A Service of Leibniz-Informationszentrum econstor Wirtschaft Leibniz Information Centre Make Your Publications Visible. zbw for Economics Gylfason, Thorvaldur; Tómasson, Helgi; Zoega, Gylfi Working Paper Around the World with Irving Fisher CESifo Working Paper, No. 5257 Provided in Cooperation with: Ifo Institute – Leibniz Institute for Economic Research at the University of Munich Suggested Citation: Gylfason, Thorvaldur; Tómasson, Helgi; Zoega, Gylfi (2015) : Around the World with Irving Fisher, CESifo Working Paper, No. 5257, Center for Economic Studies and ifo Institute (CESifo), Munich This Version is available at: http://hdl.handle.net/10419/108797 Standard-Nutzungsbedingungen: Terms of use: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your Zwecken und zum Privatgebrauch gespeichert und kopiert werden. personal and scholarly purposes. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich purposes, to exhibit the documents publicly, to make them machen, vertreiben oder anderweitig nutzen. publicly available on the internet, or to distribute or otherwise use the documents in public. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, If the documents have been made available under an Open gelten abweichend von diesen Nutzungsbedingungen die in der dort Content Licence (especially Creative Commons Licences), you genannten Lizenz gewährten Nutzungsrechte. may exercise further usage rights as specified in the indicated licence. www.econstor.eu Around the World with Irving Fisher Thorvaldur Gylfason Helgi Tómasson Gylfi Zoega CESIFO WORKING PAPER NO. 5257 CATEGORY 7: MONETARY POLICY AND INTERNATIONAL FINANCE MARCH 2015 An electronic version of the paper may be downloaded • from the SSRN website: www.SSRN.com • from the RePEc website: www.RePEc.org • from the CESifo website: www.CESifoT -group.org/wpT ISSN 2364-1428 CESifo Working Paper No. 5257 Around the World with Irving Fisher Abstract This paper aims to show why Irving Fisher’s own data on interest rates and inflation in New York, London, Paris, Berlin, Calcutta, and Tokyo from 1825 to 1927 suggested to him that nominal interest rates adjusted neither quickly nor fully to changes in inflation, not even in the long run. In Fisher’s data, interest rates have more persistence than inflation and change less than inflation over time. The Fisher effect is a misnomer unless it is taken to refer to what Fisher actually found and what his data show: a persistent negative effect of increased inflation on real interest rates. JEL-Code: E310, E430. Keywords: Fisher effect, inflation, interest rates. Thorvaldur Gylfason Helgi Tómasson Faculty of Economics Faculty of Economics University of Iceland University of Iceland Iceland – 101 Reykjavik Iceland – 101 Reykjavik [email protected] [email protected] Gylfi Zoega Faculty of Economics University of Iceland Iceland – 101 Reykjavik [email protected] 2 1. Introduction Ricardian equivalence, as you know, refers to the idea that government budget deficits do not matter because taxpayers are indifferent between debt-financed and tax-financed government expenditure: they realize that current debt needs to be serviced through future taxation and plan their saving accordingly. However, the attribution of this proposition to David Ricardo is unfair to him because, even if he exposited the logic behind it, he found the proposition unconvincing. To quote from Ricardo (1817, 254): “… it must not be inferred that I consider the system of borrowing as the best calculated to defray the extraordinary expenses of the State. It is a system which tends to make us less thrifty – to blind us to our real situation.” This short paper is intended to demonstrate anew that Ricardo is not alone, for Irving Fisher has suffered a similar treatment by his followers. The Fisher effect – through which nominal interest rates react to changes in inflation point by point so as to leave real interest rates unchanged, at least in the long run – is a comparable misnomer because, as will be shown here, Fisher’s (1930) own data on interest rates and inflation that he collected from six financial centers around the world suggest that nominal interest rates do not come close to mirroring the movements in inflation, even in the long run. Fisher (1930, 413) recorded “a great unsteadiness in real interest when compared with money interest.” Yet, many writers continue to attribute to Fisher the idea that real interest rates are immune to changes in inflation and to suggest that Fisher thought it somehow natural for real interest rates, like private consumption in the Ricardian case, to be so immune. For example, Okun (1981, 208) states incorrectly: “As Fisher saw it, an extra 1 percentage point of expected inflation raises the nominal expected rate of return on real capital assets by 1 percentage point and induces a parallel increase of 1 percentage point in bond and bill yields to keep expected returns in balance.” For another example, Feldstein and Eckstein (1970, 366) write: “The data thus confirm the two basic Fisherian hypotheses: (1) in the long run, the real rate of interest is (approximately) unaffected by the rate of inflation, but (2) in the short run, the real rate of interest falls as the rate of inflation increases.”1 In fact, Fisher thought no such thing. On the contrary, as Tobin (1987) and Dimand (1999), among others, point out, both Fisher’s theory of interest and his reading of the historical record suggested to him that real interest rates varied inversely with inflation, and that the adjustment of nominal interest rates to changes in inflation took a very long time (Fisher, 1896). In Fisher’s (1930, 43) words: “… when prices are rising, the rate of interest tends to be 1 Feldstein and Eckstein (1970) used quarterly U.S. data for 1954-1969. 3 high but not so high as it should be to compensate for the rise; and when prices are falling, the rate of interest tends to be low, but not so low as it should be to compensate for the fall.” 2 To stress the point, we revisit Fisher’s (1930) data. 2. Fisher’s Data from Six Cities: A Time Series Approach In an Appendix to his Theory of Interest (1930, 520-5), Fisher tabulates nominal interest rates as well as wholesale commodity price indices in six financial centers: New York, London, Paris, Berlin, Calcutta, and Tokyo, for a period spanning up to a hundred years from 1825 to 1927.3 During this period, for decades on end, prices in New York, London, and Berlin rose by merely a fraction of a percentage point per year while prices actually fell on average in Paris. Meanwhile, prices in Calcutta and Tokyo increased by 2.1 per cent and 3.9 per cent per year on average (Figure 1). Fisher’s informal analysis of the data confirmed his view that nominal interest rates tended to adjust only partially and slowly to changes in inflation, but he squeezed less juice out of the data than he might have. We start by studying correlations across Fisher’s data after removing trends and cycles from each time series yt with an AR(2) model, either by estimating (1) ͭ/ = + ͨ + &ͥͭ/ͯͥ + &ͦͭ/ͯͦ + / or ∗ ∗ ∗ (2) Δͭ/ = + &ͥΔͭ/ͯͥ + &ͦΔͭ/ͯͦ + #/ We interpret the estimated residuals / and #/ as the prewhitened versions of the original time series, and use them to study the magnitude and duration (i.e., cycle length) of the innovations in each series as well as their lead-lag structures and correlations across countries. Table 1 suggests cycles of about three to four years (Model 2) or more (Model 1), except Model 1 does not suggest cyclical behavior of interest rates or inflation in New York or of inflation in Tokyo. With the possible exception of the interest rate series in New York and Tokyo, the ADF test results for unit roots do not suggest that detrending the series with Model 2 is better than using Model 1 (the 0.05 critical value is -2.9 throughout). Both New York and Tokyo exhibit extreme values near the end of the sample as does London (Figure 1). Because several of the series are quite short (about 40-50 years in New York, Paris, Berlin, and 2 Fisher (1930, 494) describes the relationship between interest rates and inflation also thus: “When the price level falls, the rate of interest nominally falls slightly, but really rises greatly and when the price level rises, the rate of interest nominally rises slightly, but really falls greatly.” Here Fisher means the rate of change of the price level even if he says only “price level.” Fisher (1907, 270) made a clear distinction between the two: “Falling prices are as different from low prices as a waterfall is from sea level.” 3 The sample periods vary from place to place, see Table 3. 4 Tokyo), the cycle lengths shown in Table 1 cannot be very precisely estimated. Table 1. Cycle Length of Interest Rates and Inflation Rates 1825-1927 and Unit Root Tests AR(2) model (1) AR(2) model (2) Augmented Dickey- (Years) (Years) Fuller unit root test (Intercept, lag = 2) Nominal Inflation Nominal Inflation Nominal Inflation interest rate interest rate interest rate rate rate rate New York Long Long 3.3 3.1 -2.37 -3.46 London 17.7 8.1 3.5 3.1 -4.05 -4.98 Paris 8.1 4.2 3.9 3.2 -3.95 -3.45 Berlin 9.4 5.4 3.4 3.2 -3.39 -4.51 Calcutta 12.3 4.9 2.8 3.5 -5.12 -4.38 Tokyo 8.1 Long 4.1 Long -2.54 -3.58 Source: Authors’ computations based on Fisher’s data.