The State Statistics Service’s Review of “What is Wrong with the Calculation of GDP in Ukraine?”

The State Statistics Service’s respond to Tetyana Tyshchuk’s article “Schrödinger’s GDP: What is Wrong with the Calculation of GDP in Ukraine?”

In response to your request of 15.03.2017 regarding the article “Schrödinger’s GDP: What is Wrong with the GDP Statistics in Ukraine” by T. Tyshchuk published on 13.03.2017, we would like to inform you about the following:

Gross domestic product (GDP) is the key integrated indicator of the system of national accounts developed in line with the methodology based on the up-to-date international standards. The GDP and its components, in particular, gross value added, are calculated based on the data for the year of 1990.

The latest revision of the national methodology in line with the United Nations System of National Accounts of 2008 and the European system of national and regional accounts of 2010 was performed in 2014, at the same time as in the European Union countries.

The system of national accounts is harmonized at the level of methodology and interrelated indicators with the balance of payments statistics and monetary statistics which fall within the responsibility of the National Bank, as well as with the public finance statistics prepared by the Ministry of Finance.

Harmonization of indicators of the above-mentioned areas of statistics, use of three methods of GDP evaluation (production approach, income approach, and assessment by end use categories) with further balancing of commodity flows in the input-output tables and the use of statistical methods of data analysis ensure high quality of statistical information.”

The methodology of GDP calculation used by the State Statistics Service (SSS) meets the requirements of the IMF Special Data Dissemination Standard which the SSS subscribed to in 2003.  
Value added at factor cost (VAFC) is an indicator of structural statistics which is new for the Ukrainian statistics and was first published in 2014 over the period from 2012. The methodology of its calculation is based on the requirements of the EU structural statistics regulations and has been developed with the participation of the European experts.

Given the fact that compliance of the Ukrainian National Accounts System with the international standards has been approved by the conclusions of the monitoring missions of the IMF Statistics Department in charge of national accounts statistics, to clarify the necessity of revising the VAFC calculation method, considering the above-mentioned facts, the SSS is taking measures to assess this methodology with the participation of experts of the statistical office of the European Union.

Moreover, to thoroughly analyze the VAFC methodology, the issue of more detailed data collection of different elements of this indicator is also being worked out.  

Considering the above-mentioned, the State Statistics Service will further inform you about the recommendations to be received in the follow-up of the assessment of the VAFC calculation methodology by the European experts and will provide the information for further analysis.

Mykola Miagkyi’s Review

Giving deserved credit to the author for doing such a good research on a subject, there are two issues that are worth exploring further.

First, the article finds huge discrepancy between two seemingly identical measures of the “Value Added”. According to it, it amounts to UAH 600 bln, which is equivalent to almost ¼ of the official GDP. However, the article stops short from looking deeper into that number. I would suggest to explore what drives the discrepancy, at least at the sectoral level. Is there a particular sector that stands out, or the discrepancy is relatively consistent across sectors? While the article shows the difference for industry, it may be good to show the same gap for other sectors. This may give an insight to where to look further.

Second, the article can benefit from deeper understanding the methodology behind the calculation of the two indicators. Is there difference in samples? Are all sectors covered? How is the adjustment for informal activity being done? This may indicate an important methodological source of the discrepancy. Obviously that in countries with better quality reporting and data collection the two indicators will be closer than in Ukraine.

In general, I would not over-dramatize an issue with the discrepancy, but use it as an instrument for understanding economic data better.

Mariya Repko’s Comments

The figure of UAH 600 billion, presented as a discrepancy between two methods of calculating GDP, is astonishing — this is about a quarter of the economy. If it turns out that the statistical data may contain discrepancies like this in such an important indicator as economic value added, then almost any other statistical indicator can be called into doubt.

Inside the country, GDP is the basis for estimating and calculating the state budget, as well as making political decisions in the economic sphere. For external partners, GDP, GDP growth, and GDP per capita are key indicators of the country’s investment potential, living standards, economic conditions, and even geopolitical opportunities. This is true for the IMF and the World Bank, as well as for rating agencies and foreign investors.

Answer to the question how exactly the revision of GDP can influence payments on the GDP warrants, issued by Ukraine during restructuring of its sovereign debt, lies within the legal framework. The prospectus to issue securities says that owners of the warrants don’t have the right to dispute Ukraine’s methodology for calculating GDP or demand recalculation of payments after the revision. But such a large-scale revision may cause loss of confidence in the issuer. In any case, this will bring GDP closer to the barrier of $125.4 billion, after which, and when other conditions occur, payments can begin.

Realising the depth of the problem that Ukraine may face if the author’s calculations will be confirmed or won’t be refuted, we are eagerly awaiting an official response of the State Statistics Service of Ukraine and a reaction from the Ministry of Finance and the Ministry of Economic Development and Trade.

Yuriy Gorodnichenko’s Comments

Economic statistics are key ingredients for sound policy-making and Gross Domestic Product (GDP) is the king of economic indicators. Indeed, nearly all decisions at the macroeconomic level are based on GDP and its derivatives. Should the central bank raise interest rates? Should the government implement fiscal stimulus? How is the welfare at the aggregate level changing? Answers to all these key questions require a well-measured GDP.

Given the central role of GDP for policymaking, it is vital to have high-quality measures of GDP. Yet, measurement of GDP is fraught with a number of challenges, with a large size of the shadow economy being perhaps the most important factor. In this environment, it is inevitable that various approaches to calculate GDP can disagree and, perhaps not surprisingly, the disagreement can be substantial. However, a framework used to compute GDP should be internally and externally consistent so that in case there is a discrepancy across different methods there is a simple and intuitive way to rationalize the discrepancy. Is it so in the Ukrainian statistics?

Measurement of GDP is fraught with a number of challenges, with a large size of the shadow economy being perhaps the most important factor. In this environment, it is inevitable that various approaches to calculate GDP can disagree and, perhaps not surprisingly, the disagreement can be substantial.

Dr. Tetyana Tyshchuk does a series of checks to answer this question. Specifically, she examines whether a measure of value added based on a method that presumably should include shadow economy (the difference between gross output and intermediate consumption) is greater than a measure of value added based on method that presumably excludes shadow economy (sum of value added across firms). This is a straightforward check but she shows that the current value added statistics in Ukraine fails to pass it. Furthermore, the difference is staggering. If for a typical European country the ratio the narrow measure to the broad measure is somewhat less than one (and much less than one for countries with a large shadow economy sector), in Ukraine the ratio is close to two! Clearly, something does not add up in the national income and business statistics of Ukraine. Why does the statistical office of Ukraine have such major inconsistency? This is an open question but thanks for Dr. Tyshchuk’s important and timely work we know that we should re-examine the methods of collecting data and calculating GDP in Ukraine. This is a matter of utmost urgency for Ukraine’s statistical office.