Government regulations vary greatly in their effectiveness. In cases of ones that fail, a common theme is basing regulations on what turn out to be inaccurate models of how people respond to incentives. How can we avoid such errors when thinking about raising the minimum wage?

In the dynamic economy of the 21st century, it’s impossible to accurately predict the effects of raising the minimum wage. However, based on the lessons of yesteryear, imposing some intellectual discipline upon ourselves when we formulate our predictions will help us craft better policies.

Economists often recount the “cobra effect” when analyzing the effects of government regulations. In colonial India, in an effort to reduce the number of venomous cobras in Delhi, the government offered a bounty for dead serpents. Perhaps the idea came from drawing an analogy between cobras and criminals, where bounties have been helping authorities capture outlaws since antiquity; as such, this policy seemed entirely sensible.

But apparently, the policy backfired by motivating people to breed cobras in the pursuit of the bounty. When the government responded by suspending the bounty, the breeders released the now-worthless serpents into the city, leading to a record cobra population. In retrospect, the flaw was failing to realize that people can easily modify their conception of cobras, whereas they cannot do the same for criminals. This error was quite subtle, and I don’t blame policymakers for making it; I pay pest controllers to kill insects, and I’m confident that this hasn’t led to them breeding more insects. Crucially, the government diagnosed and corrected its error.

The cobra anecdote shouldn’t be taken as paradigmatic; many regulations work positively as planned, while others are clearly going to be disastrous but are implemented nonetheless. One of the key lessons, however, is that there are returns to thinking carefully before diving into new regulations.

In the case of raising the minimum wage, we need to demonstrate more intellectual discipline. The departure point for many proponents of the minimum wage is the assertion that if the labor market earnings of a worker fall sufficiently low, he or she will be unable to live a dignified life. Legislating a higher wage, therefore, directly rectifies this morally unacceptable state of affairs. In an effort to preemptively dismiss concerns about the impact upon employers, proponents may draw attention to the healthy profits earned by companies such as Apple and Nike to imply that they can afford to pay higher wages.

With all of that in mind, here are a few questions to ponder:

First, does the above give us any guidance on the level at which to set the minimum wage? In the United States, small- and medium-sized enterprises account for approximately 50 percent of total employment, a far cry from the “fat cat” image of corporate America. If you think that they won’t be affected, then why not set the minimum wage at $50 an hour? If you do concede that they would be hurt, then how do you intend to gauge the effects? Remember that the government represents all stakeholders, so it’s unacceptable for it to base policies purely on what one side stands to gain.

Second, is there any reason to think that people might respond to the incentives in an unexpected way, just like the cobra-hunters-come-breeders? Most people walk past impoverished panhandlers every day; if they were forced to donate $1 each time on moral grounds, would we be surprised to see people trying to avoid panhandlers, potentially leaving them with even less money than if the donation were voluntary? Analogously, many economists expect firms to look for ways to minimize hiring, such as automation and downsizing, in response to minimum wage hikes, which is to the detriment of the lowest earners.

These issues speak to a broader criticism directed toward casual proponents of the minimum wage, which is that they don’t really have a mental model of how wages are determined in the absence of minimum wages, and so they’re unable to analyze the consequences or reason why one level is superior to another. (My Mercatus Center colleague Don Boudreaux has written a lot about this.)

One of the best ways to sharpen our thinking about the minimum wage is to compare it to the alternative poverty reduction policies available and consider which is the best approach. For example, means-tested wage subsidies funded by general taxation potentially have the same result as a minimum wage for the workers, but that has a lesser effect on unemployment and on profits because the cost is spread out over the economy.

Ultimately, greater introspection as individuals will help us to collectively arrive at a consensus regarding minimum wages. If we are going to ask employers to pay higher compensation, then the least that we can do is force ourselves to think hard, so the cobra effect remains no more than a quirky anecdote.