed. Old news.”

That’s the thing. It isn’t inflation. Consumer prices in July ran 3.4% headline, 2.5% core — still above the Fed’s 2% target, but falling, not spiking. Inflation got better this summer and long-term rates got worse anyway.

That’s what makes this different from 2022. This is not a story about the cost of eggs. It’s a story about the supply of paper — a government issuing an enormous amount of debt into a market where the old reliable buyers are stepping back. Foreign holdings of Treasuries fell to $9.30 trillion in June, down $72 billion in a month, with China down near its lowest level since 2008.

Two more tells, and then I’ll get to how this is related to your money.

First: on August 19, Treasury Secretary Scott Bessent announced Treasury would at least double the size of its long-dated buyback operations — from $2 billion to at least $4 billion per operation, starting September 9. That is the government going into the market to buy back its own IOUs to prop up the price. Yields dropped nine basis points and then bounced right back the next day. When the issuer has to become the buyer, that tells you something about who else isn’t showing up.

Second: the Fed minutes released the same day showed officials discussing whether they’d need to raise rates if inflation doesn’t keep cooling.

Read those two sentences together. If your business plan has a line in it that says “rates come down next year,” that line is currently unsupported.

The Four Doors It Comes Through

Angie is about to take all four of these at once. Most owners take two or three and never connect them.

Door One: Your Own Paper.

SBA 7(a) loans float on prime, and SBA caps what a lender can add on top. At today’s 6.75% prime, the ceilings run 9.75% on loans over $350,000, 11.25% from $250,001 to $350,000, 12.75% from $50,001 to $250,000, and 13.25% at $50,000 or less. Note what that means: the smaller your loan, the more you pay. A $40,000 working-capital loan can legally cost you four full points more than a $400,000 one.

Door Two: Your Bank.

Banks bought a lot of bonds when rates were low. Rates went up, those bonds lost value, and FDIC-insured institutions were sitting on $325 billion in unrealized securities losses as of the first quarter — up $19 billion in three months. That’s paper loss, not a bank run. But a bank staring at that number gets choosier about who it lends to. That’s you.

Door Three: Your Landlord.

About $875 billion in commercial mortgages mature in 2026, most written when money was cheap and refinancing now when it isn’t. Commercial mortgage delinquencies hit 7.86% in July, up 51 basis points in a single month. If your landlord’s note reprices, you’ll hear about it at renewal.

Door Your: Your Customer.

The 30-year mortgage is 6.67%. Every household paying more for a house, a car, and a credit card balance is a household with less to spend in your shop. High rates don’t just raise your costs. They shrink your top line.

The Maryland Part, Which Is Worse

Everything above is true in Ohio too. Here’s what’s ours.

Maryland is the most federally-dependent state economy in the country — federal activity accounts for roughly 30% of state GDP, about $150 billion a year in wages, contracts, grants, and retirement checks. Federal work is about 10% of all wages earned in this state, though a smaller share of headcount — six or seven percent, depending on how you count.

Then came the retrenchment. Between January 2025 and January 2026, Maryland lost 31,100 federal jobs and 49,300 total jobs — the largest decline of any state. Estimated corporate income tax payments from federal contractors fell about 60% year over year. A University of Maryland survey of state businesses found 45% reporting harm from federal workforce cuts and 36% from procurement changes, including delayed payments.

The state’s own borrowing tells the story. Moody’s stripped Maryland of its Aaa rating in May 2025, the first time since 1973. A $1.4 billion budget gap reopened for the coming year — about five times what had been projected in April. And this spring the state stopped buying Moody’s ratings altogether, hiring Kroll instead before its $800 million June bond sale.

So stack it up. Nationally, the price of credit is rising for everyone. Locally, the biggest customer in the state is spending less. Most of the country is getting one of those. We’re getting both.

Which is Angie’s actual problem. Her loan got more expensive and a third of her customer base got less certain, in the same twelve months, and those two facts arrived from completely different directions.

The Fine Print, Because I’m Also a Lawyer

Now let me argue against myself, because this piece is worth nothing if I don’t.

Nobody knows if this becomes a crisis. Jamie Dimon has been saying for a year that “there will be some kind of bond crisis” — though the sentence he actually said ends “and then we’ll have to deal with it. It will be okay,” which is calmer than the headlines. And history is on the skeptics’ side: when S&P downgraded the United States in 2011, everyone braced for a rate spike and Treasury yields fell instead. The base case here is elevated-but-functioning markets. Not collapse.

The scariest number in my notes is an estimate. Fed staff published a note in June finding that the hedge-fund “basis trade” — a leveraged bet on the gap between Treasury bonds and Treasury futures — had grown to about $830 billion, roughly double its 2020 peak. Governor Lisa Cook has flagged it as a real vulnerability. If it unwinds fast, it makes everything above worse in a hurry. But it rests on modeling assumptions, and reasonable people put the number in different places. I’m telling you it exists, not that I can time it.

And the survey data doesn’t back my alarm. The NFIB’s optimism index rose to 99.8 in July, above its 52-year average for the first time in nearly a year. Only 2% of owners named financing and interest rates as their single biggest problem.

I’d read that second number the other way, though. In June, just 22% of owners said they borrow regularly — against a historical average of 34%. Financing isn’t the top problem because a third of the people who’d normally be borrowing have simply stopped. That’s not comfort. That’s a waiting room.

Maryland, for its part, added 17,800 jobs in the first half of 2026 and grew faster than the country. Unemployment is 4.3%. Battered, stabilizing — not falling apart.

What I’d Actually Do

Make one page listing everything you owe that floats. Every loan, line, card, and lease tied to prime or SOFR, with the current rate and the balance. Most owners cannot produce this in under an hour, which is the problem. Prime moves a quarter point and you should be able to say what it costs you by dinner.

Then model that page at today’s rate plus one point. Not because I’m predicting a hike — because the Fed just spent a meeting discussing one, and a plan that only works if rates fall isn’t a plan. If plus-one-point breaks you, you’ve found this quarter’s real project.

Fix what deserves fixing, and know which rate you’re fixing against. This is the part I most want Angie to hear. Waiting two years cost her about $2,400 a year. Picking the wrong loan product would cost her ten times that. An SBA 504 loan fixes off the 10-year Treasury for the life of the note — around 6.3% for the 20-year at August pricing — while a large 7(a) floats at 9.75% and moves every time prime does. On an $800,000 note over twenty years, that spread runs roughly $1,750 a month. Twenty-one thousand a year, sitting inside a decision she gets to make. 504 for fixed assets, 7(a) for working capital. And shop three SBA Preferred Lenders, because those SBA numbers are ceilings, not prices.

Treat federal revenue like an over-weighted position. If federal work — direct or as a sub — is more than about a quarter of your revenue, you are running a concentrated portfolio in the one state where that concentration just got tested. You don’t have to walk away from it. You have to go find commercial, state, and local work while you still have the cash to prospect.

Pick up the phone about your lease before it picks up the phone about you. If your landlord has debt maturing, or you do, start that conversation early. Extensions exist. Panic refinancing at the top of a rate cycle is how people lock in the worst terms available.

Then watch three numbers and nothing else: the 10-year Treasury, prime, and the share of NFIB owners saying they can’t get credit. A sustained 10-year above 5% means more pain in real estate. A move back toward 4% is your window to lock something in. Everything else is noise.

I don’t know whether Angie should buy that building. I haven’t seen her books.

What I know is that she should stop deciding by waiting. She’s holding out for a cut that the people who actually price long money aren’t promising her — and every month she holds, they get a vote and she doesn’t.

That great-aunt paid $25 and the government promised her $50. That gap is all a yield is: the price the world charges Uncle Sam for a promise. For most of my life it was low enough that nobody outside a trading floor had a reason to learn the word. It was a thing that sat in a drawer.

It isn’t in a drawer anymore. It’s in Angie’s loan documents, in her landlord’s refinance, in what her customers can afford this quarter.

The bond in that envelope and the note on her desk are the same piece of paper, read from opposite ends.

That great-aunt could afford not to know that.

Angie can’t. Neither can you.

Leave a comment

Sources

Bond market and rates

• U.S. Treasury, Debt to the Penny — $40.047 trillion, August 18, 2026.

• U.S. Treasury, Daily Treasury Par Yield Curve Rates, August 2026.

• CNBC, “30-year Treasury yield tops 5.33%, a new 19-year high”; Bloomberg, “US bond selloff drives 30-year yields to the highest since 2007”.

• Seeking Alpha, “US10Y hits 4.75%, highest level in 19 months”.

• Federal Reserve, H.15 Selected Interest Rates — bank prime loan rate 6.75%.

• Federal Reserve, FOMC statement, July 29, 2026; CNBC, July 2026 minutes.

• U.S. Treasury, buyback operation announcement, August 19, 2026.

• U.S. Treasury, TIC major foreign holders of Treasury securities, June 2026.

• Bureau of Labor Statistics, Consumer Price Index, July 2026.

• Congressional Budget Office, The Budget and Economic Outlook: 2026 to 2036.

• Federal Reserve, Phillip J. Monin, “Decomposing Hedge Funds’ U.S. Treasury Exposures”, FEDS Notes, June 22, 2026.

• Governor Lisa D. Cook, “A Policymaker’s View of Financial Stability,” November 20, 2025.

• CNBC, Jamie Dimon on a bond crisis, April 28, 2026.

Small business credit

• U.S. Small Business Administration, 7(a) terms, conditions, and eligibility — maximum allowable rates.

• SBA 504 debenture rates, August 2026 pricing.

• NFIB, Small Business Optimism Index, July 2026 and June 2026.

• FDIC, Quarterly Banking Profile, First Quarter 2026.

• Mortgage Bankers Association, 2026 commercial and multifamily maturities.

• Connect CRE, CMBS delinquency, July 2026 (Trepp data).

• Freddie Mac, Primary Mortgage Market Survey.

Maryland

• Brookings Institution, “Federal government cuts are testing Maryland’s economic resilience”, June 2026.

• Maryland Comptroller, Federal Spending in Maryland.

• Maryland Department of Labor, January 2026 and June 2026 employment releases.

• University of Maryland Robert H. Smith School of Business, “Maryland Businesses Express Cautious Optimism for 2026”.

• Maryland Matters: Moody’s downgrade · $1.4 billion gap · state drops Moody’s.

• Maryland Department of Legislative Services, June 2026 GO Bond Sale Report.

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice, and I’m not your lawyer — I’m a guy from Baltimore with a spreadsheet and a law degree, in that order. Every rate in this piece is a snapshot: the yields are August 17–19, 2026, prime is the Federal Reserve’s H.15 as of August 19, the SBA 504 rate is the August pricing and resets monthly, and the NFIB and FDIC figures are monthly and quarterly releases that will have moved by the time you act on them. Check the source before you sign anything.

Coffee’s on. Let’s go through the books.

Last week I told you to wait until Friday before believing anything about the July jobs trend. Friday came. The number is better than I expected, and it arrived in the same week the federal government put a fifty percent wall around a chunk of our largest export market. Both of those things are true at once, which is most weeks around here.

On the ledger this week: the country shed jobs in July and Maryland added 11,700; 50% tariffs on Canadian goods took effect at midnight Friday and Canada answers on September 8; there’s a $50,000 grant that closes this Friday if you’re on the right side of the state; the $14 million loan round is eight days in and I owe you a correction on its deadline; FinCEN just killed beneficial ownership reporting permanently; a new law lands October 1 whose required poster may not exist until November 1; and there’s a measles outbreak in three counties that your staffing plan should know about.

1. The country lost jobs in July. Maryland added 11,700.

This is the one that actually matters, and it’s the rare case where the local number and the national number point in opposite directions.

Nationally, July was bad. In its August 7 report, the Bureau of Labor Statistics found U.S. nonfarm payrolls fell by 23,000 — against a consensus expecting somewhere between 83,000 and 95,000 gains. BLS also cut its May and June estimates by a combined 103,000 jobs, while the national unemployment rate ticked down to 4.1% and labor force participation fell to 61.4% — a level the country hasn’t seen in more than five years. A falling unemployment rate alongside falling participation is not good news; it’s people leaving.

Two weeks later, Maryland’s July numbers landed on Friday, August 21: 11,700 nonfarm jobs added, with the private sector accounting for 10,900 of them. The state’s unemployment rate dipped to 4.2%. June was revised up by 300, to a net gain of 1,200.

Where the jobs were: Health Care and Social Assistance added 4,800. Professional, Scientific, and Technical Services added 3,000. Accommodation and Food Services added 1,400 — worth noting if you run a restaurant and have been telling yourself nobody’s hiring. Administrative and Support and Waste Management added 900. Manufacturing and Government each added 800. Five sectors lost ground, led by Retail Trade at -1,200, with Information down 300, Other Services down 300, Construction down 200, and Management of Companies down 200.

The longer line is the real story. Through seven months, Maryland has added 29,800 nonfarm jobs — 1.1% growth against a national rate of 0.3%. Labor force participation here is 63.8%, against 61.4% nationally. Maryland is growing nearly four times faster than the country, and doing it while still absorbing federal job reductions.

Now the honest part. Monthly state estimates get revised, sometimes hard — the national revisions above should tell you exactly how much confidence to place in a single month. And none of this says anything specific about Baltimore. The county, city, and MSA-level July data doesn’t publish until Wednesday, September 2. Anybody telling you this week what happened in Anne Arundel in July is telling you a story.

What I’d do: If you’re hiring in retail, construction, or information, your leverage went up this month and the cheerful headline hides that. If you’re hiring in health care or professional services, it did not — you’re competing, and you should be benchmarking pay rather than reposting the same listing. Either way, mark September 2 and don’t make a Baltimore-specific decision before it.

2. A 50% tariff wall went up at midnight Friday. Canada answers September 8.

Trade talks between the United States and Canada collapsed Friday night, and at the stroke of midnight — so, Saturday, August 22 — 50% tariffs took effect on roughly $20 billion worth of Canadian goods — including dairy products, alcoholic beverages, cement, and hockey equipment. U.S. Customs and Border Protection told importers its officers would be checking compliance immediately after the deadline passed. On Saturday, Prime Minister Mark Carney said Canada would match the tariffs “dollar for dollar,” beginning Tuesday, September 8 — the Tuesday after Labour Day — and named steel, dairy, appliances, agricultural equipment, pulp and paper, and electronics as targets.

I want to be careful here, because tariff coverage tends to arrive at one of two temperatures and neither is useful.

Here’s the sober version. Canada was Maryland’s leading export destination in 2024, taking $2.2 billion in machinery, pharmaceuticals, and printed material, according to a Baltimore Banner analysis of Census data. So if you export, this is real. But that same analysis found exports are only about 3% of Maryland’s GDP — 43rd in the country — and imports about 8%, ranking 24th. As a state, we are not Michigan. Our exposure runs through the Port and through federal spending far more than through trade flows.

That’s the macro. The micro is different, and the micro is where you live. If you buy cement, lumber, appliances, restaurant liquor, or dairy at wholesale, the cost line moves regardless of what share of state GDP trade represents. And note the timing asymmetry: the U.S. tariffs are already live, Canada’s arrive in two weeks. Your input costs may move before your customers’ spending does.

One caveat I’ll flag plainly: the Banner figures are 2024 Census data published in May 2025. They’re the best structural picture I have of Maryland’s trade exposure, but they predate this round entirely.

What I’d do: Pull your last three purchase orders and find out where the goods actually originate — not where you buy them, where they’re made. Most owners can’t answer that in under an hour, and that hour is the whole exercise. Then call your two largest suppliers this week and ask directly whether they’re repricing and when. Get it in writing if you can. If you have a fixed-price customer contract running past September, read the escalation clause now rather than in October.

3. There’s $50,000 on the table that closes Friday — if you’re in the right two counties

The MDOT Small Business Grant Program’s Round 2 closes Friday, August 28 at 11:59 p.m. Direct grants up to $50,000 for small businesses hurt by Purple Line construction. Awards get announced September 29.

Let me get the disqualifier out of the way first, since most of you reading this in Baltimore are about to waste three minutes: your business must be within a quarter mile of the existing or future Purple Line light rail tracks. That’s Montgomery and Prince George’s. If you’re in Hampden, this is not your grant.

If you are on the corridor, the rest of the gate: open and operational at your current location before July 1, 2023, and currently open to customers. Fifty or fewer employees. Annual gross sales not exceeding $2 million in the most recently completed five fiscal years. Active and in good standing with SDAT — a screenshot from Maryland Business Express suffices, no formal certificate needed. Plus a W-9. Funds can go to payroll, rent or mortgage on business property, supplier payments, equipment, utilities, and working capital. Excluded: cannabis, tobacco and alcohol businesses, pawn shops, gun shops, massage parlors, adult entertainment, tanning salons, gambling, and tattoo parlors.

Two things the program page says out loud that I’d rather you hear from me than learn later. First: “This is a highly competitive grant program and not all eligible applicants will receive grants.” Awards are prioritized by proximity to the most significant construction impacts in the last six months or the next six. Second: grant payments are taxable income, and they’re subject to interception if you owe the IRS or the Comptroller. Since 2025 the program has moved $2.6 million to nearly 250 businesses, which is real money and also tells you the average award is well under the $50,000 headline.

What I’d do: If you’re on the corridor, apply this week — the SDAT good-standing screenshot and the W-9 are a twenty-minute problem, and an incomplete application is not eligible for award. Note that since 2025 you have to reapply for each round, so a Round 1 application doesn’t carry over. If you miss Friday, Round 3 opens September 29 and closes November 7; put both dates in your calendar now.

4. The $14 million round is eight days in — and I got the deadline wrong last week

Housekeeping first, because I’d rather correct myself than be quietly wrong.

Last week I told you DHCD’s $14 million Small Business Direct Loan round closes at 5:00 p.m. on September 17. The program page says 11:59 p.m. You get an extra seven hours, near enough. Don’t use them. I also described the proceeds as usable for “refinancing debt,” and the page is narrower than that: refinancing debt as part of a broader project. If your plan is purely to refinance a merchant cash advance with nothing else attached, that reads differently to an underwriter than I implied, and you should raise it in the consultation rather than discover it in review.

The rest holds up, and DHCD’s new website did launch as advertised, so your old bookmark is dead. Application status is Open. Up to $2 million at 4% fixed, terms up to 30 years. Collateral required, personal guarantees required. You must be inside a Sustainable Community or Priority Funding Area. Preference goes to projects that rehab vacant or blighted property, create housing, expand fresh food access, preserve or expand child care, or demonstrate clear community need. The exclusion list still catches people: no residential projects, no nursing homes or assisted living or group homes or shelters, no liquor stores, gun shops, pawn shops, tanning salons, tattoo parlors, massage parlors, gambling, or adult entertainment.

Eight days of the window are already gone; twenty-three remain. In a scored, ranked round, that is not a lot of runway.

What I’d do: Schedule the consultation today if you haven’t. It is the only cheap way to find out whether you’re inside a Priority Funding Area and whether your project reads as “community value” to the people scoring it. And I’ll repeat last week’s warning because it hasn’t gotten less true: 4% fixed for thirty years is genuinely cheap money, and it is still your house behind the personal guarantee.

5. FinCEN just killed beneficial ownership reporting. For good.

If you formed an LLC in Maryland any time in the last few years, you probably spent an unpleasant afternoon on beneficial ownership reporting, or paid somebody to. That’s over.

On August 11, Treasury’s Financial Crimes Enforcement Network issued a final rule — effective August 14 — permanently exempting domestic reporting companies from beneficial ownership information reporting under the Corporate Transparency Act. It also exempts U.S.-person beneficial owners and company applicants from having their information reported at all, and relieves U.S. persons of any duty to update or correct what they already filed. The rule adopts, with limited changes, the interim final rule FinCEN issued back in March 2025. FinCEN has also said it will run a one-time process to remove from its database the information that wouldn’t have been required had this rule applied from the start.

NFIB — which has been pushing on this for years — puts the compliance savings at over $128 billion nationally across more than 32 million businesses and is still asking Congress to repeal the underlying mandate outright. That last clause is the part worth holding onto: the Corporate Transparency Act is still a statute. What changed is a rule interpreting it, and rules are easier to change than laws. NFIB is an advocacy organization and the $128 billion figure is theirs, not Treasury’s.

What I’d do: Stop paying anyone for BOI filing or BOI monitoring. Check whether your registered agent or formation service has a recurring BOI line item on your invoice and kill it — this is exactly the kind of subscription that survives its own purpose. If you have a foreign-formed entity registered to do business in Maryland, don’t assume you’re covered by this; the exemption is built around domestic reporting companies and U.S. persons, and that’s a question for your lawyer, not a newsletter.

6. October 1 brings a new law. Its required poster doesn’t exist until November 1.

Senate Bill 417, the Maryland Worker Freedom Act, was enacted April 28 as Chapter 221 and takes effect Thursday, October 1. It prohibits you from taking adverse action against an employee — or refusing to hire an applicant — because they decline to attend or participate in an employer-sponsored meeting where you express opinions on religious or political matters. “Political matters” expressly includes the decision to join or support a labor organization.

What’s still allowed: voluntary meetings on those subjects, and required training related to compliance with federal or state law, including discrimination, harassment, and workplace safety. What it costs if you get it wrong: the statute lets the Commissioner of Labor and Industry assess a civil penalty of up to $10,000 for an initial violation and up to $25,000 for a subsequent one, plus cease-and-desist orders, reinstatement, back pay and benefits, compensatory damages, and attorneys’ fees. You’ll see “$25,000” quoted flatly in a lot of the law-firm write-ups; read the chapter itself and the two-tier structure is right there.

Then there’s the notice requirement, which is where the calendar gets silly. You must post a model notice in the workplace and give it to each new employee on hire. The Commissioner is required to make that poster and model notice available by November 1, 2026. The law binds you on October 1. The document you’re supposed to post shows up as late as November 1.

And the caveat that may end up mattering more than any of the above: most similar state laws have been challenged as preempted by the National Labor Relations Act or as infringing employers’ First Amendment rights. Maryland’s may well get sued too. I would not plan around it losing.

What I’d do: This is a fifteen-minute fix for most small employers, so just do it. Look at any all-hands or mandatory meeting where you talk politics, elections, or unionization — and be honest that “mandatory” includes the meeting nobody’s technically required to attend but everybody understands they’d better. Make attendance genuinely optional and say so in the invitation. Leave your harassment and safety training exactly as it is; that’s carved out. Then set a calendar reminder for November 1 to check the Commissioner’s site for the poster, because nobody is going to mail it to you.

7. A measles outbreak in three counties, and what it means for your schedule

The Maryland Department of Health declared a measles outbreak Monday, August 24, in St. Mary’s, Carroll, and Cecil counties, following 14 new confirmed cases. Maryland is at 31 confirmed cases for the year. Twenty-seven of those have come since July 15, all in unvaccinated residents: St. Mary’s (13), Carroll (8), Cecil (5), and Charles (1). The four earlier cases were in Anne Arundel (2), Baltimore (1), and Prince George’s (1).

I’m including this in a business newsletter for one reason: if someone is not protected and gets exposed, there’s a 90% chance they become infected, and the public health guidance is that exposed or symptomatic people stay out of work. For a six-person shop, that isn’t a health story, it’s a scheduling story.

The honest framing, straight from the Department: no public exposures have been identified in the new cases, there’s no evidence of widespread community transmission, most cases occurred among household contacts, and overall risk remains low if you’re fully vaccinated. More than 106,000 MMR doses have been administered in Maryland this year, over 24,000 of them since July 1.

And one thing I’d rather say than let you assume: this is not somebody else’s county. Carroll is in the Baltimore metro, and it has the second-highest case count of the three. Anne Arundel had two of the four earlier cases. It isn’t a Baltimore City story, but it isn’t Southern Maryland’s problem either.

What I’d do: Nothing dramatic. If you operate in St. Mary’s, Carroll, or Cecil — or run child care anywhere in the state — know that your coverage plan has a hole in it if two people are told to stay home for a week, and figure out now who backfills. Point staff at MDH’s measles page rather than answering medical questions yourself, and let them check their own records. And if you’re the owner who hasn’t taken a sick day in four years, note that you are the single point of failure you keep worrying about.

Share

Subscribe now

The week ahead

• Friday, August 28, 11:59 p.m. — MDOT Purple Line Small Business Grant Round 2 applications close. Awards September 29.

• Wednesday, September 2 — Maryland’s July substate employment data: county, city, and MSA. This is when Baltimore and Anne Arundel become real numbers instead of vibes.

• Thursday, September 3 — Monthly Labor Review, Maryland’s narrative writeup of the same data.

• Monday, September 7 — Labor Day.

• Tuesday, September 8 — Canada’s retaliatory tariffs take effect on steel, dairy, appliances, agricultural equipment, pulp and paper, and electronics.

• Thursday, September 17, 11:59 p.m. — DHCD’s $14 million Small Business Direct Loan round closes. Twenty-three days from today.

• Friday, September 18 — Maryland’s August statewide employment report.

• Tuesday, September 29 — Purple Line Round 2 awards announced; Round 3 applications open, closing November 7.

• Thursday, October 1 — Maryland Worker Freedom Act takes effect. The poster you need follows by November 1.

See you next Monday.

Sources

• Little change in nonfarm payroll employment in July 2026 — U.S. Bureau of Labor Statistics, The Economics Daily

• U.S. jobs report July 2026: Payrolls fell 23,000 — Yahoo Finance (national July figures, prior-month revisions, participation rate)

• Maryland employment rose by 11,700 jobs in July, unemployment rate drops — Nottingham MD, August 21, 2026 (sector detail, June revision)

• Maryland Job Growth Outpaces US in July — The Southern Maryland Chronicle, August 24, 2026

• Maryland Employment and Unemployment Press Release Schedule — Maryland Department of Labor (statewide, substate, and Monthly Labor Review dates)

• As Canada readies retaliatory tariffs, Mark Carney says his nation is ‘at war’ with U.S. — NPR, August 22, 2026 (50% tariffs, $20 billion, covered goods, September 8 retaliation, CBP bulletin)

• Maryland trades with over 200 countries. Trump’s tariffs could change the numbers. — The Baltimore Banner, May 27, 2025 (2024 Census trade data; Canada as leading export partner; trade as share of Maryland GDP)

• MDOT Small Business Grant — Purple Line MD (Round 2 dates, eligibility, exclusions, competitiveness, 2026 round schedule)

• Maryland Business Express Entity Search — Maryland SDAT (good-standing documentation)

• Small Business Direct Loans — Maryland DHCD (application window, terms, eligible uses, exclusions)

• Maryland DHCD Announces Small Business Lending Program, Competitive Round for Low-Cost Loans — Maryland DHCD, July 8, 2026

• FinCEN Ends Beneficial Ownership Reporting Requirements for U.S. Companies — Morrison Foerster

• U.S. FinCEN Issues Final Rule Ending Beneficial Ownership Reporting Requirement — Sidley Austin

• BOI Final Rule is a Major Win for Maryland Small Businesses — NFIB, August 13, 2026

• Chapter 221 (Senate Bill 417) — Maryland Worker Freedom Act — Maryland General Assembly, enacted April 28, 2026

• Forthcoming Maryland Laws Target Employer Practices and Labor Rights — Ogletree Deakins, April 14, 2026 (penalties, exemptions, preemption challenges)

• Maryland Worker Freedom Act: Maryland Employers Face New Limits on Mandatory Meetings — Rubin Fortunato (November 1 notice deadline)

• Maryland Department of Health declares measles outbreak in three Maryland counties — Maryland Department of Health, August 24, 2026

• Measles — Maryland Department of Health

Crabcakes & Commerce is written for Maryland small business owners. Nothing above is legal, tax, or financial advice — I’m a guy from Baltimore with a spreadsheet and a bar card. Program details, dates, and rates can change; verify with the agency before you act.