A New Measure of Government Efficiency: The Post Office

Rafael La Porta measures government efficiency by sending letters to fake addresses.

Most poor countries are badly governed. Economists and political scientists have long debated why this is the case. Some argue that governance is worse in poor countries because they tend to be non-democratic and less accountable to their citizens. Another view holds that the productivity of public services in developing countries is low for similar reasons as that of private firms: low education levels, poor infrastructure and lack of technology.  Still another reason is poor management, which could be the result of patronage in hiring and promotion, low public sector wages that fail to attract talent, and a dearth of incentives, supervision, and monitoring

This ambiguity has presented a challenge for government efficiency researchers. Moreover, their usual tool—surveys of citizens—has been ineffective in deciphering which individual factors are in play, since people often lump together a number of different issues together in responding to questions about government efficiency, and tend to mix personal experiences with policy views.

Now Noble Foundation Professor of Finance Rafael La Porta and colleagues have found a new way to measure efficiency through the use of a universal government service: the post office. By sending 10 letters to fake addresses in 159 countries and measuring how long it took for the letters to be returned to Hanover, N.H., La Porta was able to gather data on efficiency in a manner in which neither corruption nor politics would be a factor. He simply measured whether postal employees were doing their job.

“It’s an environment where corruption doesn’t matter since a postal employee in the Philippines cannot ask me for a bribe,” says La Porta. “Throwing out the letter also serves no political purpose, so we’re able to better understand the extent to which more capital investment or better management affect efficiency.”

In conducting the experiment, La Porta’s team sent two letters to non-existent business addresses  in the five largest cities in each country. Each envelope contained an address listing a common personal name in the country along with a generic business name – such as “Inventory Technology Partners.” The made-up businesses were located on made-up street addresses that used the names of Nobel laureates in economics and famous classical composers, and each envelope contained a single-page form business letter. The return address was listed at the Tuck School of Business along with bold letters stating “Please Return to Sender if Undeliverable.”

Since all of the countries in the experiment are members of the Universal Postal Union, their postal workers have the same obligations to return undeliverable mail within one month of it entering the country. In practice, only 59 percent of the letters were returned within a year, and only 35 percent were returned within three months. Countries that returned all of the letters sent to them included not only countries often associated with efficient government, such as Canada, Norway and Finland, but also El Salvador, Uruguay and Barbados. Many of the 42 countries that did not return any letters were in sub-Saharan Africa but the group also included Russia, Egypt and Cambodia. 

La Porta’s team then examined the data to find correlations with factors such as the number of post offices and postal employees per capita, the use of postal code databases, and whether a country uses the Latin alphabet. These measures of resources and technology explained slightly less than half of the variation in countries’ postal performance.

They then checked the return-to-sender results against measures of professionalism and meritocracy in a country’s civil service. La Porta and company also examined to what degree management skills may have had an effect, by comparing the data against measures of management quality in each country’s public and private sector. Both factors had a significant impact on the likelihood a letter would be returned to sender.

The results suggest corruption and politics may not play as large a role in government inefficiency as some have argued, since the corruption-immune postal data is strongly correlated with standard measures of government efficiency, such as number of days to adjudicate a court case.

“The overall story is that government is not that different from the private sector,” says La Porta. “The indication is that the postal service gets better with development not only because there are more resources but also because people learn how to better manage large organizations.”