Why Labor Shortages Kill Some Firms While Others Raise Wages: Japan's Paradox

Putting Two Japanese Data Points in Context

Tokyo Shoko Research reports 5,373 bankruptcies (liabilities ≥10M yen) in the first half of FY2026, up 3.8% year-on-year, the highest since 2013; 90.6% had fewer than 10 employees. Meanwhile, Ministry of Health data shows 121,000 high-school graduates seeking jobs versus 499,000 openings (4.12x ratio) in March 2026. Daiwa House raised new-graduate monthly pay from 250,000 to 350,000 yen (40% increase) in 2025, though this reflects a salary-structure revision, not a nationwide 40% jump. University graduate job-offer ratio was 1.62x for the class of 2027. These figures calibrate the paradox: the same labor shortage lets capable firms bid up wages while pushing marginal firms out of business.

The Missing Step: Can the Business Still Profit After Raising Pay?

A small restaurant operating on old wages loses staff to competitors offering better pay. To retain workers it must raise wages, but customers resist menu-price hikes and rent/material costs don't fall. The firm posts vacancies but cannot fill them—not because no workers exist, but because at that location, hours, and pay package, the firm's ability to pay does not match workers' alternatives. Tokyo Shoko Research classifies 240 bankruptcies in H1 FY2026 as "labor-shortage-related," of which 130 cited rising labor costs. This is a small share of total bankruptcies but shows that labor shortage means not just unfilled posts but also profit erosion from retention costs.

A Simple Model: Two Lines That May No Longer Overlap

Define per-employee monthly revenue v, non-labor variable cost c, and allocated fixed burden f (rent, interest). The maximum employable cost per worker is v - c - f. This includes wages, bonuses, employer social insurance—not just take-home pay. If the market cost to hire and retain a suitable worker is w, break-even requires w ≤ v - c - f.

Two illustrative firms (units: 10,000 yen per employee-month):

Firm A: v=34, c=8, f=2 → max employable cost = 24

Firm B: v=42, c=10, f=3 → max employable cost = 29

Initially both can retain staff at w=22. When market w rises to 25, Firm A loses 1 per employee-month while Firm B still has a 4 margin. The same "labor got more expensive" expands Firm B's advantage but hits Firm A's survival boundary. Firm A could raise prices or output to lift v to 35, or cut c or f, but such adjustments need time and conditions; if they lag, the firm may cut hours, shrink, or exit. This is not a bankruptcy prediction—cash reserves, borrowing, and cash-flow timing also matter—but it explains the counter-intuitive pattern: the tighter the labor market, the harder it becomes for a segment of firms to operate.

Interest Rates, Prices, and Labor Shortages Squeezing the Same Budget

Rising material costs increase c; higher borrowing costs increase f; competition for workers pushes up w. When sales prices cannot adjust accordingly, the gap between maximum employable cost and required wage widens. The Bank of Japan's October 2026 regional report confirms this divergence: many firms raise wages to attract and retain staff, while others struggle to pass on costs and cannot fund wage hikes. Therefore, rising bankruptcies cannot be attributed solely to rate hikes; the impact depends on each firm's debt level, whether rates have reset, and its pricing power.

Do Workers Automatically Flow to Better Jobs After Bankruptcies?

BOJ working paper (Abe, Hirano, Kaihatsu 2026) uses micro wage data and a firm-heterogeneity model to show how past wage rigidity hindered labor reallocation and how easing that friction could raise wages and productivity. However, the analysis does not incorporate firm entry and exit, nor does it prove that every current bankruptcy is efficient "creative destruction." A restaurant worker cannot become a semiconductor engineer overnight; local residents may not move to Tokyo. Most bankrupt firms are tiny; their exit does not instantly release enough suitable labor to fill vacancies. Training, relocation, care arrangements, and job-matching services determine whether "open positions" and "available workers" actually connect. Community store closures may also leave gaps in daily-life services.

Conclusion

Policy should support productivity improvement, better trading conditions, and helping workers transition to better jobs—not treat low wages as the default way to preserve unviable businesses, nor celebrate bankruptcies as pure efficiency gains. Japan's twin headlines remind us that evaluating the labor market requires asking not only how many jobs exist, but whether those businesses can sustainably employ people at decent wages. When workers gain more choices, firms must answer not just "do I need people?" but "can my business afford the price of keeping them?"

References

[1] Tokyo Shoko Research, 2026 H1 nationwide corporate bankruptcy statistics, 2026-10-08. Bankruptcies include legal and other restructuring forms, not strictly equivalent to legal "bankruptcy." [2] Ministry of Health, Labour and Welfare, 2026 March high-school/junior-high new-graduate public employment agency job-offer/job-seeker/placement statistics, end-March basis. [3] Recruit Works Institute, "University Job-Offer Ratio Survey (Class of 2027)," 2026-04-23. Different scope and method from high-school statistics. [4] Daiwa House Industry, "Wage Level Adjustment and Starting Salary Increase Announcement," 2025-01-20. [5] Tokyo Shoko Research, "240 Labor-Shortage Bankruptcies in Fiscal H1," 2026-10-05. [6] Bank of Japan, "Regional Economic Report: October 2026 Branch Managers' Meeting," 2026-10-08. Interview summaries, not causal identification. [7] Abe, N., Hirano, R., & Kaihatsu, S. (2026). What Prevents Productivity Gains from Translating into Wages: Lessons from Japan's Deflationary Period. BOJ Working Paper 26-E-12. Authors' views do not represent official BOJ stance.