In 2016, the Liberian government transferred the responsibility of running ninety-three government schools to eight different organizations (some NGOs and some for-profit operators) and, remarkably, ran an RCT to evaluate the impact. The results were mixed. Students’ results in those schools were somewhat better on average, but the private schools also spent a lot more money per pupil (double what the normal students got), so the playing field wasn’t quite level. Moreover, four of the eight organizations did little better than public schools. Bridge Academy, the star provider, had good scores, but only after receiving considerable outside money and dropping all students in excess of their class size cap.18 Another provider, the US charity More Than Me, got itself embroiled in an egregious sexual abuse scandal.19 There was no miracle cure.
THE CORRUPTION OBSESSION
Part of the root of the skepticism of government is a widely shared obsession with corruption in government across the world. Perhaps it is because the idea of government officials living the easy life on taxpayers’ money offends people, and therefore is often at the heart of political campaigns. The presumption is that if there was enough political will, corruption could be made to go away. There is of course a lot of truth to this. How can you expect corruption to go away when heads of governments are themselves up to their elbows in gravy?
But the view that all it takes to root out corruption is the will to do it misses the key point about the sources of corruption and our ability to control it. It is often precisely because governments do things the market will not touch that they become susceptible to corruption. Take the example of a fine for polluting. The polluter would gladly pay someone in the pollution control office a portion of the fine to make the evidence go away. But would things improve if a profit-maximizing private firm was collecting fines? Probably not, since they like money at least as much. Moreover, as the history of private tax collection (“tax farming”) tells us, incentivizing private agents to collect taxes (or fines) runs the risk they will extort those who are not liable as well.
Or consider a slot in the best public schools. It is very tempting for a school official to accept a payment to open a “side door” for a rich but unqualified student, and it is rumored to be a common practice in China’s top high schools. But this is not about government per se; it is about rationing. Whenever a good is rationed, the temptation to just pay one’s way in is very strong. This was made abundantly clear by the admission scandals that shook elite private universities such as Stanford and Yale in 2019; parents who were wealthy but not wealthy enough to pay the full “price” of back door entry for their offspring (say, a building for the university) worked with a consultant who offered a more affordable side door (e.g., a bribe to the sports coaches).
The broader point is that our social objectives often push us to not follow the market’s dictates. There is no pure market solution for collecting fines, and the reason why public schools have low fees and private universities do not charge the price the market would bear is because we want poor but talented kids to be able to get the best opportunities. But whenever anyone tries to get in the way of the market there will be a temptation to cheat. Since it is in the nature of the government’s job to interpose itself in front of the market, the fight against corruption in government will be an uphill and continuous battle, even with the best intentions.
Moreover, fighting corruption is by no means costless. In Italy, an umbrella government organization called Consip was set up in reaction to a succession of corruption scandals. Its job was to purchase supplies on behalf of government departments. What it purchased changed from time to time; as a result, sometimes government departments had to provision certain things on their own, while at other times they relied on Consip. When government departments had access to Consip, they used that option most of the time, but it ended up costing the government substantially more for exactly the same products, because usually there was a cheaper version of the product on the market. In other words, the departments could have bought what they needed more cheaply, but they chose not to exercise that option when Consip was available. As a result, on net Consip turned out to be a money loser. Trusting government officials to do what they had always done, without constraining them, would have been a better idea.20
Why did almost everyone use Consip when it was available, even though they knew they could get products cheaper elsewhere? Probably because they knew that this way they were protected from any accusation of corruption. There is nothing special about government officials in their desire to check all the boxes to avoid trouble. Doctors in the United States recommend too many tests to avoid malpractice suits, for example. And large companies that use a single mandated travel agent for all their staff travel almost surely lose money on most tickets, since the agent does not search for the best deal. But this limits the risk that employees make money on the side.
This illustrates a broader point. The current fashion in fighting corruption is transparency, the idea that the workings of governments should be available for scrutiny by outsiders like independent public auditors, the media, and the public. There is solid evidence that in many situations transparency helps. In particular, informing the ultimate beneficiaries about the difference between what they are entitled to and what they are getting is a powerful instrument for fighting corruption.21 However, as the Consip example makes clear, there is also a downside to transparency. Monitoring often relies on outsiders limited in their ability to understand the bigger picture or evaluate how well the overall social objectives are being served; the most they can do is to verify that due process is followed. In turn, this means bureaucrats tend to focus a lot on checking off the right boxes to avoid attracting attention. This creates a specific bias toward following the letter of the law, even when its spirit is somewhere entirely different.
Ultimately, the portrayal of bureaucrats and politicians as either bumbling idiots or corrupt sleazeballs, for which economists are probably partly responsible, is deeply damaging.
First, it prompts a knee-jerk reaction against all proposals to expand the government, even when government is clearly needed, like in the United States today. In our survey of US respondents, trust in bureaucrats is as low as trust in economists: only 26 percent of our respondents trust civil servants either “somewhat” or “a lot.”22 This probably explains why so few people think government can be part of the solution.
Second, it affects who wants to work for the government. Attracting qualified people is essential to a well-functioning government. But to a talented young person in the United States, a career in government, given its reputation, is unappealing. Neither of us has ever had an undergraduate about to receive their diploma tell us they were headed to a career in government. This kind of sorting can turn into a vicious cycle. If only the less able work in government, we get an ineffectual government no one of talent would want to join. In France, on the other hand, there is prestige attached to working for the government, and the best and brightest do so.
The image of the government also affects the honesty of those who want to work for it. A study in India replicated the Swiss experiment with bankers we discussed in chapter 4,23 where participants (in this case, college students) were asked to privately roll a die forty-two times and record what numbers they got each time; the reward was half an Indian rupee for a one, one rupee for a two, one and a half rupees for a three, and so on. Students were free to lie about the numbers they rolled, and roughly the same proportion as in Switzerland did. But, just as those who were reminded of their identity as bankers cheated more in Switzerland, in India students planning to work for the government cheated more.24 In contrast, when the study was again replicated in Denmark, which is justifiably proud of its social sector, researchers found the exact opposite as in India: those planning to join the government were much less likely to cheat.25
Third, if it is assumed most people in government are either venal or lazy (or both), it makes sense to try to remove all
decision-making power from them (and thereby banish all creativity and all creative people). This has a direct impact on what government officials can do. In a recent experiment in Pakistan, providing a bit more flexibility to the procurement officers of hospitals and schools by giving them some free cash to spend on basic supplies greatly improved their ability to negotiate low prices, leading to big savings for the government.26
Putting too many constraints on government officials and government contracts can discourage talent when it is the most needed. Despite the fact that the United States is the world leader in computing, none of the big tech firms chose to bid on contracts to set up the computer system supporting Obamacare. The reason was apparently that there were so many boxes to check off to be a government contractor that very few firms were willing to do it. The Federal Acquisition Regulation has eighteen hundred pages. So, in order to win a government contract, it is more important to be good at paperwork than to be able to do the job.27 In the development world, the contractors that systematically bid for and win the USAID contracts are known as “Beltway bandits.” It is very difficult for other organizations to get a piece of that action, even when they have relevant experience on the ground.
Finally, and perhaps most importantly, the mantra that government is corrupt and incompetent has produced the kind of jaded citizenry who can react to news of shameless corruption among its elected leaders with a shrug, from Washington, DC, to Jerusalem and Moscow. They basically have learned to expect nothing else, and stop paying attention. Perversely, the obsession with petty corruption is breeding room for venality on a grand scale.
AMERICA FIRST?
The United States seems to be at an impasse. Forty years of promising that good things are just around the corner have created an environment where too many people trust no one, least of all the government. The growing economic and political influence of the rich, the result of the pursuit of the elusive elixir of growth, has combined with anti-government sentiments the rich carefully have cultivated to head off any attempts to rein in their growing wealth. The government is chronically broke because it is politically impossible to raise taxes, and even the most socially minded of the young have become convinced that government is terminally uncool, and so head to private foundations if they do not actually give up and join an “impact” fund, or an unabashedly money-making venture. And yet the only possible way out involves a much expanded role for the government.
It is possible this is also the shape of the future in many other countries. While the rise was less spectacular than in the United States, inequality has also increased in France. Between 1983 and 2014, the average income of the richest 1 percent has risen by 100 percent and that of the richest 0.1 percent by 150 percent. Since GDP growth has been slow, standards of living for most people except the rich have tended to stagnate: over the same period, income increased only by 25 percent (less than 1 percent per year) for the remaining 99 percent.28 This has fueled growing mistrust of the elite and the rise of the xenophobic Rassemblement National. The recent round of tax reforms undertaken by the centrist Macron government has made tax less progressive: the flat tax was raised, the wealth tax is gone, and taxes on capital have been pared back. The official justification is that this is necessary to make France able to attract capital away from other countries. It may well be true, but it runs the risk of forcing other countries in Europe to cut taxes as well, prompting a race to the bottom. The American experience warns us this may be very hard to reverse. European countries need to cooperate to hold the line on their taxes.
Developing-country governments raise even less money than the United States. The median low-income country raises less than 15 percent of GDP in taxes as compared to nearly 50 percent in Europe (and 34 percent in the OECD on average). To some extent, the underdevelopment of the tax system is a consequence of the nature of the economy; a large part of the economy is taken up by tiny firms or remote farms whose income is hard to verify. But to a large extent the low level of taxation is a political choice. India and China offer an interesting contrast. Historically, most citizens in both countries had too little income for it to be worth taxing them. But as incomes grew, India kept raising the threshold above which people had to pay income taxes—on budget day, when new tax rates are announced, the raised threshold is often headline news. As a result, the share of the population that paid any income tax remained stable around 2–3 percent. In China, where the thresholds were not adjusted, the fraction of the population subject to the income tax went up from less than 0.1 percent in 1986 to about 20 percent in 2008. Income tax revenues in China boomed, from less than 0.1 percent of GDP to 2.5 percent in 2008, while in India they have stagnated at around 0.5 percent of GDP. More generally, tax revenues as a share of GDP have been stable at about 15 percent of GDP in India for many years now, while they are above 20 percent in China, giving China the option to invest more and/or carry out more social spending.29 The new Goods and Services Tax in India is supposed to help by making it harder to evade taxes, but being a more or less proportional tax on purchases, it has very little redistributive effect.
Moreover, very much like the United States, India has not been very successful in using taxation to limit the ballooning of top-income pre-tax inequality. According to the World Inequality Database, the share of the top 1 percent of income in India’s GDP increased from 7.3 percent in 1980 to more than 20 percent in 2015. In China, where there was a bit more effort, it still went up, but by less, from 6.4 percent to 13.9 percent.30
The interesting counterexample here is Latin America, for many years the example everyone used for growth with exploding inequality (which then turned into inequality with no growth), where the recent decades have seen a significant reduction in inequality. This was partly driven by rising commodity prices, but also in part by policy interventions, higher minimum wages, and large-scale redistribution in particular.31
The way redistribution was expanded in those countries is instructive. The political opposition to transfer programs in Latin America is couched in terms of the moral and psychological consequences of giveaways, much like the US conversation about welfare is dominated by the fear of abuse and laziness. From the beginning, Santiago Levy, an economics professor who played a very important role in setting up Progresa, the transfer program in Mexico that provided the blueprint for many others, was very conscious of the need to get buy-in from the right.32 The program emphasized a social quid pro quo. The transfers were quite explicitly conditional: the families had to take their children to the doctor and send them to school to get the money. A randomized controlled trial proved that those given access to the program had better child outcomes.33 Probably as a result, these programs have proved durable. For decades, successive governments have sometimes changed the name of the program (Progresa became Oportunidades and then Prospera) but not much else. In 2019, the new left-wing Mexican government seems to be on the way to replace the program with a similarly generous program with fewer strings attached.
In the meantime, the conditional cash transfer program (CCT) had been imitated all over the region and beyond (all the way to New York City). Originally, most of the programs adopted similar conditionalities, and often paired the programs with RCTs. These series of experiments had two impacts. First, they demonstrated nothing terrible happens when one gives cash to the poor. As we will see in the next chapter, they don’t drink it all and they don’t stop working. This was instrumental in shifting the public perception on redistribution all over the developing world. In the 2019 elections in India, both major parties, for the first time, made a cash transfer to the poor a central element of their platform. Second, as countries started to experiment with the model and try out variants of it, it became clear the poor don’t need as much handholding as the design of the original CCTs implied. There has been a complete turnaround in the public conversation on redistribution, and the Progresa experiment and its successors contributed a lot to it.
The battl
e against growing inequality has not been permanently won even in Latin America. The top tax rates remain low and top incomes are not systematically going down (since 2000, in the World Inequality Database, they are completely flat in Chile, rising in Colombia, bouncing all over the place in Brazil).34 But the experience of Progresa highlights the notion that careful program design will be key to breaking open the seeming impasse in the United States and similar issues that might come down the pike elsewhere.
Figuring this out may be one of the greatest challenges of our time. Much greater than space travel, perhaps even than curing cancer. After all, what is at stake is the whole idea of the good life as we have known it. We have the resources. What we lack are ideas that will help us jump the wall of disagreement and distrust that divides us. If we can engage the world seriously in this quest, and the best minds in the world to work with governments and NGOs and others to redesign our social programs for effectiveness and political viability, there is a chance history will remember our era with gratitude.
CHAPTER 9
CASH AND CARE
MANY VISITORS to the northern Indian city of Lucknow visit a gigantic eighteenth-century Indo-Islamic monument in the middle of the old city called the Bada Imambara. It is unusual among buildings from that period, being neither a fort nor a palace, nor a mosque nor a mausoleum. Guides tell many stories about it inflected, no doubt, to suit the tastes of the audience—Abhijit was told it was a part of the kingdom’s defense against the encroaching British Raj, despite not looking remotely like a fort. In fact, it was built by the king of Awadh, Asaf-ud-Daula, in 1784 to provide jobs to his starving subjects because crops had failed.
Good Economics for Hard Times Page 33