China: Proposed Statistical Reforms

China is a vast country that is inherently difficult to quantify and measure. Beyond the socioeconomic flux it finds itself experiencing due to development, China has a highly varied geography and a huge and diverse population. Accurately accounting for such variety and rapid change would be challenging for any government, but China's sprawling bureaucracy also has an incentive to tailor its statistical data for political needs. In particular, the provincial governments, which manage their own statistics, regularly alter numbers to present themselves in a better light and meet central government demands. They downplay some problems, exaggerate others and almost always seek to post high growth figures. Bureaucrats that report better economic indicators tend to be promoted, which encourages the fudging of statistics. This is the first problem that National Bureau of Statistics (NBS) Director Ma Jiantang addressed on Jan. 28, pointing to 13,500 incidents of false statistics-compiling in 2009. Ma said that the current way of calculating gross domestic product (GDP), which entails provinces reporting their own statistics to the NBS for approval, leads to local meddling. Ma announced that the NBS is working toward creating a unified calculation scheme. In 2007, for instance, the provincial GDPs summed up to 27.5 trillion yuan ($4 trillion) while the NBS reported the national GDP as being 24.9 trillion yuan ($3.6 trillion). The huge discrepancy (2.6 trillion yuan, or $340 billion) would likely have been worse if not for the fact that the NBS reviewed the provincial statistics before publishing them. The second problem Ma addressed was China's practice of reporting major economic indicators only by showing year-on-year change. Ma said this would also be revised, and pointed out that showing economic change on a month-on-month or quarter-on-quarter basis provided a much more nuanced and timely picture of what was actually happening in the Chinese economy. Year-on-year change only reflects the data in relation to the same period of the previous year, rather than showing sequential change. This is especially problematic when there is volatility — as there often is — that can completely distort the picture. In other words, when one month's statistics are compared to those of the same period of the previous year, it may show a smoother line on a chart between the two points, but neglects the (often significant) variations that happened in between. (click here to enlarge image) Compare month-on-month and year-on-year statistics in regard to China's exports (see chart). From March 2009 through December 2009 China saw several (non-consecutive) months in which exports grew. But while the month-on-month calculations showed these months' positive changes, the year-on-year ones continued to depict export change in the negative range, simply because the total value of the exports still fell below the values during the same period the year before. In other words, the year-on-year picture was unable to convey the more immediate reality of intermittently rising exports. Because of this effect, year-on-year measurements do not reflect seasonal trends (such as rising exports for the Christmas season in Western markets), nor do they reflect breaks in trends (such as the point in March 2009 when exports surged for the first time since the recession began). All in all, the year-on-year picture is one that neglects the volatility of what was actually happening to exports in real time. With China's exports, both the month-on-month and year-on-year changes can be calculated from absolute values. In other categories, China leaves no alternative to the year-on-year picture. Such is the case with the Chinese consumer price index (CPI), the standard measure for price inflation; every month China presents year-on-year change in CPI only. The index uses the same month of the previous year as a base, but this means that the base is constantly changing, and hence there is no absolute value on which to independently calculate month-on-month change. It is not clear how the National Bureau of Statistics will reform these practices, but providing month-on-month changes in CPI would present a more realistic picture of the overall changes in prices across China's economy. (click here to enlarge image) Of course, Beijing's purpose is not to meet international standards and provide more transparency for outsiders. These are considerations only insofar as they may attract more investment and positive press. Rather, the point here is extending the central government's eye into the provinces, gaining more transparency within China and limiting the provincial governments' ability to massage the numbers. Beijing's tightening control over information about its disparate regional economies only incidentally aids external analysis of China's economy. Such statistical reforms — and others like it — could do wonders for the Chinese government's ability to paint a quick and accurate picture of what is happening on the ground, a necessary prerequisite if it is to even have a chance at crafting policies that address its deep economic imbalances. Implementing such statistical reforms successfully is not guaranteed; there is inertia in the existing system. And even successful statistical reforms will not change the fact that China fudges numbers. Controlling information is a critical component of Beijing's social control, which can be compromised only at the risk of overall destabilization. Rather, improving statistical reporting will merely give Beijing the prerogative to handle all the fudging itself, rather than getting misled by the provinces.