Return to Sender
What undeliverable mail can teach us about economic growth.
Photograph by Justin Sullivan/Getty Images.
The allegedly poor performance of “the government” is a staple of conservative rhetoric, while progressives retort that private sector success is typically built on a solid foundation of public services and infrastructure. Common sense, however, suggests simply that the quality of government services varies. The United States Navy is a fantastically high-performing agency. When I visited Stockholm, employees of the metro system did a great job of helping me understand how to use the city’s bikeshare system, even though they had to speak in a foreign language and I was clearly not eligible to vote for their bosses. But when a few years back I needed to get a replacement recycling bin from the D.C. Department of Public Works, my then-roommates and I were plunged into a Kafkaesque nightmare. Even though these differences in performance level are clearly real, there’s very little effort to measure them.
A paper posted last week to the National Bureau of Economic Research website aims to counter that with a simple test. The vast majority of countries have signed international agreements committing them to the principle that undeliverable letter should be returned to their senders. So the researchers deliberately mailed letters to fake businesses around the world to see which postal agencies return them, and how promptly. Many of the results are about what you’d expect—Germany is efficient and well-governed—but there are some surprising high-performance standouts, including dark horses Uruguay and Algeria.
The basic experiment was simple enough. Alberto Chong, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer mailed 10 letters to fake businesses in each of the 159 countries.
Each letter was correctly formatted and featured both a return address and a clear request to “please return to sender if undeliverable.” The letters all featured a real city and real (and appropriate) ZIP Code (or similar), paired with the name of a fake company and a nonexistent street address. The idea was to ensure that the letters would make it through the automated scan-and-sort portion of the delivery process and find their way into the hands of a human letter-carrier. The question then is what happens when a carrier is faced with an undeliverable piece of mail: Does she throw it out, or does she do what she’s supposed to do and initiate the process for returning it?
This is, the authors note, an interesting test of government performance in part because there’s no place for bribery. Shakedowns are an important aspect of public sector dysfunction in many parts of the world, but it’s not logistically possible to make return of undeliverable mail contingent on the payment of a bribe.
The experiment revealed amazingly wide variation in postal performance. Twenty-one countries returned all the letters within a 12-month span. Four of them—the United States, El Salvador, the Czech Republic, and Luxembourg—got it all done within 90 days. There were 16 countries, by contrast, that didn’t send any of them back within a year. Those are mostly African countries, but Cambodia, Tajikistan, and Russia also make the list of shame.
The meat of the analysis, however, is the number-crunching that followed. Controlling for other variables, it turns out that you’re more likely to get your letter returned from a country that uses the Latin alphabet. That’s about what you’d expect, considering that the envelopes were addressed using the Latin alphabet. Then again, international postal conventions commit all countries to delivering Latin-addressed mail so the fact that personnel appear not to be properly trained in its use is telling. In addition, various measures of “postal resources” are positively correlated with proper return of the letters. Countries that have more postal workers per capita, more post offices per square mile, and more detailed postal databases all do better. Money invested in postal systems, in other words, isn’t “wasted” in any narrow sense: More resources really do make more productive postal agencies.
These linguistic and technological factors statistically explain 41 to 46 percent of the observed variance in postal performance. The rest seems to be the secret sauce of management.
What’s interesting here is that good performance is strongly correlated with general indicators of economy-wide management skill. Countries with higher-ranked business schools, better-educated managers as revealed by census data, and high scores on surveys about managerial willingness to delegate and capacity to innovate all do well at delivering the mail. In other words, politics may matter less than you think, and “the government” isn’t that different from the private sector. Well-managed, high-productivity postal services exist in the same place as well-managed, high-productivity private enterprises. Given this, it is also not surprising that they found no difference in efficiency between countries that have government postal monopolies and those that have competitive mail delivery.
Researchers often puzzle over the fact that on the quality of government services almost invariably increases with per capita income. This, the authors point, is “surprising if one focuses on the uniqueness of government, but makes perfect sense once it is recognized that government is subject to the same dynamics as the private sector.” In other words, government agencies are people and tend to share the characteristics of the overall society in which they’re embedded. Countries well supplied with capital and skilled workers manage to do a decent job of delivering the mail. Those that can’t get letters sent where they belong tend to be countries where the private sector doesn’t work well either. The stellar performance of El Salvador’s postal service may be an indicator that this small, poor country devastated by recent civil war has some of the underlying qualities needed to make it much richer.