One of the most distortionary things in politics is to frame issues in terms of who wins and who loses. Almost any major policy debate you can think of invokes images of people who lose and people who win. It’s natural to see why framing things in these terms is so appealing:
- Consolidating the debate to winners and losers allows us to economize on time spent debating, in much the same way that stereotypes allow us to economize on scarce information to make an evaluation. It’s much easier (intellectually lazy?) to decide whether you like one group over another than it is do sound economic analysis on policy. Humans are social animals and we prefer making evaluations based on social standing over logic or cost-benefit reasoning.
- Not all groups exhibit the same propensity to vote, so by framing issues in terms of winners and losers, politicians can pander to groups who vote en masse, potentially at the expense of other well-represented groups. Also related here is simply that some groups are much larger in number than others and hence capture more votes.
But there are serious problems with framing things in this way.
The first is simply that economic exchange is not a zero-sum game; most exchanges in a modern society are a positive-sum game. Evolutionary psychology creates a strong tendency for humans to believe that games are zero sum, since for most of our evolutionary history they were (and in many cases were actually negative-sum). But with the advent of economic growth driven by specialization and trade, wealth and well-being can be created out of thin air. The millions of economic transactions occurring daily produce an enormous number of small net-positive exchanges which create wealth. Those who create wealth - whether on a small or large scale - largely do so not at the expense of others. In the real world, often it’s winners vs. winners, not winners vs. losers. People simply win at different magnitudes.
Secondly, identifying who wins and who loses is different than identifying the magnitude by which the winners win and the magnitude by which the losers lose. In committing this error, most policy discussions simply boil down to which group we feel more compassionate for as a victim, even if the degree by which some group “loses” in aggregate is far more than the degree by which the other group “wins”.
Third, identification of who is the winner or loser may actually be incorrect (or, more ambiguous than is commonly thought). To give an extreme example, let’s imagine a law was passed tomorrow granting a minimum wage of $20/hr. Most would believe the winners are the poor and the losers are companies employing minimum-wage workers. Let’s leave aside identification of the loser there to the next point, but it’s by no means a foregone conclusion that the poor “win”, as such an enforcement of wage above where many workers’ productivity lie is likely to cause higher unemployment (this is basic supply/demand/elasticity economics). So even within the proclaimed group of winners, some people lose big time by loosing their jobs, but many others win by getting higher pay. Since the laws of economics are more powerful than the laws of politicians, the real minimum wage is always $0. Someone who makes $0 (job lost) is clearly worse off than someone who happens to keep their job and make $20/hr. So who wins on net? It’s fairly ambiguous and depends on 1) the nature of each employee’s utility function, and 2) the percentage of jobs lost. To be clear, I’m not debating the validity of minimum wage as a policy tool here; I’m merely pointing out that even in such a simple example, identification of winners and losers is often fraught with error or at best uncertainty.
Fourth, even if the identification is correct, it may be incomplete: there may be others winners or losers that haven’t been identified. Most instances of this involve the well-known tendency for humans to focus more on benefits or costs that are high in magnitude but low in number than those that are low in magnitude but high in number. This mis-identification of costs and benefits occurs almost any time the government is asked to intervene to grant privileges onto some special class of people, company, or industry, where there are usually more (but more diffuse) losers than are commonly thought.
Back to the minimum wage example, although Wal-Mart and other companies certainly do lose, they aren’t the only losers - and I would argue not anywhere near the biggest loser. Firstly, shareholders of such companies ultimately lose, since corporations are a legal abstraction of shareholders, many of whom are not considered to be in the same group of people as corporate management so abhorred. Even more importantly, consumers across the board lose a result of higher prices they pay for merchandise, especially goods and services that are very labor-intensive. What’s ironic here is that in many cases, consumers purchasing goods/services produced by minimum wage workers are more likely to be poor themselves, imposing a more disproportionate cost on low-income consumers - the very people advocates claim to be helping.
Another example is trade restrictions favoring some particular industry. If we slap tariffs on food imports, lots of American farmers may win (and the presumed loser is foreign food produces, which is true). But in reality the biggest loser by far - loosing by far more than the American farmers do - are American consumers, having to pay individually slightly higher prices for food which in aggregate is an enormous amount of money. But these consumers are diffuse; they incur costs that are low in magnitude but extremely large in number. The political incentives of such groups who may be small in number but suffer large gains or losses individually to organize them also make them more vocal, which is perhaps why they are not identified as losers to begin with.
Fourth, framing a debate in terms of winners and losers tends to fallaciously see policy affecting those groups as being a static system, whereas it’s actually quite dynamic: policy decisions often affects the behavior of individuals in ways unaccounted for, altering a) the very relationship in the system they seek to change, b) incentives in the economy as a whole, outside the winning/loosing groups. This is perhaps the most important insight from economics that policy makers and voters routinely misunderstand - comparing initial policy intentions with actual results reveals the law of unintended consequences quite well. This is why economists know to evaluate policy in terms of the incentives created or destroyed (and post-hoc, the actual results that took place), not on the proclamations of those making legislation, or the intentions embodied therein.
One example of (a) is tax policy, whereby increasing marginal tax rates on one group of people but not others is often thought to be simply a matter of “math” (just multiply the marginal tax delta by income and that’s your increase in tax revenue). But the very change of marginal tax rates creates behavior changes that most assuredly increases revenue by less than you’d expect by straight “math”. This effect can in rare cases be so extreme, it actually swamps out the straight math calculated revenue increase entirely, as people work less or are incentivized to shift income in such ways to avoid taxation beyond the simple marginal increase. Though estimating these behavior changes may be difficult, and their magnitude debatable, it’s important to understand that they do exist and do distort the relationship.
I encourage you to think about these issues next time a politician or someone else frames a policy debate in terms of winners and losers; reality is much more nuanced than many would have you believe. An understanding of economics, statistics, and cognitive psychology goes a long way to seeing public policy issues clearly. We must always be aware of our own biases and natural tendency to commit cognitive errors, for our natural tendencies and inclinations are often dead wrong; economics is often very non-intuitive. Failing to do so produces bad policy and impoverishes our well-being more than would be necessary.