Measuring Argentina's GDP: Myths And Facts

Ariel Coremberg - January 2014


Speed Read
  • Since 2007, official economic statistics in Argentina, particularly on consumer inflation and GDP, have been subject to political manipulation.
  • This paper reproduces Argentine national income from 2007 using standard methods and original sector data and finds that declared GDP is 12.2% higher in 1993 prices due to political intervention.
  • The paper finds that the distortion is mainly due to changes in accounting methodology across industries and not to changes in inflation estimates.
  • The reproduced GDP data dispels the myth that Argentina has been the fastest growing South American economy in recent years.


The Problems with Argentina’s Official Economic Data
During the last two decades, Argentina has experienced economic instability which has had a strong impact on the sustainability of the country’s long-run growth. In addition, there have also been a number of government actions that cast doubt on the veracity of official economic statistics. In 2007, the administration decided to suppress reported inflation by intervening in the calculation of the official Consumer Price Inflation Index (CPI) estimated by the National Statistics Institute (INDEC), an interference that still continues. Since 2007 several academic and private analysts have estimated that the actual CPI has been considerably higher than the one reported by the official series. In consequence there have been accumulating gaps between the official estimation of inflation and alternative ones and this also distorts the measurement of other important economic indicators, such as the poverty rate and the distribution of income. At present, for example, Argentina has an official poverty rate at a lower level than Sweden.

Political intervention in the production of statistics began in January 2007 with the CPI, and a few months later, the wholesale price index (WPI) was also modified, as well as also the official Household and Employment, Manufacturing Survey. This has had a profound influence on many other interlinked Argentine economic indicators. One of the reasons for the political intervention was to reduce the amount of the public debt indexed by the CPI “to increase public savings”. Paradoxically, at the same time, the positive upwards bias on GDP generated some substantial “extra payments” of public debt linked to economic growth.

This political manipulation of statistics has created an incentive to undertake alternative “unofficial” estimates of GDP in order to analyze the growth profile, productivity and competitiveness of Argentina more accurately The author has carried this out using the ARKLEMS database, resulting from a project organized by a team of Argentinean academics and researchers from the University of Buenos Aires. The project measures and analyses the sources of economic growth, productivity and competitiveness of the Argentinean economy at the macro and at the industry level. 

The purpose of this paper is to present an indicator of economic activity that reproduces the GDP growth of Argentina from 1993 to 2012 at a very detailed industry level, following the same traditional sources of information and methodology used by the Argentine National Accounts System during the twenty five years previous to the INDEC intervention of 2007. By showing this, we will also be able to measure the distortions induced by the political intervention.

It is necessary to distinguish “legitimate” measurement errors which can bias output and productivity measurement from manipulations aimed at showing political results. The latter problems have been well documented for countries such as Greece where Sturgess (2010), for example, reports distortions related to the measurement of public debt and public finance indicators, in order to show a public deficit figure below the 3% of GDP required to enter the Euro zone.

Argentinean CPI distortions have been reported in Cavallo (2012) , but many academics share the view that not only the CPI but also real GDP growth has been significantly lower in Argentina than what official publications report. Up to now these differences have not been systematically estimated, verified and reported. 


Reproducing GDP: The Methodology 
At present the GDP of Argentina is estimated following an old approach: using a volume index at 1993 prices, while national accounts in developed countries, and also in several Latin American countries, have progressively moved to the so-called “chain indices.” The National Accounts of Argentina also lag behind in terms of updating the base year (it uses an old set of relative prices and weights), but the main problem rests in the credibility of the data. 

Most components of Argentine GDP are estimated using volume index indicators and not by deflating the value of production or value added at current prices. This methodology was adopted more than two decades ago, and it is applied to the main basic series that constitute GDP in Argentina. Those volume index indicators do not belong to INDEC, but come mainly from other more representative public and private surveys carried out by different institutions. Checking up the consistency of actual GDP estimations is therefore a possible but difficult task, which demands time and also a detailed knowledge of the data sources and the methodology used for estimating the GDP figures. 

In order to do that checking, we reproduced the GDP of Argentina following the traditional National Accounts methodology. We compiled published and public series that made up every industry value added volume index of GDP at 4-5 digit of the ISIC classification from 1993 up to the present. The main result of this procedure is a series that replicates almost exactly the official Argentine GDP growth from 1993 to 2007. After that year, however, an important gap appears. This gap increases over time as is shown in Figure 1.

Official GDP shows a positive gap of 12.3% in 2012 with respect to our ARKLEMS reproduced GDP. During the period of political intervention in the production of official statistics, reproduced Argentina GDP grew by 15.9% between 2007 and 2012 (3% annually), a fall in the rate of growth of GDP with respect to the earlier period 2002-2007 of 47% (8.1% annual rate). However, after intervention official figures show a higher, almost doubled rate of GDP growth between the years 2007-2012: 30% (5.3% annual rate).


Figure 1: Argentina GDP 1993-2012, Volume index 1993=100.



Stylized Facts and Myths
Political interference in the production of statistics has created a number of myths about Argentina’s growth, which can be dispelled using our GDP estimations. Three myths are evaluated below which we call the methodological myth, Argentina’s resurrection myth and the myth that Argentina is a Latin American growth champion. 

The Methodology Myth: Official Argentine GDP data has a positive bias upwards because official INDEC estimation deflates value added at current prices with a manipulated consumer price index.

In reality the positive biases in the present official GDP are not due to a consistent application of a new methodology, but to the withdrawal of the traditional national accounts methodology and the application of discretionary non-reported criteria. The revision reports carried out by our study through traditional methodology and data sources, shows that only financial intermediation (5% of GDP) was affected by using inaccurate deflators . Furthermore, analyzing the contribution of every industry’s value added gap, financial intermediation explains only 27% of the total GDP gap between official figures and reproduced data. The other main sectors explaining the gap are trade (28%) and manufacturing (20%) while other services (9%), real estate (8.2%) and hotels and restaurants (4.5%) are also important. The rest of services and goods production explain a minor share, compensated by minor negative gaps from public administration and agriculture. 

To summarize, our GDP estimations show that the resulting gap does not depend on official CPI index manipulation instead, 70% of the total gap is due to discretionary intervention on individual industries by changing the original national accounts methodology.

The Resurrection Myth: Recent Argentina growth episode was the highest growth acceleration of for many decades.

It is important to analyse the GDP cycle to study whether an economy is recovering from a recession or a previous crisis, and whether it is starting a process of growth acceleration. It is important to assess whether an economy is growing in the long run, beyond a cyclical recovery (Hausmann et.al. 2005). One of the stylized facts of the commodity price boom that took place between 2002 and 2011 is that the economic growth of Latin America and Argentina occurred after a deep economic depression during 1998-2001. But there are a number of questions that need to be answered? How much of the economic growth was really due to the boom and how much was due to a “recovery effect”? By how much did GDP accelerate, in comparison to the previous positive growth phase that occurred during the so-called “Washington Consensus period” in the 1990´s? The answers to those questions depend on the consistency of the GDP series used.

The periods of analysis have been chosen in order to compare the actual boom that which took place during the reforms of 1990-1998 which lasted until the negative shock of 1998 followed by the depression of 1998-2002 with later periods. Moreover, comparing that crisis phase 1998-2002 to 2002-2012 allows an analysis of the impact of the present boom of commodity prices on the recovery after the crisis. We also report data for the periods 2002-2007 and for 2007-2012, taking into account that 2007 is not technically a cyclical peak, but rather is the year when the political intervention of official statistics began.

The Table below shows the GDP performance in the periods previously defined according to the official INDEC statistics and to our alternative GDP ARKLEMS reproducible estimations.

                     Source: ARKLEMS. GDP growth at producer prices.

The growth episode from 2002 to 2012 measured by reproducible GDP is similar to the previous positive phase in Argentine history (1990-1998). However, official figures show that the recent growth episode was much stronger than the previous one. The GDP estimates show “Chinese” style growth rates”, at approximately 8% annual growth, only during the period 2002-2007 (47.6% cumulative rate). But after 2007, Argentina suffered an important GDP slowdown: 15.9% of accumulated growth (3.0% annual rate). The official GDP performance, however, shows almost double the rate of growth at 29.4% (5.3% annual rate) compared with the reproducible GDP.

As has been demonstrated in Coremberg (2011,2012), an analysis of Argentina’s growth profile by source showed an unsustainable growth extensively based on factor utilization which was accompanied by a Total Factor Productivity slowdown during the recent growth episode. The reason for the slow-down is not only based on unsustainable growth and the inefficiency of productive factors, but mainly on lower growth performance that it is not recognize in official statistics. For example, Argentine GDP grew between peaks only by 2.5%, according to ARKLEMS estimation (instead of the official 3.4%). 

The Growth Championship Myth: Argentina is the Latin American country which has had the highest growth acceleration in the region during the last decade.

According to official figures, Argentina was the “growth champion” in the whole region over the period 2002-12 with an impressive cumulative growth of 99%, more than double the region’s average. However, if we use our ARKLEMS GDP estimation, instead, the growth performance of Argentina is substantially lower (71%, or nearly 30 points less than the official figures) and the country lies behind Peru and Uruguay in the rankings. This comparison is shown in Figure 2.


Figure 2: Latin America GDP Growth, 2002-2012 (compound rate, %). 
            Source: ARKLEMS and ECLAC.


Conclusion
This paper has made an exhaustive revision of the Argentina National Accounts' methodology in order to reproduce GDP series since 1993 and check economic growth after political interventions in the production of official statistics. 

Our series reproduces Argentina’s GDP growth closely from 1993 up to 2006. Since then, the difference between the official series and ARKLEMS reproducible series accumulates a large positive bias upwards due to intervention.

The divergence between our results and official figures is due to the withdrawal of the traditional GDP measurement methodology in almost every sector of GDP. The paper shows that these distortions are not based on deflating value added by industry at current prices levels with manipulated price indices but mainly on discretionary intervention in every industry component of the GDP, not only financial sector but also trade and manufacturing and the rest of services and goods production sectors, with the objective to be to show a higher GDP growth. 

The official series present a picture of GDP growing at higher rates during the recent recovery period (2003-2012) than in the previous one (1990-1998), and they make Argentina lead the GDP performance of the region. But our research demonstrated that ARKLEMS reproducible GDP for recent growth episode had a similar performance to the previous positive cycle, 1990-1998. Technical analyses based on typical National Bureau for Economic Research (NBER) cycle decomposition updated by Hausman et.al. (2005), shows that Argentina does not have a sustainable growth acceleration.

The paper showed that Argentina growth was important during recent growth episode, but contrary to official figures, the country was not the growth champion of the Latin America region. Argentina has also one of the largest GDP volatilities in the whole region, showing a very slow performance, lower (next to Mexico) than the region and Brazil, when the comparison is realized between peaks of economic cycle (1998-2012). 



References
Cavallo, Alberto (2012). Online vs Official Price Indexes: Measuring Argentina′s Inflation - Journal of Monetary Economics. December 2012

Coremberg, A. (2011): “The Argentine Productivity Slowdown.The challenges after global financial collapse”, World Economics 2011. Vol.12, nº4. http://www.world-economics-journal.com/Contents/ArticleOverview.aspx?ID=481

Coremberg, A. (2012): Measuring Productivity in Land Rich Economies. The ARKLEMS+LAND Project, WorldKLEMS 2nd Conference, Harvard University

Haussman R., Prichett L. and Rodrik D. (2005): Growth Accelerations, Journal of Economic Growth, December 2005, Volume 10, Issue 4, pp 303-329

Sturgess, Brian (2010): Greek Economic Statistics: A Decade of Deceit. So how come the rating agencies missed it again? World Economics, Vol. 11 No. 2 April–June 2010



Notes
1 The methodology is based on KLEMS framework (Capital, Labor, Energy, Material and Service Inputs) in coordination with the WORLDKLEMS Project lead by Pr. Dale Jorgenson (Harvard University), Marcel Timmer (Groningen University) and Bart Van Ark (Conference Board and Groningen University).

2 Another important and methodological consistent alternative estimation have been made by G. Bevaqua, former Director of CPI as CPI GB, which is issue by email distribution.

3 After 2007, many consultants and experts have been fined by the government; the Argentinean justice system has quashed penalties only recently after more than four years of trials and judicial conflicts


4 The International Standard Classification of All Economic Activities

In the case of restaurants and non-regular passenger transportation, the traditional methodology uses a demand function approach (because there are no direct surveys on those industries), so index prices do not enter directly in the estimation unless through the relative prices of those categories in the CPI. But this methodology was abandoned in 2003, before the the intervention. 

1 The methodology is based on KLEMS framework (Capital, Labor, Energy, Material and Service Inputs) in coordination with the WORLDKLEMS Project lead by Pr. Dale Jorgenson (Harvard University), Marcel Timmer (Groningen University) and Bart Van Ark (Conference Board and Groningen University).

2 Another important and methodological consistent alternative estimation have been made by G. Bevaqua, former Director of CPI as CPI GB, which is issue by email distribution. 

3 After 2007, many consultants and experts have been fined by the government; the Argentinean justice system has quashed penalties only recently after more than four years of trials and judicial conflicts

1 The methodology is based on KLEMS framework (Capital, Labor, Energy, Material and Service Inputs) in coordination with the WORLDKLEMS Project lead by Pr. Dale Jorgenson (Harvard University), Marcel Timmer (Groningen University) and Bart Van Ark (Conference Board and Groningen University).

2 Another important and methodological consistent alternative estimation have been made by G. Bevaqua, former Director of CPI as CPI GB, which is issue by email distribution. 

3 After 2007, many consultants and experts have been fined by the government; the Argentinean justice system has quashed penalties only recently after more than four years of trials and judicial conflicts