What "Michigan Ranks #5" Actually Tells You

A note on reading business-survival rankings — and what Michigan's own trend line shows once you look past the leaderboard

September 2026 · Wayne Fu · Prompted by a Detroit News media request

In late August, a Detroit News reporter reached out about a new study ranking Michigan No. 5 in the country for five-year business survival — the share of new businesses still open five years after opening. It's a nice headline. It's also, on its own, close to meaningless. Here's why, and what the data actually says once you ask better questions of it.

The number behind the headline

The underlying source is real: the U.S. Bureau of Labor Statistics' Business Employment Dynamics program tracks cohorts of establishments from the year they open and measures what share are still active one, two, three, four, and five years later. A private report (Build Your Store) ran the numbers and put Michigan's five-year survival rate at 54.7%, good for 5th nationally — behind Minnesota (56.2%), a Pennsylvania/West Virginia tie (55.9%), and Illinois (55.2%). That's a real, defensible number. The problem is what people do with it next: turn a noisy, moving measurement into a fixed identity ("Michigan is a top-5 state for business survival").

A rank is not a measurement

When I pulled Michigan's rank across recent years for the reporter, it looked like this: 11th in 2015, 27th in 2020, 6th in 2023, back to 11th in 2024. That's not a state whose fundamentals are swinging wildly year to year — it's what happens when roughly 50 states are packed into a narrow band, mostly between 45% and 57%. A move of one or two-tenths of a percentage point can flip a dozen spots in the standings. Rank amplifies noise that the underlying rate doesn't actually contain. Treat any single year's rank as a snapshot of measurement error, not a report card.

Compare yourself to the field, not just to your past self

This cuts both ways, and it's the piece most coverage skips. A state can post a genuinely higher number than last year and still fall in rank — because everyone else moved too. Michigan's own five-year rate held in a strong, fairly stable range from 2022 through 2024, yet its rank still slipped from 6th to 11th over that stretch, simply because other states improved faster. The reverse is also true: a flat or even declining rate can still look like "climbing the ranks" if peers are declining faster. Either way, the story is really about the field, not about you — which is exactly why the national average has to be part of every comparison, not an afterthought.

The selection problem hiding underneath the number. There's a subtler issue, and it's the one I kept coming back to in the interview. A state's survival rate isn't a clean measure of "how good this place is to run a business" — it's partly a measure of who bothered to start a business there in the first place. Make it very easy and cheap to open a shop, and you'll pull in more casual, under-capitalized, first-time entrants — some of whom were always more likely to fold, regardless of the local economy. Make it slower and more regulatory-heavy, and only more serious, better-capitalized founders bother — which mechanically pushes the survival rate up, even if the state itself isn't "better" for business. Economists call this a selection effect (formally, a version of the Heckman selection problem): the group you're measuring outcomes for isn't a random sample, it self-selected in a way that's correlated with the outcome you're measuring. In plain terms: easy-entry places look artificially worse, high-barrier places look artificially better, and neither necessarily says anything about how healthy the underlying business climate is. A high survival rate can mean "great place to build a business" or it can mean "hard enough to get started here that only the most determined try" — the raw number can't tell those apart.

Read your own line, not the leaderboard

The best antidote to rank-chasing is to stop looking at rank at all and look at a state's own trajectory over time — which is what the tool below is built for. Pick a few states, or just leave Michigan selected, and watch the line rather than a single year's position.

BLS Business Employment Dynamics · Table 7

Cohort Survival Explorer

Of the private-sector establishments that opened in a given year, what share were still operating five years later? Pick any states below to compare their 5-year survival rate across opening cohorts from 2010 through 2020.

5-year survival rate by opening cohort
x-axis: year the cohort reached its 5-year mark (opened 5 years earlier)
Source: U.S. Bureau of Labor Statistics, Business Employment Dynamics, establishment age & survival tables, Table 7 (state files, bls.gov/bdm/bdmage.htm), retrieved August 2026. Each point is the share of establishments opened in a given March-ending year that were still active exactly five years later — e.g. the "2020" point on the x-axis is the March 2015 cohort measured at March 2020. A few states' raw BLS series contain isolated jumps from underlying data revisions (noticeable in Washington, Virginia, and a handful of others) — those are reproduced as published, not smoothed over.

So — is Michigan actually doing better?

Trace Michigan's own line and the honest answer is: modestly, and mostly it's a recovery, not a breakout. The rate dipped to about 49% for the cohort measured in 2020 — the pandemic-era low the Detroit News piece flagged — then climbed back to roughly 55% by 2025. Net across the full decade, that's close to a one-point gain (from about 54% in 2015 to about 55% in 2025), against a national average that itself only moved from about 51% to about 51%. Michigan has held a fairly steady edge over the national line for most of the decade — it just spent 2020 on the wrong side of it. That's a real, if unglamorous, story: Michigan mostly re-set after a pandemic dip and is back in its usual range, holding a consistent few-point premium over the country as a whole. It is not a state whose business-survival fundamentals have suddenly leapt forward, and it isn't a top-5 state in any way that should be read as durable or structural.

Which is close to what I told the reporter directly:

"The percentage is higher, that's it. Doesn't mean it's easier or good to run a business." Wayne Fu, quoted in The Detroit News, Aug. 28, 2026

None of this is a knock on Michigan, or on the reporters who cover these studies — a ranking is a genuinely useful hook for a story about real entrepreneurs doing real work (the piece profiled a Roseville painting company that grew from three people to 53 employees, and a Ferndale butcher-shop-turned-restaurant built with help from local financing programs). The point is narrower: the number that makes the headline — this year's rank — is close to the least informative way to read this dataset. The trend line, compared honestly against the national trend and read with an eye toward who's self-selecting into the sample, tells you far more.

Sources & further reading: BLS Business Employment Dynamics — Survival Table 7 (by state) · Build Your Store, "Business Survival Rates by State" · Myesha Johnson, "Why Michigan Is a Top State for New Business Survival," The Detroit News / TCA Regional News, Aug. 28, 2026.