Things that are anti-inductive confound our attempts to understand them. It's not just that these things are hard, like proving Fermat's Last Theorem or figuring out general relativity. At least those things, once solved, stay solved. If physics were anti-inductive, then the universe would adjust to us stoopid huminz figuring out general relativity by changing the rules as soon as they are discovered.
Social reality is anti-inductive. As soon as we figure out that the stock market goes up every Friday and dips every Monday, everyone rushes to sell Friday and then buy their shares back every Monday. This bids down the price of stocks on Friday and raises them again on Monday, defeating the theory that we just discovered. This pattern is generally true when the reality you're trying to theorize about is composed of intelligent creatures that can use the theory to plan their behavior (or who know that other intelligent creatures are planning on using that theory to predict their behavior).
People adjust their behavior to thwart superficially sensible government policies. Why don't we just find people who have low incomes and directly give them money? Well, because that creates a subsidy for being underpaid and underemployed. If we reward people for being "needy", we incentivize people to compete on needs rather than on productivity. People who go to college earn more than people who don't, so why not just send everyone to college for free? Well, because stamping "Graduate" on a bunch of people's foreheads degrades the value of existing college degrees without necessarily making the new graduates any more productive. If a college education will pay off for an individual student, that student will be able to finance the expense with their future earnings (by borrowing now and paying back the loan later). Money helps people buy things, so why not just print off a bunch and give it away? Well, because with more money chasing an unchanged amount of goods, producers and sellers will raise their prices. At best this will be a wash, and at worst you become Zimbabwe. Prices are rising, so why not just fix them in place to keep everything affordable? Well, because if prices are rising, that probably means the actual underlying costs of the inputs are rising, too. By preventing the price from rising, you are basically telling producers and sellers, "You must sell at a loss." Which is in effect telling them, "You may not sell at all."
Some of these are more controversial than others. I think nobody doubts that printing lots of money causes inflation, though some people nit-pick about whether price controls have the effects that economists predict. (They most certainly do, although the magnitudes may be difficult to measure.) But this post isn't just about bad government policy.
Should you bail out your alcoholic friend who just got a DUI and let him sleep on your couch for a week? For the third time? For the tenth? And maybe it's not for a week this time but a whole month? You may begin to suspect that such a person is using his friends to support a very bad habit and some tough love might be in order. Should you loan this person money for a needed liver transplant? Or should you think, "This person 'spent' his liver with his eyes wide open."
Should you loan your friend money because the utility company is going to shut her lights off? Maybe. But you might be thinking, "What brought us up to this point? Wasn't there something you could have done to avoid this? If you had asked me three months in advance, might I have said, 'Please take the following steps to avoid financial ruin, or I won't feel obligated to help you.'?"
It's really, really hard to help people. Some people know just how much of an "other people's help" budget they will get, and they spend the whole thing. Increase the budget, and they simply spend more. Economists call this moral hazard, the tendency to behave recklessly when someone is promising to rescue you from a wreck. This is why most interventions don't work. You can try to get the incentives right. You can give away only boring stuff, like bunk beds in austere public housing and tasteless-but-nutritious food rather than cash, such that people don't try to exploit your generosity. You can make giving conditional on some kind of buy-in by the recipient. Think of the sweat equity that Habitat for Humanity demands of its partner families. Some kinds of interventions work better than others, for sure. But if you put free stuff or free help out there, people will try to get it. It may mean deliberately earning less and being a less productive citizen in order to stay below some cutoff point. Or it may mean completely and deliberately crippling your life until someone takes pity on you and helps. We don't have a good theory telling us when people will give up in the "competing on needs" contest and return to self-help. Some will go all the way to the brink and even end up killing themselves; some will realize their bluff has been called and pick themselves up. Some perfectly deserving individuals will scoff at any offer of charity and insist entirely on self-help. Any intervention that promises to help people should count on a fair mix all these behaviors and try to minimize self-sabotaging "competing on needs."
These problems are anti-inductive because people are smart. A government program or private charitable plan to give away $1 billion could easily induce more than $1 billion in social costs due to destructive competition. We don't just get "reality as it stands plus $1 billion worth of mitigation." We get "reality as it stands plus $1 billion worth of mitigation plus sympathy-inducing self-sabotage directed toward collecting the $1 billion bounty." We need to be far more careful than we have been. We need to stop having knee-jerk "Here's-a-huge-problem-OMG-throw-money-at-it!" responses to social problems.
No comments:
Post a Comment