Geek Squad

Why did Nate Silver get it right in 2012 when so many others were so wrong?

Conventional wisdom held that in last year’s U.S. presidential campaign, a leader saddled with a weak economy running against a smart, accomplished businessman backed by hundreds of millions of dollars did not stand a chance. But data-driven geeks like Silver nailed the election forecast while many old-fashioned pundits flopped.

One reason President Barack Obama won a convincing victory that surprised so many was his campaign’s quiet, methodical use of a combination of tools from the world of behavioral science: highly motivated volunteers armed with sophisticated technology and field-tested messages to deliver voters. Call it the rise of evidence-based campaigning.

This scientific approach to winning elections is just the tip of what should now become a rapidly expanding iceberg. To start fixing some major global problems — from balanced budgets to climate change — both in the United States and around the world, we should be using the same methods that worked in the campaign. It’s time to bring in the geeks — and apply behavioral science to find solutions to persistent problems.

Economists have long had a strong influence on government policy, while other fields of social science are rarely paid much heed. The problem is that economists tend to have one solution for everything: If you want to discourage something, raise the price; if you want to encourage it, lower the price (or offer a subsidy). But prices, while important, are only one aspect of policy design. A better approach would take into account everything from understanding how lazy we really are when it comes to filling out forms to why we cheat or skimp on our taxes.

Consider the goal of helping people save for retirement, a problem governments all around the world are facing because of the underfunding of both public and private pensions. In creating special savings vehicles such as 401(k)s and IRAs in the United States, it has long been thought that the tax shelter such funds offer is crucial to their success. However, the design of such plans, what Harvard University law professor and former White House advisor Cass Sunstein and I have called "choice architecture," also matters. Setting up a plan so that workers are automatically enrolled (with the ability to opt out) and giving workers the opportunity to automatically increase their contributions when they get a raise (a plan called "Save More Tomorrow") have proved to be highly effective methods of increasing savings rates. And a recent study by Harvard economist Raj Chetty and his colleagues, using Danish data, has shown that these automatic features are actually more important than the tax subsidy in increasing savings rates.

And let’s forget the debate over raising or lowering taxes for a second. What about actually collecting them efficiently? The team of behavioral scientists created by British Prime Minister David Cameron (to whom I am an unpaid advisor) has shown that we can do a better job of collecting money from those who are behind on their taxes simply by improving the wording of the letter that is sent to taxpayers in arrears. Through randomized controlled trials, the British government has been able to show that a well-crafted letter reminding people that most of their neighbors pay their taxes on time increases the response rate of those procrastinators who owe money by 15 percentage points. These methods more than pay for themselves in terms of increased tax revenues and, when eventually rolled out across Britain, could produce more than $50 million in extra revenue annually, essentially at no cost. Isn’t that an approach that all governments should consider?

Another way to increase tax revenue without increasing rates would be to give more of the roughly 8 million immigrants who are employed illegally in the United States a way to obtain legal work permits. Today, most of these workers are paid off the books. But payroll taxes often go unpaid for home workers — nannies, gardeners, painters — even if the employees are citizens. While some of this tax avoidance is a conscious effort to save money, many would gladly pay the taxes if the paperwork were not such a damn nuisance.

The solution is to apply the single most useful bit of psychology one can ever learn: If you want to encourage people to do something, make it easy — or even better, automatic. Imagine, instead of the annoying reams of forms it now takes to set up a legal payroll system, that you could go online, enter an employee’s Social Security number and bank account information, and then let a program fill out all the paperwork and pay the relevant taxes. Some private companies offer these services, but governments looking for all the revenue they can get their hands on should offer an app that makes it easy.

Other countries have already found innovative ways to chip away at the gray economy. In China, for example, restaurant bills are often paid in cash, and taxes are avoided. To encourage customers to ask for a receipt, which ensures that the transaction is legally recorded, the government has been experimenting with giving restaurants receipts that double as government-issued lottery tickets — a ploy that has had significant success in the Philippines and South Korea. Because customers want the lottery ticket, they ask for their receipt. Voilà! Customers are happy, and the government gets a 17 percent increase in sales tax revenue.

There are other simple ways to encourage good behavior. Take the environment. When you see reports of the entire Arctic ice cap melting last summer, the problem can seem way too big for a single person to impact, regardless of how often we bike to work or how many newspapers we recycle. But that’s not true, especially if we find ways to make it easy. Not all of us can afford to buy a Prius, but have you mastered your programmable thermostat at home? Personally, I find these devices to be instruments of torture. While newer, user-friendly models are now coming to market, millions of homes already have programmable thermostats that are going unused. (Recent research published in the journal Building and Environment found that 90 percent of such thermostats are not programmed properly.) Now picture an army of high school geeks, paid a few bucks by the government and supervised by their science teachers, reprogramming thermostats during the summer. Not only would the kids earn some money for college, but we would all save some energy.

When it comes to efficiency, wasteful government spending is always at the top of the blame chart. And the Pentagon’s $600 billion annual budget is perhaps the best place to start to look for some savings. But taking the ax to the military requires a strong political stomach. What if there were another way? One of the earliest findings in behavioral economics is what we call the "endowment effect." The idea is that once we have some object in our possession, we are highly reluctant to give it up, even if we would have little interest in actually buying the thing again if we lost it. It could be your car, your house, or your husband.

The same principle applies to many things that the government owns. One example is military bases. If we could do it all over again, would we pick the spots we now have for those facilities, some of which are located on land that would be worth hundreds of millions of dollars? Consider California’s Camp Pendleton: It occupies 17 miles of prime Pacific coastline, and some 125,000 acres inland, between San Clemente and northern San Diego County. Do tanks really need clear skies, 70-degree year-round temperatures, and Pacific Ocean views to practice their maneuvers?

In the twilight of the Cold War, the United States decommissioned dozens of military bases using a procedure that wisely avoided getting Congress involved. But the method had one noteworthy flaw: The market value of the land was not among the criteria used to decide which bases to close. Even when the land was sold off, the price the government could get was not a factor. And today, if a branch of the military decides it no longer needs a base and sells the land, none of the proceeds go back to that branch of the military, or even to the military itself.

Let’s fix this. If it were up to me, I’d cut a significant chunk of the Pentagon’s budget over the next 10 years, but tell the Defense Department’s beleaguered accountants that they can recoup some of those cuts by selling off some of their pricey properties and using that money to restore the most essential parts of the budget that would otherwise be cut.

These are the sort of win-win ideas that the Obama administration and other governments around the world can start to apply to the problems of government — policies that take us away from partisan bickering to common-sense solutions. But to do that, we need more than politicians thinking they run the world. We need an openness to new thinking and to the findings of behavioral science. We need geeks.