Why firms are hoarding labour through downturns instead of cutting jobs

The pattern that governed labour markets through every downturn since the early nineteen-eighties appears to have broken. Firms across developed economies are holding onto workers even as demand softens and unit labour costs climb, behaviour that would have been dismissed as irrational during previous cycles. What was once the first line of adjustment—headcount reduction—has become a last resort, and the implications stretch from central bank policy rooms to the boardrooms now wrestling with stagnant productivity.

The numbers tell an unusual story. Employment levels in major economies have remained remarkably stable even as growth indicators have weakened and profit margins have come under pressure. Unemployment rates have edged up only modestly, and job openings, while down from their peaks, remain elevated by historical standards. The swift, deep cuts that characterised downturns in the past have given way to hiring freezes, reduced hours, and wage restraint. Redundancies are happening, but not at the scale or speed that past relationships between output and employment would predict.

Demographic reality is the most fundamental driver. Across Europe, North America, and parts of Asia, working-age populations are shrinking or growing far more slowly than in previous decades. The pipeline of new entrants is narrower, and the experience of the past few years has convinced many executives that labour scarcity is not a temporary aberration but a structural condition. Letting people go in a downturn no longer guarantees the ability to rehire them when demand returns. The workers may have moved industries, retired early, or found employers willing to offer more stability.

The pandemic recruitment crisis left a deep imprint. Firms that slashed payrolls in early twenty-twenty found themselves unable to rebuild teams quickly when demand rebounded. Supply chains faltered not only because of logistics but because there were not enough workers to staff warehouses, factories, and service counters. The scramble to hire drove wages up sharply and forced companies into bidding wars for talent. Senior managers who lived through that period are now reluctant to repeat the experience. The cost of losing institutional knowledge and the time required to bring new hires up to speed have been reassessed, and the calculus has shifted in favour of retention.

Hiring and training costs have risen substantially, and not only in nominal terms. Regulatory complexity, compliance requirements, and the administrative burden of onboarding have all increased. Technology roles require longer ramp-up periods. Even in sectors once considered low-skill, the expectation of digital literacy and adaptability has raised the bar. The fixed costs associated with each employee have grown, making the decision to cut and later rehire more expensive than it once was. Retaining a worker through a slow patch, even at reduced utilisation, can be cheaper than the round-trip cost of dismissal and replacement.

There is also a reputational dimension. In tight labour markets, how a firm treats its workforce during downturns becomes a signal that prospective employees notice. Companies that maintained payrolls during the pandemic earned goodwill; those that did not faced scepticism when they later sought to hire. Employer brand, once a concern mainly for consumer-facing businesses, now matters in the competition for talent across sectors. The cost of damaging that brand is harder to quantify but increasingly factored into retention decisions.

Yet this shift is not without consequence. When firms hold headcount steady while output falls, productivity—defined as output per worker—declines mechanically. Labour hoarding smooths employment volatility, but it also means that workers are producing less per hour. Unit labour costs rise because wages continue even as the revenue generated per employee drops. In a healthy expansion, productivity gains help absorb wage increases. In a slowdown with sticky employment, that cushion disappears.

The productivity problem is not self-correcting. Wage growth alone cannot solve it; in fact, continued wage increases without corresponding output gains worsen the margin squeeze. Firms can try to offset the pressure by raising prices, but in a weak demand environment that option is limited. They can cut other costs—capital expenditure, marketing, research—but those cuts carry their own long-term risks. Or they can accept lower profitability and hope that demand recovers before financial pressure forces harder choices.

For central banks, labour hoarding complicates the interpretation of employment data. Historically, a resilient labour market signalled economic strength and justified continued monetary tightening. But if employment is being sustained through reduced hours, wage restraint, and managerial caution rather than robust demand, the signal is muddier. Unemployment may stay low even as economic activity weakens, delaying the recognition of a downturn and potentially keeping policy tighter for longer than underlying conditions warrant.

The political economy also shifts. Governments have grown accustomed to pointing to low unemployment as evidence of economic health. If joblessness remains subdued even as growth slows, the pressure to respond with fiscal stimulus may be less acute. Yet the lived experience of workers—stagnant wages, fewer hours, diminished prospects for advancement—may not align with the headline figures. The gap between labour market statistics and economic sentiment could widen, with consequences for policy credibility.

There is a counter-argument worth taking seriously. Labour hoarding may be a rational response to genuine uncertainty about the depth and duration of a slowdown. If firms believe the weakness is temporary, retaining workers preserves optionality and avoids the costs of adjustment in both directions. In that view, the behaviour is not a structural break but a reasonable hedge in an unusually uncertain environment. The question is whether that uncertainty will persist, and whether firms can afford to wait it out.

The risk is that labour hoarding becomes a trap. If demand does not recover quickly, firms will eventually be forced to cut, but by then the productivity losses will have accumulated and financial buffers will have eroded. The delayed adjustment could be sharper and more disruptive than if it had happened earlier. Alternatively, if firms adjust by other means—automation, offshoring, or simply accepting lower growth—the employment stability may prove durable but at the cost of dynamism.

What remains uncertain is how the next genuine recession will unfold. If labour hoarding persists, unemployment may rise more slowly than in past downturns, but the eventual adjustment could be more abrupt once firms exhaust their capacity to absorb losses. If the pattern breaks and firms revert to swift cuts, it will suggest that the recent behaviour was contingent on the specific circumstances of the post-pandemic period rather than a lasting shift.

The stakes are considerable. A labour market that no longer responds predictably to demand shocks challenges the frameworks that guide monetary policy, fiscal planning, and business strategy. Productivity growth, already weak across much of the developed world, faces a new headwind if firms prioritise employment stability over output efficiency. The question is not whether labour hoarding is good or bad in isolation, but whether economies can sustain it without sacrificing the dynamism that underwrites long-term prosperity.

This article was produced with AI assistance and reviewed against our editorial standards.