Scientific rigor
Historical perspective
Subtle irony
Imagine you've believed your whole life that the earth is flat. Then someone shows you pictures from space. That's pretty much how economists studying labor markets feel right now. For decades, we were fed simple truths: raise the minimum wage, and people will lose their jobs. Simplify bureaucracy, and businesses will step out of the shadows. The labor market, we were told, follows the laws of supply and demand, just like a stall selling tomatoes at the market.
Except, life turned out to be far more complex than the textbooks suggested. And when researchers began to systematically gather data from hundreds of studies across different countries and eras, the picture changed beyond recognition. What seemed to be the iron laws of economics turned out to be more like our collective illusions – convenient, easy to grasp, but wrong.
Hide-and-Seek with the State: Why Businesses Go into the Shadows
Let's start with the thorniest issue: the informal economy. In developing countries, anywhere from a third to two-thirds of all economic activity happens «in the shadows.» People work without contracts, companies don't pay taxes, and no one reports to the state.
The classic explanation sounded convincing: entrepreneurs want to be legal, but the government gets in their way. Too much paperwork, too many taxes, too many complicated rules. Back in the 1980s, the Peruvian economist Hernando de Soto even calculated how many days it took to register a business in Lima – a figure that shocked the world. The conclusion seemed obvious: simplify the procedures, and businesses will come into the light on their own.
Sounds logical, right? Except when governments around the world started doing just that, nothing happened. Or, almost nothing. Meta-analyses – studies that combine the results of dozens of other studies – revealed something astonishing: simplifying registration has virtually no effect on the size of the formal sector. You can cut the process from a month to a single day, reduce the cost tenfold, and still, the lion's share of businesses will remain in the shadows.
Why? Because reality is more complicated. Most informal entrepreneurs aren't just sitting around, waiting for permission to go legit. They are either deliberately evading taxes (the parasitic version) or are simply so unproductive that they can't afford official status, even with minimal costs (the survival version).
Think of a street vendor in Lyon. He sells chestnuts from a cart. Even if you make registration free and instant, he still has to pay taxes, social security contributions, and comply with health codes. His tiny business simply wouldn't survive under that burden. He's not in the shadows because the state is in his way – he's in the shadows because his business is too fragile for the formal world.
So what does work? Surprisingly, the most effective tool is old-fashioned monitoring and enforcement. When the state starts actively identifying unregistered companies and creates a real risk of punishment, the picture changes. But there's a subtlety here: enforcement should focus on getting firms to register in the first place (the extensive margin), not on checking if they've registered all their employees (the intensive margin). The latter can be disastrous: a company might just go fully into the shadows, firing all its official staff.
Imagine a small café. The owner is registered, but only three out of ten employees are on the books. If an inspector starts aggressively demanding that everyone be officially employed, the owner might just close the legal entity and open a new café entirely in the shadows. But if the state focuses on ensuring that all cafés are at least registered, the effect will be much better.
The Minimum Wage: When the Textbooks Lie
Now, let's talk about the minimum wage. If you've ever studied economics, you were taught something like this: the labor market is in equilibrium. There's a demand for workers and a supply of them. A fair price – a wage – is established. If the government steps in and sets a minimum wage above that price, companies will hire fewer people simply because they can't afford it. It's the classic case of the road to hell being paved with good intentions. You wanted to help the poor, but you left them jobless instead.
This logic is so simple and compelling that it was repeated for decades. It became something of an economic religion. And here's the surprising part: when researchers started collecting real-world data from hundreds of studies across the globe, from the US to Brazil, from the 1920s to today, it just didn't hold up.
The key metric here is the «own-wage elasticity of employment.» It sounds intimidating, but the concept is simple: by what percentage will employment fall if the wage increases by one percent? Classical theory predicted a significant negative effect. Reality showed something else.
The average value across all studies is -0.13. What does that mean in plain English? Imagine the government raises the minimum wage by 10%. Workers should get 10% more money, but some of them will lose their jobs. Well, the job losses would amount to about 1.3%. So, for every extra €100 that workers receive, about €13 is lost due to job cuts. The other €87 stays in people's pockets.
For many categories of workers, the effect is close to zero. In other words, raising the minimum wage simply redistributes money from employers to employees, with almost no impact on the number of jobs. Yes, there are vulnerable groups: teenagers looking for their first job, the restaurant business with its low profit margins. The effects are more noticeable there. But for the majority of workers, the picture is completely different from what the textbooks predicted.
So why were economists so wrong? Here we encounter a phenomenon that permeates all of economic science: publication bias. It works like this: you conduct a study. If the results are «expected» – the minimum wage reduces employment – a journal will gladly accept your paper. If the results are zero or unexpected, your chances of publication plummet.
It's not a malicious conspiracy; it's just human psychology. Journal editors, peer reviewers, and researchers themselves – we all subconsciously look for confirmation of what we already «know.» Unexpected results raise suspicion: maybe you messed up the methodology? Maybe the data is bad? But if everything turns out «as it should», there are fewer doubts.
As a result, a distorted picture accumulates in the scientific literature. When meta-analyses begin to correct for this bias using statistical methods, the effects of the minimum wage on employment surprisingly shrink to almost zero, and sometimes even become positive.
Think about what this means. For decades, politicians were afraid to raise the minimum wage, fearing unemployment. Millions of people lived on less than a living wage because economists assured them that helping would only make things worse. But in reality, it was possible to help. We were just looking at the world through the lens of a theory that didn't work.
Who Really Decides What You Get Paid
Now for the most interesting part: monopsony. You've probably never heard the word, even though you live inside this phenomenon every day.
In the classic economic model, the labor market works like this: many employers compete for workers, and many workers choose among employers. The result is a fair wage – exactly what the worker produces. If someone tries to pay less, the worker will simply go to a competitor.
Sounds wonderful. Except in reality, it's not like that at all. Changing jobs isn't like picking apples at a market. You can't instantly switch from one employer to another. You have to search for openings, go to interviews, maybe even relocate. It takes time, money, and causes stress and risk. You have a family, ties to a place, personal circumstances. And there might not be many employers in your field – especially in small towns or specialized professions.
All of this gives employers power. They can pay you less than what you actually produce because they know it's not that easy for you to leave. This is monopsony: market power on the side of the buyer of labor.
How can this power be measured? Economists look at what's called «labor supply elasticity.» Essentially, it's a measure of how willing workers are to switch jobs if another place pays more. If elasticity is high, the market is competitive, and employers can't keep wages below a fair level. If it's low, employers have a lot of power.
The data shows that this elasticity is frighteningly low. The average is around 1.4, according to direct estimates. What does this mean? If a company raises its wages by 10%, it will only attract 14% more workers. That's very low. It means employers have enormous power: they can pay you significantly less than you're worth, and you'll still stay because your alternatives are limited.
Other estimation methods yield higher figures – around 14. But even that is far from the ideal of perfect competition, where elasticity would be infinite. The most balanced studies point to a value of around 7. This means that, on average, workers are paid about 12% less than what they actually produce. That 12% stays with the employer – simply because they have the power to pay less.
This is especially noticeable in certain professions. Nurses, teachers – where there are few employers and switching to another field is difficult. Or women with children, whose mobility is more limited than men's. Monopsony falls unevenly on the shoulders of different groups, worsening social inequality.
And here's where something amazing happens: the theory of monopsony explains why the minimum wage doesn't destroy jobs. If your employer is paying you less than you're worth because they have market power, a minimum wage increase simply cuts into their excess profits; it doesn't make you unprofitable. You still produce more than you're paid. In fact, in some cases, a wage hike can even increase employment because more people will want to work.
This turns the whole classical logic on its head. It turns out that when the state sets a minimum wage, it isn't meddling in a perfectly functioning market. It's correcting an already distorted market where employers are abusing their power.
When the Numbers Lie: The Problem No One Talks About
Now let's talk about why economic science was wrong for so long. And why you, an ordinary person, should care.
Imagine reading the news: «Scientists prove coffee causes cancer.» The next week: «Scientists prove coffee protects against cancer.» A month later: «Coffee has no effect on cancer.» You start to suspect the scientists themselves don't know what they're talking about. And in a way, you're right, but the problem is deeper than it seems.
Economic research suffers from the same ailment. Take any topic – the minimum wage, the impact of education on income, the effect of tax breaks. For every topic, there are dozens, hundreds of studies. And their results are scattered across the entire spectrum: from a «huge positive effect» to «no effect» and a «significant negative effect.»
Why does this happen? Partly because the world is genuinely complex and contextual. A minimum wage might work differently in Lyon and Marseille, in 2010 and 2025, for waiters and for programmers. But there's another reason: a systematic bias in what gets published in scientific journals.
Here's how it works. A young researcher studies the impact of some policy. They conduct a complex analysis, months of work. The result: the effect is almost zero, statistically insignificant. What next? They try to publish the work in a good journal. But the editor says, «Well, this isn't very interesting, is it? You don't have a clear conclusion.» Peer reviewers nitpick the methodology: «Maybe you just have bad data? Or the wrong model?»
The same researcher conducts another study. This time, they get a clear result: the effect is negative, statistically significant, just like in the textbook. The journal accepts the article with open arms. The reviewers are pleased: «Now this is real science!»
But here's the irony: the first study might have been correct, while the second was just a lucky fluke, statistical noise that looked like a pattern. Yet the scientific literature fills up with the latter, while the former remain in a desk drawer.
Multiply this by thousands of researchers over decades of work, and you get a systematic distortion. The literature becomes filled with studies showing «interesting» results, while papers with null effects are almost nonexistent. When policymakers and the public look at the «scientific consensus», they see a skewed picture.
Meta-analyses can correct for this, statistically, using special methods. When they are applied, the picture changes dramatically. Those very effects that seemed robust and proven shrink to almost zero. And sometimes, they even flip signs.
This doesn't mean all economists are liars or incompetent. It means the scientific process is designed in a way that systematically distorts our understanding of reality. It's not a malicious conspiracy – it's a structural problem that is rarely discussed with the general public.
Putting It All Together: A Unified Picture from Scattered Facts
Now for the most interesting part: how does this all fit together?
For a long time, economists studied these three topics – informality, the minimum wage, and monopsony – as separate problems, as if they were different worlds. But when you start to look closely, you see they are all pieces of the same puzzle.
Here's how it works. Imagine the economy as an ecosystem of companies. They are very different: from tiny enterprises on the verge of survival to successful, highly productive firms. These companies decide whether to operate legally or in the shadows, how much to pay their workers, and how many to hire.
The weakest companies can't afford to be legal – even minimal costs would kill them. They stay in the shadows, pay pennies, and balance on the edge. Mid-tier companies could formalize but are reluctant. It's more profitable for them to stay in the shadows until the risk outweighs the benefit. For strong companies, being legal is actually an advantage: it gives them access to credit, government contracts, and better workers.
Now, add monopsony. Even legal companies pay workers less than what they produce – simply because they can. Workers lack real alternatives, especially if competition in the local labor market is low.
What happens when the state introduces a minimum wage? In the classic model, it's a catastrophe leading to mass layoffs. But in the real world, where employers pay below a fair level due to their market power, the effect is completely different. The minimum wage simply reduces their excess profits. Employment hardly falls and might even rise, as more people are willing to work for a more decent wage.
What happens when the state tries to pull businesses out of the shadows by simplifying registration? Almost nothing, because the problem isn't the complexity of the procedures but the fact that many firms are simply too weak for the formal world. But if the state strengthens enforcement, increasing the risks of operating in the shadows, the picture changes. Mid-tier companies start to formalize because the risk outweighs the benefit.
These aren't separate phenomena – they are an interconnected system. And any policy that ignores these connections is doomed to fail or achieve only partial success.
Let's take a real-world example. A country decides to reduce informal employment. The classic approach: let's simplify bureaucracy! The result: almost zero. Another approach: let's lower taxes for small businesses! The result: expensive and ineffective, because the weakest firms still won't survive, and the strong ones would have formalized anyway.
Now, imagine a comprehensive approach. The state strengthens enforcement, raising the risks of operating in the shadows. At the same time, it introduces a reasonable minimum wage – this forces legal companies to pay decently but doesn't strangle them, because they were already paying below a fair level. As a result, more companies go legal, workers earn more, the economy becomes more transparent, and job losses are minimal.
This isn't a fantasy. These are the conclusions drawn from hundreds of studies when you look at them not in isolation, but together.
What This Means for Us
Why does all of this matter? Not just because it's interesting to learn how wrong economists were. It matters because these mistakes have cost millions of people real money, opportunities, and a decent life.
How many times have you heard, «We can't raise the minimum wage – people will lose their jobs»? It was repeated like a mantra. As a result, workers spent years, decades, living on wages that weren't enough for a normal life. And they could have been earning more – without catastrophic consequences for employment.
How many times have governments tried to fight the shadow economy by simplifying registration? They spent millions on bureaucratic reforms. And the effect was minuscule because they were targeting the wrong problem.
How many times have employers justified low wages with «market conditions»? As if that's just the price of labor, and nothing can be done. But in reality, they were simply using their power, knowing that workers had nowhere else to go.
Our ideas about the economy aren't just abstract theories. They form the basis for laws, policy decisions, and public opinion. When these ideas are wrong, the consequences are very real.
And here's what's encouraging: we are learning. Meta-analyses, systematic reviews, and corrections for publication bias are tools that help us clear our understanding of distortions. They show us that the world is more complex than simple models suggest, but it's not unknowable.
The minimum wage doesn't work like it does in the textbook – and that's good news. It means we can help workers without fearing mass unemployment. Informality won't disappear by simplifying bureaucracy – but that's also good to know, because it allows us to focus on what really works: stronger enforcement and creating conditions for productivity growth.
Labor markets aren't perfectly competitive, and employers have power – and acknowledging this fact opens the door for meaningful interventions that don't «distort the market» but rather correct existing distortions.
Belief and Evidence
Let's go back to the beginning. Money exists because we believe in it – that's my favorite idea. But economic theories also exist because we believe in them. And when belief diverges from reality, we need to change our beliefs, not deny reality.
For decades, economists believed in a certain picture of the world: competitive markets, rational decisions, simple cause-and-effect relationships. This belief shaped policy and affected millions of lives. And then the data showed that the picture was incomplete, and in some places, just plain wrong.
This isn't a failure of science – it's a triumph. Science works exactly this way: we propose hypotheses, test them, make mistakes, learn, and adjust. The problem arises when we cling to old ideas even when the facts tell us otherwise.
We are now at an amazing moment: the data we've accumulated allows us to see reality more clearly than ever before. Meta-analyses are clearing the picture of noise and distortion. Integrated models are revealing the connections between phenomena that once seemed isolated.
What's next? That depends on whether we can change not only our theories but also the institutions that produce knowledge. As long as scientific journals prefer «interesting» results over null findings, the distortion will persist. As long as politicians look for simple answers to complex questions, simplistic models will continue to circulate.
But there is hope. More and more researchers are becoming aware of publication bias and are working to correct it. More policymakers are turning to evidence instead of ideology. More people – perhaps even you, reading this – are starting to ask uncomfortable questions and demand honest answers.
The labor market doesn't obey the textbooks. But that doesn't mean it's unknowable. It means we need better textbooks – ones that describe the world as it is, not as we've grown accustomed to imagining it.
And when we understand this – when we accept that the minimum wage doesn't destroy jobs, that employers have power, that the shadow economy requires complex solutions, and that our data is skewed by systematic errors – then we can start building policies that actually work. Not for idealized models, but for the real people who face the reality of the labor market every day: looking for work, negotiating a salary, trying to make ends meet.
Because economics isn't about numbers in a report. It's about us: our hopes, fears, decisions, and mistakes. And the more honest we are with ourselves about how it works, the better we can change it.
