Browsing by Author "Saarinen, Jaakko"

Economic theory states that the financial markets are forward looking, and that the price of any asset equals the discounted present value of the income generated by holding that asset. The future is, however, uncertain which means that expectations play a large role when investors ponder over what the price of an asset should be. When considering government debt obligations, or bonds, the flow of future income is known with some degree of certainty, since the coupon payments are set in the terms of any bond issue, contrary to e.g. dividends from holding a stock which depend on the company’s success. This means that the uncertainty regarding the real income generated from holding a bond stems from, among other things, the possibility of a sovereign default and inflation. Hence the price of a government bond, and thus the market interest rate (the yield), which are inversely related, reflects the investors’ expectations about the future prospects of a country. It is because of these reasons, that many financial economists have studied the dynamics of the yield curve, which plots the market rates of bonds against their remaining maturities to redemption, and the relationship between the shape of the yield curve and future economic growth. The shape of the yield curve in most of the studies is measured by the difference, or the spread, between long term- and short term market yield of government debt. This relationship between the contemporaneous yield curve shape and subsequent economic growth has been confirmed by a number of studies internationally. However, the results have varied between countries and also in time, which means that the predictive power of the yield curve is probably not structural, but rather depends of country specific characteristics and the type of monetary policy practised by the central bank. While the international literature on the subject is vast, there have not been many studies investigating this relationship with Finnish data. This serves as the main motivation behind this thesis. The macroeconomic explanation for why such a relationship should exist is based on a model suggested by Arturo Estrella in 2005. The model is constructed from an IS-curve, a Phillips curve and the reaction function of the central bank, as well as the minimisation problem of the central bank’s loss function. This means that the model takes the prevalent monetary policy in to account when considering the ability of the yield curve to predict future growth. This study employs Finnish quarterly level of GDP data which spans from 1975 to 2011. The contemporaneous yield spread between a 10-year government bond and a 3-month market interest rate (a proxy for the 3-month T-bill rate) is used as a predictor in an OLS-regression to investigate the predictive power with many forecasting horizons and four different model specifications. The time series of the growth rate of GDP shows persistence, and for this reason also the contemporaneous growth rate of the GDP is also included as a predictor. The regressions are first run on the whole sample period, and also on a sub-period 1987 – 2011. This is because before the middle 1980’s the financial markets in Finland were heavily regulated, and hence the interest rates could not effectively reflect the market participants’ expectations about the future. This study finds that the yield spread is able to predict GDP growth rate in Finland for up to three years in to the future in the latter sample period. The results from the whole sample period were quite poor. Moreover, the contemporaneous yield spread is a better predictor of future growth than the contemporaneous growth rate itself. When considering a rather simple indicator of future growth, these results are encouraging. This is quite impressive especially since many of the international studies find that the predicting power of the yield curve has diminished since the 1980’s, which is contrary to the findings of this study.