5:25 PM Sep 21, 1995
IMF DATA CREATE PROBLEMS AT UNCTAD
Geneva 21 Sep (Chakravarthi Raghavan) -- Financial data provided by the International Monetary Fund, and their interpretation and methodologies used, proved a sore point of contention at the just concluded UNCTAD Trade and Development Board session. While ultimately, a concord of sorts was reached between Argentina and the UNCTAD secretariat over this issue, the subject though appears to raise many problems, according to statisticians and economists elsewhere. The Board session was marked by some differences between Argentina and the UNCTAD Secretariat over the data in the TDR relating to Argentine capital flows and the problems it was facing. The differences related mainly to the capital inflows into Argentina in 1993, not on the actual figures, but on their interpretation and meaning. The TDR, in discussing the over-reliance of the Latin American region in short-term capital flows to finance the current account deficits, with such flows (by their very nature) going to finance of consumption rather than investment, had dealt with what was seen as an exception to the rule in the case of Argentina, namely a net portfolio disinvestment of nearly nine billion dollars in 1993. The Argentine delegate, Amb. Sanchez Arnau took particular exception to this "huge error", and challenged the data, noting that as a recent World Bank report had said, in fact Argentina had a net gain of ten billion dollars in that year. He referred in this connection to several bond issues of Argentina in the international markets that year, issues by the government and a number of public sector bodies which had resulted in a large positive inflows. Arnau also complained that the TDR had posed the problems facing Argentina now, and in the TDR had said that half of Argentina's exports were now going to Brazil and this may not be sustained. However, less than 30% of Argentine exports went to the Mercosur partners (Brazil, Paraguay and Uruguay combined) and no greater than 25% to Brazil. Roger Lawrence Director of the division said this last was an editing error, and what the TDR was trying to convey was that 50% of the increase in exports by Argentina (in coping with the aftermath of the Mexican crisis) has gone to Brazil -- where the conditions were changing. But on the statistical data issues raised by Argentina, Lawrence in a first response said that the data used by UNCTAD about the portfolio investment outflows were derived from the IMF's International Financial Statistics Yearbook (which provides a uniform and comparable set of figures, across the years and across countries). However, Lawrence noted, there was no difference between Argentina and the UNCTAD Secretariat and its TDR that Argentina had received a positive 9-10 billion dollar inflows in that year. According to the IFS data (on page 191 of the yearbook), the portfolio investment n.i.e. (not included elsewhere) is shown for that year to be 9 billion dollars, while there is shown separately a figure of inflows amounting to $19.5 billion as 'exceptional financing'. The IFS data itself does not explain what this "exceptional financing" is and why it is not treated as part of the portfolio investment figures in that year for Argentina. But Arnau came back to cite the statistical annual publication of ECLAC (the Santiago-based UN regional body) which in the Balance of Payments data records a net portfolio investment of slightly over $18 billion for Argentina in 1993, a direct investment of $6.2 billion and a negative $11.8 billion of 'other long term capital, and suggested that there should be a corrigenda to the TDR. At the Board's final session on Wednesday, Lawrence reverted to the issue and said that the secretariat recognized that the differences between its own figures and those of Argentina were the result of differences in methodology in the definition of categorisation of financial flows. Therefore, he added, the TDR was to be understood as indicating that Argentina had overall received a net inflow in 1993. Not cited by either side in the exchanges, but adding to the confusion over data and its interpretation, was the IMF's Balance of Payments data in the IMF's annual BOP data yearbook published around the same time as the IFS yearbook. The IMF's BOP data is taken as sacrosanct, for example, at the WTO in relation to current account and BOP situations of countries. The BOP year book data, at variance with the IFS data, at first sight seems to bear out the Argentine complaint. It shows a $18.67 billion figure as inflows through other (non-Brady) bonds issued by Argentina and a negative $8.18 billion figure for redemption of securities. With one or two other related items, this records positive inflow of foreign financial flows -- of the same order cited by Argentina. This IMF-BOP year book figures seemed to be near about the figures in the ECLAC document cited by Arnau who pointed out that the ECLAC secretariat harmonises its own data with national data from governments of the region before publication and thus was most authoritative. But according to UNCTAD sources, the IMF has since clarified that IMF has decided that its IFS statistical classification was the proper one, and the figures in the BOP yearbook was an error (in terms of the different classifications) and has been or was being corrected. How the IMF would do this correction, whether by merely revising the figures (for 1993) in the 1995 BOP year-book, and with or without a small footnote of the revision, or with some more prominent explanation remains to be seen. Also remaining to be seen is whether the IMF Executive Board will take note of this and/or whether Argentina will raise this at the IMF? Observers here note that there is no dispute at all about the correctness of the figures of inflows into Argentina, cited by Arnau from national data. The problem seemed to be the IMF's classification of the data and any lack of explanation of this by the IMF so far nor any explanation, even by footnote, of the nature of that inflow by sale of bonds meriting its being treated as exceptional. Argentina's sensitivity over the different categorisation of its inflows and possible inferences could be seen in terms of the very difficult and belt-tightening measures at home being taken by its government to overcome the after effects of the Mexican crisis. Though no explanation of the amounts raised by Argentina is available, experts make some intelligent guesses of its possible nature. They suggest that the amounts raised by it in 1993, which the IMF chose not to include in portfolio investments, but as 'exceptional' flows, in pure commercial and sound external finance management is no doubt that something Argentina could be congratulated and which Argentina might rightly claim is indicative of the market's confidence. After the rescheduling of its external debt under the Brady scheme, and the issuance of Brady bonds which restored its sovereign borrowing status on international capital markets and improved its borrowing margins (over Libor), coupled with the reduction of US interest rates in 1993, Argentina probably restructured its debts by floating new bond issues and using it to pay off, costlier older bonds and loans. In fact it would have been stupid of Argentinean authorities if in 1993 they had not taken advantage of the relatively easier conditions to restructure their debt and thus benefit from lower rates and margins. Should this be viewed as new inflows or exceptional refinancing or what? If the IMF had taken a consolidated view of the inflows and outflows on this score to record a positive inflow, the Argentinean experience of short-term inflows financing the trade deficits and consumption expenditure would have been similar to that of the other Latin American economies until the Mexican crisis. But without an IMF explanation and a practice of regularly in its publications, statistical or analytical, explaining its premises, the differences of interpretation and assumptions would remain. Economists and statisticians elsewhere, at other UN bodies and in non-governmental institutions, say that while providing a standard and detailed financial data and analysis, by not fully explaining its variations and/or its assumptions, IMF data and conclusions often create difficulties for outsiders who work on those figures, particularly if they disagree with. They cite in this connection, the IMF assumptions, not clearly spelt out, for example, in estimating capacity-utilization in countries, an ingredient in assessing the inflationary risks of the annual gdp growth and whether there should be monetary tightening or relaxation. While every economist and econometrist has to make some assumptions, where data are not easily available, they normally spell out their assumptions clearly -- which enables others to agree or disagree and say why the assumptions could be wrong. Also, recently, the Bretton Woods institutions, faced with a negative or bleak situation in the transition economies (which have not been able to recovery quickly, as originally anticipated, from the shock therapies) have been arguing that the GDP figures in national data do not fully reflect the actual situation on the ground, and that the data need to be 'corrected' through calculations of the shadow economy. Some economists say that under IMF guidance, some of the national authorities of the Baltic and East Europeans to do the same, particularly on the contribution to GDP of the financial sectors, and thus arrive at a positive GDP growth, involving some times as much as a 15-20 percent adjustment -- but which are not spelt out. Experts like Prof. Jan Kregel at the John Hopkins University Campus at Bologna Centre in Italy say this is nothing unusual and notes that Italy a few years ago corrected its GDP figures for its shadow or black (unreported incomes etc) economy, with the GDP going up by 15-20 percent. But the problem, Kregel explains, arises from the fact that there is no rule of thumb which could be applied across countries or even within countries over time for such estimations, but rather a 'rule of nose' guesstimate: those estimating the shadow economy having to 'smell' their way into what they think is happening. Since all 'shadow economy' transactions are cash transactions -- whether labour or self-employed workers not reporting or small entrepreneurs under-reporting cash receipts -- the normal methodology for calculations is to take some historical figures of money in circulation and its ratio to GDP, and use it, on the basis of money now in circulation to correct the GDP, he says. Similarly, since the construction industry is one of the sources of such 'leakage' -- particularly construction or repairs and improvements to private homes, the cement imports plus production is similarly used. And for transition economies, whose past data is seriously questioned, the adjustment becomes very iffy. But unless this is all clearly spelt out, and available easily in the data figures, it is difficult to judge between an honest guesstimation and one of 'massaging' of data to justify an ideological stand, and thus leading to arguments about data and policy conclusions and issues of credibility. The only way would be for a clear 'understanding' among statisticians and the various international organizations on these matters, and even more an agreement to specifically state the assumptions and basis of calculations in arriving at data. That day though may be still far off - when the BWIs and the UN systems will cooperate on equal basis.