Fool The Markets? Creative Accounting, Fiscal Transparency And Sovereign Risk Premia
German Institute for Economic Research (DIW Berlin)
Guntram B. Wolff
Deutsche Bundesbank; University of Bonn – Center for European Integration Studies (ZEI)
We investigate the effects of official fiscal data and creative accounting signals on interest rate spreads between bond yields in the European Union. Our model predicts that risk premia contained in government bond spreads should increase in both, the official fiscal position and the expected ”creative” part of fiscal policy. The relative importance of these two signals depends on the transparency of the country. Greater transparency reduces risk premia. The empirical results confirm the hypotheses. Creative accounting increases the spread. The increase of the risk premium is stronger if financial markets are unsure about the true extent of creative accounting. Fiscal transparency reduces risk premia. Instrumental variable regressions confirm these results by addressing potential reverse causality problems and measurement bias.
Fool The Markets? Creative Accounting, Fiscal Transparency And Sovereign Risk Premia – Non-Technical Summary
A number of empirical and theoretical papers show that EU Member States in various cases have used questionable accounting practices and data interpretations as well as temporary measures to beautify fiscal data in the context of the requirements of the Maastricht Treaty and the Stability and Growth Pact. For the European Union, recent research by Dafflon and Rossi (1999), Milesi-Ferretti and Moriyama (2004), Koen and van den Noord (2005), and von Hagen and Wolff (2006) confirms that fiscal policy figures of EU countries are purposely beautified to circumvent the constraints on deficits and debt in order to officially comply with the Stability and Growth Pact (SGP).
The reaction of financial markets to this creative accounting is an important policy topic. If financial markets do not price in the de facto deterioration of the fiscal position due to creative accounting, while punishing official deficit data, risk premia could be lowered by shifting deficits to creative accounting. The lower interest rate would provide an incentive to governments to beautify their fiscal data. To our knowledge, no study so far analyzes whether financial markets take note of fiscal window-dressing when pricing government bonds. This is the purpose of our study. In particular, we study whether spreads react, besides official fiscal data, to stock-flow adjustments or to an alternative measure of creative accounting by Koen and van den Noord (2005).
Furthermore, we investigate, in how far fiscal transparency affects risk spreads. Kopits and Craig (1998) argue that international financial markets are likely to demand lower premiums from governments that are forthcoming about their fiscal position and risk. The argument is that markets can be more certain about a fiscally transparent government’s ability and willingness to service its obligation. A more transparent budget process in addition helps financial markets to detect creative accounting more easily and to assess the true fiscal position of a country. This might increase the spread since more creative accounting becomes known to the markets.
We develop a portfolio model of interest differentials based on Bernoth, von Hagen, and Schuknecht (2004). In this model, interest rate differentials increase with a relative worsening of the fiscal position due to an increase in the government’s default probability. The model is augmented to account for fiscal creative accounting and fiscal transparency. Creative accounting appearing in the media constitutes a news signal. The more reliable this signal, the greater will be the effect of creative accounting on the expected fiscal position of a country. Creative accounting news should therefore increase the default risk premium. Fiscal transparency should reduce spreads by lowering the uncertainty of fiscal policy. In addition, it influences the relative information content of the official and the news signal as more transparent countries probably provide more reliable official data but also the quality of the news signal increases.
The empirical results confirm the hypotheses derived from the model. Creative accounting increases risk premia. The gimmickry events, that make it in the financial news, have significant punishing effects on risk premia. This is especially true, if a country is intransparent, as financial markets then take gimmickry as a ”tip of the iceberg” signal. Creative accounting thus increases the cost of borrowing significantly, if it becomes known. This holds especially if financial markets are unsure about the true extent of creative accounting. Fiscal transparency is connected with lower risk premia. Deficits and creative accounting are penalized less in EMU. Instrumental variable regressions, addressing potential simultaneity and attenuation biases, confirm the results.
See full PDF below.