What the pandemic reveals about the risks that people are willing to take at work.

When the workplace becomes less safe, how much do people need to be paid to stay? Millions of workers suddenly faced that question in the starkest terms in 2020. For many, showing up to work meant exposure to a deadly virus, but reducing their working hours meant a sharp drop in income.

That uncomfortable trade-off lies at the heart of our recent study in which we present a new way to put a price tag on something that economists have long struggled to measure: the value of non-monetary aspects of work.

Jobs offer more than just a wage. Some are safer, more flexible, more predictable. Others offer status or purpose, or simply the ability to work without getting sick. These "non-wage amenities" matter deeply to workers, but they're notoriously difficult to measure. You can't see them on a payslip. You often can't compare them across jobs. And asking workers directly - perhaps via surveys - can yield vague or inflated answers.

Knowing how much workers value workplace amenities helps firms to assess what changes are worth making

Our research develops a method that overcomes those limitations by looking at how American workers made high-stakes decisions during the pandemic when they faced trade-offs between losing income and getting exposed to Covid-19.

During the crisis, Congress passed the Federal Pandemic Unemployment Compensation programme, introducing a weekly $600 supplement to regular unemployment benefits. But the payment came with a strict earnings limit. Even a small amount above the threshold disqualified someone from receiving the full benefit.

This created a stark trade-off for many hourly workers: either they could avoid taking on shifts that would push their earnings above the threshold; or they could work more hours, earn higher wages and forgo the benefit while accepting the associated health risks.

Predictably, many workers adjusted their hours to keep their entitlement, creating a visible pile-up of earnings just below the limit - a phenomenon known in public finance as "bunching at eligibility thresholds". We use the bunching concept to measure the amount of income that workers are willing to give up for better job conditions. Whether it is worthwhile to work more depends not only on the amount of extra earnings, but also on the type of non-wage experiences that workers would have.

When jobs entail high risk, it is more appealing to reduce work hours and remain below the benefit eligibility threshold. While this leads to earning losses for hourly workers, it also reduces the risk of falling sick.

We expected more bunching during periods when the Covid-19 risk was high. But this insight extends beyond that particular experience. When the non-monetary aspects of work are unattractive, reducing hours is less of a sacrifice and workers have greater incentive to bunch. By contrast, they are less likely to do so when working extra hours comes not only with more income but also with large non-wage returns from working.

When the non-monetary aspects of work are unattractive, reducing hours is less of a sacrifice

From behaviour to value

Our study shows that the value of amenities such as safety can be deduced from workers' responses to benefit eligibility thresholds. When amenities improve, bunching decreases and when amenities deteriorate, it increases. This method translates observed changes in behaviour into a monetary value for workplace safety.

Workers were willing to give up around 9% of their pay to reduce the risk of death by one in 100,000

We find that workers were willing to give up around 9% of their weekly pay to reduce the risk of death by one in 100,000. To make this number tangible, consider the difference in workplace risk between a librarian and a roofer in normal times. A roofer would have to be paid 30% more than a librarian to make the average person indifferent between the two jobs.

Economists often use a concept called the "value of a statistical life" to estimate how much people are willing to pay to reduce mortality risks. It doesn't reflect the worth of any one person's life. Instead, it captures how much many people would collectively pay for small safety improvements that, on average, save one life.

In this study, the implied statistical value of life was around $5.6mn - rising to $8mn when accounting for likely misperceptions of Covid-19 risk. This aligns with values used by US agencies when weighing the costs and benefits of safety or health regulations.

But these numbers are significantly higher than the approximately $3mn figure currently used by the UK government, suggesting that the British public may value safety more than current policy guidelines assume. The numbers also highlight a potential mismatch: the hazard pay offered by many employers was below the value workers seemed to place on avoiding Covid-19 risk.

Our study suggests the compensation required to offset fully the cost of facing Covid-19 risk would be around $4.80 per hour - an amount far larger than the compensation that was offered in practice. On average, firms offered 11 cents more per hour, and even some of the most publicised hazard rates at major retailers topped out at between $2 and $4. Hazard pay thus fell short of compensating workers fully for the risks they faced.

What makes this approach powerful is that it doesn't rely on comparing wages across different jobs or industries. Instead, it observes how the same workers respond to changing conditions - before and after an amenity shifts in value - and infers what they are willing to give up to retain it.

This allows researchers, firms and policymakers to measure the value of job attributes that don't appear in pay packets: safety, schedule control, predictability or the ability to work remotely. It's a method built on revealed preferences - what people actually do, not just what they say.

Why this matters

For businesses, these insights offer a guide to what employees value most when they have choices. If wages are hard to move, improving non-wage conditions may offer a cost-effective path to attracting and retaining workers, especially in tight labour markets. Understanding how workers respond to changing job conditions can help employers to design roles that align better with people's preferences.

For governments, the implications are equally important. When workers are willing to give up significant income to avoid certain risks or conditions, that signals a high social value to improving those conditions. Cost-benefit analyses of workplace safety regulations, paid leave mandates or investments in public health infrastructure depend on credible estimates of the value that workers place on these interventions. Our study offers a new way to obtain such estimates.

The pandemic didn't just reveal vulnerabilities in public health systems and supply chains. It exposed something deeper: the high cost that workers silently bear when job conditions deteriorate.

By capturing the value that workers place on safety - not just in theory, but in real behaviour - this study provides a powerful tool for measuring what's often difficult to quantify precisely. In doing so, it brings us closer to a more complete picture of the modern labour market: one that allows us to value not just what people earn, but what they endure, prefer and protect when they clock in.

---

21 October 2025     Paper Number CEPCP713

Download PDF - Putting a value on safety

Download Press Release

This CentrePiece article is published under the centre's Labour programme.

This publication comes under the following theme: Work and wellbeing