As someone with a relatively distant relationship with horses, I have occasionally seen some with blinders on and wondered exactly why, when I have never noticed any other animal with blinders on. So when my curiosity got the best of me, I did some checking.
It turns out that because horses are prey animals, they have eyes on the sides of their heads, to increase their ability to detect, and therefore escape, predators (whose eyes are forward-focused for targeting prey). But that also means horses can be easily distracted by things occurring on the periphery of their vision. Consequently, when people want horses to focus on a particular task, blinders can reduce the distractions and make them succeed better in doing it.
However, I have been struck by the fact that there is no parallel for humans. Putting blinders on people, as is often attempted by those trying to “sell” public policies to them, does not help people more successfully attain their goals. Such blinders do the opposite, increasing the likelihood that people will be preyed on. The reason is that blinders are put on horses to make them more productive in situations where the threat from predators has been eliminated or controlled, but when making public policy, individuals are surrounded by political predators.
We Must Consider All Aspects of a Policy Change
The key is that, as economists have long known but others often overlook, it is impossible for a policy change to alter just one incentive “story.” When one incentive changes, others must as well. Therefore, anytime a policy salesman presents an “analysis” as if there is only a single issue, or a few issues, when there are in fact far more, they are trying to put blinders on citizens, to make sure they don’t notice the cheats.
The best-known effort at blinder-proofing citizens from being turned into unwitting prey is Henry Hazlitt’s Economics in One Lesson. The lesson, stated in the opening chapter (followed up with chapter upon chapter of illustrations) is that “the art of economics consists in looking not merely at the immediate but at the longer effects of any act or policy; it consists in tracing the consequences of that policy not merely for one group but for all groups.” That two-part lesson merits serious attention.
Looking at the Long Run as Well as the Short Run
Hazlitt’s lesson that we must look at the often very different long run effects of a policy as well as the short run effects is illustrated by rent control. Because rental housing tends to be long-lived, rent control’s erosion of the rental housing stock does not appear very quickly, allowing proponents to allege the effect doesn’t exist (especially with all the other variables that changes could be attributed to), rather than “it doesn’t exist in a major way, yet.” However, the long-term effects of rent control have been likened to widespread bombing, guaranteeing that a solely short-term focus will dramatically distort accurate evaluation. Similarly, Keynesians’ determination to increase consumption in order to stimulate the economy has reduced saving, dramatically eroding investment, where the cumulative effect of depressing the growth of the capital stock now severely hampers our productive potential. As Hazlitt put it, “Today is already the [adversely affected] tomorrow which the bad economist yesterday urged us to ignore.”
Looking at all Groups, Not Just Some Groups
Hazlitt’s lesson that we must look not only look at the effects on one group, but for all groups, is illustrated by virtually every government expenditure program not financed by the beneficiaries (such as bridges whose costs are paid by users’ tolls). Since government has no resources of its own, expenditures require that revenues be raised from its citizens. One cannot just analyze the consequences of government spending, as if other things are equal, because they cannot be. One must also analyze how government acquired the resources, and the consequences that result. If resources come from taxes, those effects must be incorporated (including the costs to society from the distortions created in addition to the direct costs of the taxes). If resources come from government borrowing, which is essentially just a government commitment to increased future taxation, the present effects (crowding-out of private investment by the increased borrowing) and future effects (of the higher taxes necessary pay for the borrowing) must be incorporated in the analysis. The alternative is blinding oneself to the possibility of a reasoned evaluation.
Even the language used to sell policies can blind us to questions about effects on others. For example, there is a sense in which “we as Americans pay our Social Security taxes and we as Americans get the benefits” is true. However, early generations got far more in benefits than costs, forcing later generations to pay far more than they will get as benefits. Analyzing this with “we”” language hides the massive redistribution. Similarly, programs supported as helping groups like “the poor,” such as minimum wages and housing subsidies, actually impose very large costs on a substantial subset of “the poor” (e.g., those who lose their jobs or who don’t get funding for their housing but face higher rental prices raised by subsidies others get), which cannot be squared with helping “the poor” as a group. Even words like “need” focus attention solely on beneficiaries, where the implication that “therefore such persons shouldn’t have to pay for what they get” precludes most people from asking how person A’s need justifies government taking person B’s resources to fund it without their consent. Patriotism as a rationale for protectionism similarly frames the issue as if only American and foreign producers are affected. American consumers are ignored. But protectionism is our government colluding with our producers to raise prices, hurting all our consumers. And patriotism cannot justify some Americans ripping off other Americans.
Incentives Matter
Economics in One Lesson offers many more illustrations than those provided here, making it a “must read” for anyone who wants to understand policy rather than parrot misrepresentations. However, I believe that a third part should also be added to his two-part lesson. The art of economics must also recognize and accurately evaluate all the margins at which supposed beneficiaries’ incentives are changed. That is, those people policies are intended to benefit typically have multiple incentives changed, not just one, and the effects of all of them need to be recognized.
The food stamp program (now the Supplemental Nutrition Assistance Program, or SNAP) offers an instructive example. It was intended to increase food consumption of low-income people. Many supporters have portrayed that as the sole area of concern and treated it as obvious that subsidizing their food purchases will substantially increase their food consumption.
However, food stamps have much smaller effects on food consumption than the value of food stamps given out. The major reason is that, since almost all recipients get less in food stamps than they would have spent on food anyway, the food stamps can be substituted for dollars spent on food, freeing up income to be spent however recipients choose. In addition, the taxes to fund the program reduce other people’s food purchases, leading to very small effects on the agricultural sector, another major reason cited for supporting SNAP benefits.
In addition, since benefits are reduced as incomes rise, the food stamp program acts like an income tax, reducing incentives for recipients to become more productive. For those in other assistance programs that also reduce benefits as incomes rise (e.g., public housing or rental assistance), those disincentive effects are magnified.
Further, over the years, asset tests have been applied which have kept people from putting together sufficient resources to go to school and begin learning their way out of poverty or to start a small business and begin earning their way out of poverty. Their incentives have moved recipients’ income into unreported area (leading “poor” people to be much better off than official data suggests), and their assets into similarly concealable forms (e.g., cash kept at home), but which are far less productive than traditional investments. Program rules have even been so stringent that at times recipients could not own cars valuable enough to be reliable. Some people who receive unemployment compensation also qualify for food stamps during that unemployment, increasing their replacement rates and decreasing their incentives to find work.
Using blinders to ignore a large number of relevant issues over-sells SNAP’s effectiveness by precluding effective analysis. And similar ignored effects are common for government programs.
Blinders are for Horses, not People
Putting blinders on horses can make them more successful at their intended tasks. In contrast, putting blinders on people makes them less successful in achieving what they want, but gives those who wish to predate on them via government more of what they want. So when you see a political predator coming at you, bearing blinders, do what horses without blinders would do—run for your life, liberty and happiness.
