What Happened to the GDP Data of Turkey?

What Happened to the GDP Data of Turkey?

The National Accounts of Turkey have been recently revised by the Turkish Statistical Institute, resulting in remarkable updates on the Gross Domestic Product (GDP) series. Based on this revision, the Turkish economy seems to have painted a rosy picture in the post-crisis period until 2015. While the real GDP growth rate has become 7.4 percent on average between 2010 and 2015, the GDP per capita has been adjusted from $9,261 to $11,014 for the year 2015. The graph below shows the changes of interest.

In this sense, the surprisingly sizable updates on the GDP data of Turkey have caused debates both in the country and the world. This article attempts to explore the story behind the new national accounts of Turkey and its implications.


Source: TurkStat

For one thing, it is an international practice to periodically revise national accounts in order to have healthier measurements in terms of temporal dynamics. This is exactly why the Turkish Statistical Institute (TurkStat) hit the road in 2013, when it planned for the adoption of the European System of Accounts (ESA 2010). Accompanied also by some accredited international collaborators in early stages, the institute pursued the goal of raising the quality of the accounts to meet global standards. In this context, a main methodological change observed in the new series appears to be the “chain linked volumes” applied to the accounts within the ESA 2010 framework. This method computes the production volume for each year based on the prices of the previous year, instead of the “fixed year” technique used in the old accounts. So, TurkStat no longer employs the prices of 1998 for the calculation of real GDP, but takes advantage of a lagged chain methodology that is assumed to reflect the economy better. Moreover, some revisions other than the calculation technique also seem to play visible roles in the new series. These are mainly the changes in the definitions of the GDP components, shaped by the ESA 2010. One conspicuous example is the inclusion of the R&D expenditures into the group of gross fixed capital formation.

On the other hand, while the aforementioned revisions are known to depend on the ESA 2010 practice of the EU, the resulting gap in the Turkish accounts seems to require a further explanation. This is because the methodological changes outlined above justify only a part of the differences emerged between the old and the new series, particularly in the years after 2010. At this point, one integral modification applied to the system helps explain the surprising updates on the data to a great extent: Improvements made in data sources.

Here is how: Before the recent update, TurkStat was taking advantage of short-term indicators and surveys to elicit data for the calculation of GDP. So, what has shaken the numbers in the new accounts seems to be the integration of the “administrative records” into the system, allowing for a more comprehensive measurement. In this regard, data sources of the national accounts in Turkey now expand to certain major public bodies, including The Revenue Administration, The Social Security Institution as well as The Banking Regulation and Supervision Agency. In fact, this adjustment also helps understand why the bounce in the national accounts of Turkey is considerably higher than the one experienced by the EU upon the introduction of the ESA 2010 in 2014: Whereas Turkey failed to incorporate the administrative sources of interest into the accounts until recently, the EU had largely achieved it long ago.

Having briefly underlined the fundamental dimensions of the recent update, it is now important to understand the dynamics of the amended performance in the Turkish GDP. In this regard, an analysis of the GDP aggregates reveals that a striking component causing the gap between the old and the new figures is the gross fixed capital formation. In other words, investments happen to have contributed to the Turkish GDP considerably more than previously estimated. While one factor behind this finding is the treatment of the R&D expenditures, a notable divergence between the two series in the post-crisis period is mainly explained by the investments in construction, which in fact makes sense considering the huge growth of the sector in the country lately.

The resulting revision enlarges the slice of investments in the GDP of Turkey, while downsizing the share of household consumption. At this point, it is worth to note that such an updated picture looks more compatible within the group of the high middle-income economies, where Turkey belongs to. In terms of sectors, on the other hand, the increase in the share of construction is followed by the manufacturing industry.

What these details basically tell is that the change in the national accounts of Turkey not only arises from the introduction of the ESA 2010, but mostly represents certain improvements in data sources. In this regard, while the integration of the administrative records to the system is a milestone, the striking difference witnessed upon the update refers to some inaccurate measurements in the past, underlining the significance of data quality. Correct measurement of GDP matters not only to evaluate the welfare in a country but also to formulate appropriate policies. Therefore, now that the GDP data have ended up with notable changes, the coherence degree of certain past economic policies in the country has somehow become an enigma.



Hatice Karahan received her PhD degree in Economics from Syracuse University, where she conducted research for the Center for Policy Research. She is an associate professor of macroeconomics and currently heads the Department of Economics and Finance at Istanbul Medipol University. Besides, she serves as a consultant for leading business associations in Turkey and writes columns on economics for prominent Turkish newspapers and magazines on a regular basis.