Every year, the World Bank makes corrections in the data that feeds it’s doing business rankings, some more noticeable than others. And every year it also changes its methodology. Yet somehow these are not put under spotlight.
In 2016, for instance, its raw data said that it takes 50 days to register a property and 594 hours to pay taxes. The revised data for 2016 reported next year said that it took 155 days to register property and 307 hours to file taxes. Unlike most Pakistani statistical organisations, the World Bank Doing Business team publishes their corrections as well (credits for which); the corrections for DB 2018 can be seen in the table, that cause corresponding changes in the Ease of Doing Business Scores or the Distance to Frontier Scores (DTF).
The DTF scores show the distance of each economy from the “frontier” economy, which represents the best performance observed on each of the indicators across all economies in the Doing Business (DB) sample since 2005. While rankings are indeed competitive and being ranked 80 is better than being ranked 150, it is not the end all be all.
Interactions with market intelligence companies reveal that investors looking to set up shop in a developing or emerging economy compare raw data of Doing Business and look at DTF scores, because it is possible that two countries ranked 50 and 90 have their DTF scores in the top percentiles. Therefore, what matters the most is that regulations in Pakistan are closest to the best performance, and that the raw data that the World Bank takes is correct.
To be fair, World Bank’s Doing Business publishes more than 24,000 indicators (nearly 120 indicators per economy) each year, using more than 117,000 data points, each of which is made available on the Doing Business website. Running such a huge exercise is always prone to mistakes; the bank eventually corrects those mistakes and flags it as well. For instance, it recently highlighted that “the correction rate between Doing Business 2018 and Doing Business 2019 is 5.6 percent”. However, it states that this correction rate reflects changes that exceed 5 percent up or down. BR Research estimates that even changes less than 5 percent could also be material depending on the distribution of DTF scores; therefore, the World Bank should report even those changes that are less than 5 percent.
Incorrect data, however, is not the only problem here. Changing goal posts is another one. In DB 2019 alone, there have been three changes in methodology; one each in ‘registering property’, ‘paying taxes’, and ‘enforcing contracts’ sub-indicators. While the World Bank publishes its change in methodology, and the corresponding changes in Doing Business ranks and DFT scores, the media doesn’t necessarily report those.
For example, in 2015, Pakistani media went hysterical over the country’s drop from 128 rank (in the year before) to 138 out of 189 countries. The Doing Business team had explained that based on new methodology, Pakistan was not ranked 128 the year before and was in fact ranked 136, and therefore the fall was of 2 ranks rather than 10. But who reads such explanations.
There are two key lessons to be had here. First, the World Bank needs to ensure lower correction rates, and ensure public dissemination of data corrections and impact of methodological changes via detailed press conference in countries where financial media isn’t strong enough to understand the nuances. The same lessons apply to relevant government organisations – such as the Board of Investment in Pakistan – to clarify the same, and flag errors in raw data. Second, while improvement in yardstick is appreciated, too frequent changes, as is the case with Doing Business, take away the very essence of a yardstick from such indicators. One ends up chasing a moving target!
Comments
Comments are closed.