Winners, Losers and the Beauty of Impact Analysis

Imagine an average developing country whose government decides to pay for the bus tickets of all its high-school students. If it can afford it, what would be wrong with that? It would make it easier for teenagers, especially those of modest means, to attend school and graduate. It would also make them -- and their parents -- much happier. That would surely be the outcome of the new policy, and many people would support it. But, wait. What would be the impact? That is, what would happen now that would not have happened if the policy had not been adopted? Well, very likely, most teenagers would now consume more beer -- yes, the new subsidy would free them from using their own bus money to commute and, you know, they would use it for what they like best -- and their parents least. You get the point: outcome is not the same as impact. You may support the former, but hate the latter.

It gets better. A mildly corrupt president has enough public funds to build a school or to buy jewelry for his new wife, but not both. Just when he was grudgingly going to do the right thing, along comes a well-intentioned international development bank that wants to promote education by making loans to build schools. The president, full of smiles, signs the loan in front of the cameras and the school is built. Outcome of the bank's intervention: more classrooms. Impact: more bling for the First Lady. By financing the school, the bank actually funded corruption.

You see, you can analyze the impact of almost everything governments try to do -- from bailing out their banking systems to creating jobs to protecting the environment. And, handy in these days of austerity, you can do the same for whatever governments stop doing -- say, paying generous pension benefits. In fact, the technology for assessing the impact of economic policy has been around for years -- economists love it, if only to prove how simple-minded some politicians can be. But it was rarely used outside rich countries. Now it is. What caused the change? Data.

To detect impact, we need to know what governments, people and enterprises do with their money -- we need to know their budgets. In advanced countries that is not a problem -- if you live in an OECD economy, chances are that consumer research companies already know more about your spending habits than you do. And in those economies, governments publish their fiscal budgets and enterprises file their tax returns. The novelty is that the same is beginning to happen in the developing world. Recent improvements in household and industrial surveys are giving us a better, more comprehensive and more frequent reading of how firms and people -- including the poor -- allocate their resources. Credit-rating agencies and watch-dog NGOs are doing the equivalent to governments by pushing them to open up their books. Today, the fiscal accounts of Chile, Brazil, Mexico, Slovenia, South Africa, Sri Lanka and many others are posted on the web. They are far from perfect and not always easy to understand. But the trend is clear.

So, armed with numbers, impact analysis is becoming a game-changer in the development profession. Not only is it telling us what government action really does -- as compared to what it means to do -- it also tells us who will win and who will lose with it. Reforms become easier once you know who will support them, who will oppose them, and who should be compensated. Take the case of subsidies that keep gasoline prices low. The poor use public transport, and doing away with subsidies on gasoline will make it harder for them to get to work. Since they need to work, they may cut back on, say, preventive healthcare for their children (how about that for an impact?). You will be forced to compensate them. But it is the rich who drive the big cars. When you pay to make gasoline cheaper for the well-off, you are in effect paying for the wine, gym membership, cable TV or whatever else they do with the cash they do not have to leave at the petrol station. [You think this is farfetched? There is a Latin-American country that spends more on subsidizing gasoline for the wealthiest twenty percent of its population, than on all its social programs put together.] Now you have a map of where the resistance to reform stems from, and can figure out how to overcome it. You can also use the map to redress gross injustices -- to bring equity back on the agenda, if you will.

Whether reforms follow or not, the uncovering the true meaning of what our leaders do, has irresistible appeal in-and-of itself. No wonder impact analysis has become almost an academic obsession to the new generation of economists -- especially those that want to change the world (which is most of them). Just the World Bank alone is investing $20 million of European donors' money in finding the impact that its projects will actually have on poverty -- before they are launched. Other institutions are doing the same. The point is that the technical reasons that kept us blind to the real effects of policies and programs exist no longer.

But, will impact analysis change the way policies are made in practice? Will evidence rule politics? Slowly, it will. As democracy takes hold, education spreads, and information flows faster, farther and cheaper, even poor countries will see change in the political discourse -- some, in Africa, are seeing it already. Of course, catchy slogans will have to be spun to communicate dry statistics to voters. That's not trivial, but it's certainly doable. We will all be better off for it.