Regional Evolutions: Blanchard and Katz (1992), reconstructed and extended to 2025

Blanchard and Katz asked whether a state that loses jobs gets them back or loses workers instead. They took annual state data for 1947–90, all relative to the US average, and traced shock responses with autoregressions and a VAR. The jobs never come back: employment settles 1.3 percent below its old path, while unemployment and participation recover in five to seven years as people move away. Our extension through 2025 finds similar broad adjustment patterns, with migration absorbing a smaller share of the initial job loss in the periods studied after 2008.

The paper's main claims

1. Relative employment shocks are permanent: a 1 percent loss ends 1.3 percent down for good.
2. Year one: unemployment up 0.32 points, participation down 0.17.
3. Both vanish within 5–7 years.
4. Out-migration absorbs 0.65 of each lost job in year one.
5. State growth persists: 1970–90 on 1950–70 has slope 0.70.

What we were able to do

We assembled a state panel from public BLS and BEA series retrieved through FRED: payroll employment, unemployment, labour force, population, and per-capita income. The unemployment data used here begin in 1976. The wage comparison uses per-capita income as a proxy for manufacturing production-worker earnings, and the trivariate analysis uses resident population as a proxy for working-age population.

We wrote new Python code to implement the paper's regressions, autoregressions, and vector autoregression, using pandas, NumPy, SciPy, and statsmodels. We then extended the public-data panel through 2025 and estimated the responses for several periods.

What we got when we re-ran it

We reconstructed the state panel from publicly available BLS and BEA data and compared our estimates with the paper's published results. The reconstruction closely follows the paper's main pattern: a local job loss leaves employment persistently lower, while unemployment and labor force participation gradually recover and migration accounts for much of the adjustment. The employment responses in Table 1 are especially close: at year 20, both the paper and our reconstruction give 1.53 for the response to an initial one-unit employment increase.

The largest difference in the labor-market adjustment results is how the initial response divides between unemployment and participation. After a 1 percent job loss, we estimate a first-year unemployment increase of 0.20 percentage points, compared with 0.32 in the paper, and a participation decline of 0.49 points, compared with 0.17. The implied first-year migration share is much closer: 0.60 of each lost job, compared with 0.65. Both analyses show lasting employment losses and eventual recovery in unemployment and participation, although our estimated employment loss is deeper and unemployment takes longer to recover.

These comparisons use reconstructed series and, in some cases, shorter sample periods or different measures. In particular, our wage comparison uses per-capita income rather than the paper's manufacturing wage measure, so it is not a direct replication of that result. The tables below show the estimates and data differences explicitly.

Table 1 side by side.

Table 1. Univariate Models of Relative Employment, Unemployment, and Wages

Notes: Paper columns are the published values. Source (paper): estimates of univariate equations using data described in the paper's appendix; periods of estimation are 1952–90 for employment and wages and 1972–90 for unemployment; standard errors of the coefficients are in parentheses. Our columns: employment 1952–90 from FRED/BLS reconstructions (n = 1,953); unemployment 1976–90 LAUS only (n = 663), against the paper's 1972–90 panel with constructed pre-1976 data; wage proxied by per-capita income (n = 1,985). Replicated impulse responses are re-indexed to the paper's convention with Year 1 = 1.00 at impact. The regression standard error row and the unemployment Year 20 cell were not stored by our run (n.a.).

Our version of Figure 7.

Notes: paper values are the published text quantities; ours are from the trivariate VAR re-estimated on 1978–90 with LAUS/CPS-derived series and resident population proxying working-age population. The split of the first-year adjustment between unemployment and participation does not replicate on our household-basis series, though the implied net out-migration share (0.60 against 0.65) and the employment path (peak, hump shape, permanence) do; the paper's own footnote 37 reports that the household basis gives a smaller initial migration share (0.40) than its establishment-basis headline (0.65).

And the opening scatter.

The same analysis on today's data

Extending the analysis through 2025 preserves the broad adjustment pattern, with differences across periods. Employment losses remain persistent, while unemployment and participation adjust over time. On a comparable household employment basis, the implied migration share of the first-year adjustment falls from 0.38 in 1978–90 to 0.24–0.28 in the periods we study after 2008. The relationship between state employment growth in successive periods also weakens. The tables report estimates for each period, with 90 percent bootstrap intervals in brackets. The adjustment quantities come from the paper's trivariate VAR:

$$\begin{aligned} \Delta e_{it} &= \alpha_{i10} + \alpha_{i11}(L)\,\Delta e_{i,t-1} + \alpha_{i12}(L)\,le_{i,t-1} + \alpha_{i13}(L)\,lp_{i,t-1} + \varepsilon_{e,it} \\ le_{it} &= \alpha_{i20} + \alpha_{i21}(L)\,\Delta e_{it} + \alpha_{i22}(L)\,le_{i,t-1} + \alpha_{i23}(L)\,lp_{i,t-1} + \varepsilon_{u,it} \\ lp_{it} &= \alpha_{i30} + \alpha_{i31}(L)\,\Delta e_{it} + \alpha_{i32}(L)\,le_{i,t-1} + \alpha_{i33}(L)\,lp_{i,t-1} + \varepsilon_{p,it} \end{aligned}$$

with $\Delta e_{it}$ relative employment growth, $le_{it}$, $lp_{it}$ the relative log employment and participation rates, and shock $\varepsilon_{e,it}$ ordered first.

Permanence first.

Notes: paper columns are the published Table 1 impulse responses; ours re-estimate the same pooled autoregressions on the modern windows shown, with 90 percent state-cluster bootstrap bands in brackets. The sweep stored only the horizons shown; the wage column is not extendable because no modern equivalent of the paper's manufacturing production-worker earnings series exists.

The Figure 7 quantities era by era.

Notes: household employment basis, 90 percent state-cluster bootstrap bands in brackets; window labels follow the paper's sample notation but, except for 1978–90, are our era choices. Anchor: the paper's printed 1978–90 values (establishment basis) are a +0.32 point first-year unemployment rise, effects gone in five to seven years, a −1.3 percent employment plateau, and a 0.65 first-year migration share (0.40 on the household basis of its footnote 37, the basis our panel uses). The full-sample first-year unemployment response of 0.32 landing on the printed +0.32 is a coincidence of the pooled sample, not a replication.

On the payroll basis the migration share falls 0.77 to 0.63, and the model refits with a halved migration parameter.

And the scatter, pair by pair.

How this relates to later research

Our finding of a smaller migration response is qualitatively consistent with Dao, Furceri, and Loungani (2017, Review of Economics and Statistics), who extended the Blanchard–Katz framework and found that interstate mobility in response to relative labor demand conditions had weakened since the early 1990s. They also found stronger migration responses during recessions, underscoring the importance of distinguishing changes across decades from changes over the business cycle.

Ganong and Shoag (2017, Journal of Urban Economics) study the related decline in state income convergence and migration toward high-income places, linking those changes to housing costs and supply restrictions. Their income-convergence measure differs from our comparison of employment growth across successive periods. Our extension through 2025 adds more recent evidence on regional adjustment; it does not by itself test their housing mechanism or provide a like-for-like quantitative comparison with either study.

Notes

These are the judgment calls and data limits behind our numbers, so you can decide how much weight each result deserves.

• Table 1's wage column is a per-capita-income proxy (we did not use the paper's manufacturing wage series); the published overshoot (1.07 at year 2) comes out 0.82.
• The paper's unemployment sample is 1972–90 with constructed pre-1976 data, ours uses LAUS data for 1976–90 (n = 663). The unemployment response is initially larger in our reconstruction (1.07 vs 0.90 at year 2), then falls to 0.23 vs 0.29 at year 5.
• Our household-basis series miss the year-1 split (+0.20 vs +0.32, −0.49 vs −0.17) and deepen the plateau (−1.8 vs −1.3; payroll recovers −1.5); the paper's footnote 37 separately reports that the first-year migration share depends on the employment measure (0.40 on a household basis vs 0.65 on an establishment basis).
• Unstored cells (the standard error row, unemployment Year 20, some extension horizons) show n.a., never interpolations; impulse responses are re-indexed to Year 1 = 1.00.
• Okun coefficients and state-level long-run responses are left off (non-comparable units; near-unit-root explosions); everything here is pooled.
• All windows except 1978–90 are our era choices.

Data sources for the extension studies

The extension studies pulled fresh public data on August 24, 2026, all of it from FRED (Federal Reserve Bank of St. Louis). In full:

• U.S. Bureau of Labor Statistics, Local Area Unemployment Statistics: state unemployment rate, state civilian labor force, and state labor force participation rate, 50 states and DC, monthly, seasonally adjusted, 1976 to 2026, retrieved from FRED.
• U.S. Bureau of Labor Statistics, State and Area Employment: total nonfarm employees by state, monthly, 1939 to 2026, retrieved from FRED.
• U.S. Bureau of Labor Statistics, national series PAYEMS (total nonfarm payrolls, from 1939) and the Current Population Survey series CE16OV, CLF16OV, CIVPART and UNRATE (from 1948), retrieved from FRED.
• U.S. Bureau of Economic Analysis, per capita personal income by state (annual, 1929 to 2025) and for the United States (series A792RC0A052NBEA), retrieved from FRED.

This reconstruction uses public BLS and BEA data; comparisons with the paper use its published estimates. The theory update fetched no new data; it works entirely from the living update's panel.

The full replication working directory, including both extension studies, is public at github.com/living-sciences/living-regional-evolutions.