Hard times for the world economy bring hard times for the economic profession too. As noticed by Axel Leijonhufvud at the ouset of the ongoing financial crisis, “Abundant compliments and congratulations have been exchanged among academic economists and central bankers over the apparent successes of inflation targeting in recent years. Are these well deserved? Or have we benefited from what may turn out to be a historically unique conjuncture?“. This site itself witnesses that the debate over the possible fallacies in the theoretical paradigm underlying modern central banking has already started.
·The monetary interest rate should be anchored to the NAIRI (non-accelerating-inflation rate of interest), which is given by the “natural rate of interest” (the real rate that is associated with output being at its potential level) and the target inflation rate set by the central bank. Deviations of the monetary rate above or below the NAIRI trigger deceleration or acceleration of inflation.
· The monetary interest rate should be raised above, or pushed below, the NAIRI as inflation (and output) accelarates or decelerates with respect to the target rate.
It is worth stressing the corrolary that the three arguments of the standard Taylor rule, the NAIRI, the cyclical position of ouptut with respect to potential (output gap) and that of inflation with respect to the target (inflation gap) convey to the central bank all the necessary information to stabilize the economy optimally. In particular the evolution of credit and financial variables is only relevant to the central bank insofar as they impinge upon inflation. And the theory predicts that if they go out of line, they do impinge upon inflation. CPI stability goes hand in hand with asset price stability. The additional argument is that detecting overlending and/or asset price bubbles directly is difficult and may lead to wrong policy decisions.
As a result of the association of this paradigm with the age of “Great Moderation” – the sustained growth and employment with low and stable inflation that blessed most of the world economy in the 1990s – the profession thought that the right macroeconomic recipe had been found. The fact is that since 2000 the world has witnessed two financial boom-bust cycles of major order of magnitude with widespread macro-consequences originating in the private sector of the best developed economy in the world. In a careful analysis of the emprical regularities displayed by macro-financial crises, well before the current one, Borio and Lowe emphasized that the years of the conquest of inflation “will in all probability also be remembered as those that saw the emergence of financial instability as a major policy concern, forcing its way to the top of the international agenda […] Ostensibly, lower inflation had not by itself yielded the hoped-for peace dividend of a more stable financial environment” (p. 1)
Borio and Lowe also outline a few stylized facts: “Widespread financial distress typically arises from the unwinding of financial imbalances that build up disguised by bening economic conditions […] Booms and busts in asset prices […] are just one of a richer set of symptoms […] Other common signs include rapid credit expansion, and, often, above-average capital accumulation“ (p. 1; underline added)
Viewed through the original Wicksellian lenses, these patterns are less puzzling than they appear through the NNS lenses. Rather, it is the current theory and practice of monetary policy, namely inflation targeting of the Taylor-rule type, that appears to be suffering from three interacting bugs.
1) Targeting the NAIRI, that is the natural interest rate, is hazardous. Wicksell was well aware that the natural rate is subject to unobservable shocks and fluctuations (e.g. 1898a, pp. 82 ff.). Modern real business cycle theory of course subscribes to this view. Keynes was even more radical, casting doubts on the existence itself of a single, general-equilibrium real interest rate. Up-to-date econometric research is by no means encouraging about the possibility that central banks can ever obtain all information necessary to target the NAIRI sensibly (see e.g. Caresma et al. for a recent survey). If the central bank happens to target the wrong NAIRI, it may drive the economy on the wrong track. Faia and Angeloni impute the “too low for too long” US policy rate to a misperception of the NAIRI. On the presumption that yields on AAA long-term corporate bonds convey some market information about the NAIRI, the chart below gives a pictorial representation of this misperception hypothesis.
Figure 1. U.S. Federal Funds Rate and Yields on AAA long-term corporate bonds, 2000:1-2008:4 (monthly data)
Source. FRED Online database, Federal Reserve of St. Louis.
It is also odd that economists who argue that detecting asset price bubbles is difficult at the same time urge central banks to target the NAIRI. For in the general-equilibrium economy they assume, there exists a one-to-one relationship between the natural interest rate underlying the NAIRI and the equilibrium value of capital assests. Curiously, Wicksell used just the housing sector to illustrate his theory (1898a, p. 75 ff.). The natural rate is given by the ratio of the annual rental of a building to its market value such that demand and supply of houses remain in equilibrium and the price level remains constant. If the market interest rate at which anyone can borrow falls below the natural rate, then buying houses becomes profitable, their demand price rises and so do wages and prices in the housing sector. The market value of houses increases and goes on inflating as long as there is too low a market interest rate, excess lending and overinvestment. Thus, the debate whether the Taylor rule should include asset prices as a separate item or not is somewhat nonsensical. For targeting the NAIRI is the same thing as targeting the nominal value of the equilibrium value of capital assets. And if one believes that asset prices should be dropped because it is difficult to detect whether they are on a bubble or not, the same should hold for the natural interest rate.
2) In the previous quotation from Borio and Lowe I have underlined the role of financial imbalances because this is a key element, if not the key element, that should lead the search for the bugs in the operators’ programme that macroeconomists have delivered to the policy makers. First of all, financial imbalances, more precisely saving-investment imbalances, are indeed the driving force behind Wicksell’s interest-rate theory of the price level. In Wicksell’s theory, the consequence of the market interest rate on loans being lower than the natural rate is that savers wish to save less whereas investors wish to invest more. In modern parlance we would say that neither side of the market can achieve intertemporal equilibrium of plans. The ensuing “cumulative process” of increases in the prices of both goods and assets is a disequilibrium phenomenon, the symptom that excess investment is being accommodated at the “wrong” market rate and the economy driven out of the intertemporal equilibrium path (e.g. Wicksell, 1898a, pp. 75 ff.). On the other hand, changes in the price level are a means to re-equilibrate the economy only if, and to the extent that, they induce the market interest rate to close the gap with the natural rate (Wicksell, 1898a, pp. 80 ff.).
However, saving-investment imbalances are not contemplated in the NNS paradigm, simply because intertemporal disequilibrium falls outside the methodological boundaries of the discpline. As recognized by R. Lucas
“The problem is that the new theories, the theories embedded in general equilibrium dynamics […] don’t let us think about the US experience in the 1930s or about financial crises and their consequences […] We may be disillusioned with the Keynesian apparatus for thinking about these things, but it doesn’t mean that this replacement apparatus can do it either” (p. 23, underline added)
Indeed, ruling saving-investment imbalances out of the theory constitutes a major weakness because they are a logical implication in any theory based on the distinction between the market interest rate and the natural rate, which is, by definition, the interest rate which equates saving and investment in the general equilibrium. To put it very simply, the point with the market interest rate being too low today is not only that there is excess demand today but also that there will be excess supply tomorrow. Overlooking this fact seriously limits the central bank’s ability to understand and steer the evolution of the business cycle.
3) A major consequence closely related to the current situation concerns the “missing inflation” puzzle. As indicated by Borio and Lowe, and by the previous chart, financial imbalances may build up disguised in a low-inflation, high-output environment. Which, probably, has misguided US monetary policy as well as capital asset valuations for a number of years in a row (Leijonhufvud, 2007). One reason (perhaps deeper than sticky prices, or the large supply of cheap goods from India and China mentioned by Faia and Angeloni) may be that as long as firms overinvest, the stock of physical capital and productive capacity increase. Wicksell assumed continuous full employment of resources, so that additional supply would never materialize. But, as was pointed out by Hyman Minsky, in a more probable world with idle and mobile resources, the counterpart of the low interest rate may be a positive supply-side effect. As a result, output grows, excess demand is offset over time, and CPI inflation is damped. The central bank and the business community as a whole believe they are living in the best possible world so that S&P inflation occurs. If this is the boom-phase, however, overinvestment and intertemporal disequilibrium mean that installed productive capacity will eventually be in excess of the absorption of the economy. This sets the stage for the bust-phase of revenues falling below debt service or expected dividends, losses and sharp correction of capital asset values. The serious problem that this pattern of boom-bust cycles poses to inflation targeting is that CPI inflation is not the right signal to look at in these circumstances.
Some tentative conclusions. Real disequilibrium business cycles disguised behind Great Moderation seem a recurrent pattern that seriously challenges the present state of the art and science of monetary policy. Probably, there has been overoptimism and overconfidence in the NNS paradigm as if it were the “End of History” for central banking. To say the least, granted that all the deceitfully “natural” variables (NAIRI, NAIRU etc.) that are plugged into NNS monetary policy rules are hard to detect and chase reliably, it is recommendable that central banks resort to simpler and more robust adaptive rules based on observable cycle indicators of the type suggested by e.g. Orphanides et al. Also, more open-minded monitoring of credit and financial variables seems necessary (with the public being aware that the central bank does monitor these variables and take them into account seriously). The apparently odd idea of the ECB of controlling the interest rate and then monitoring also monetary aggregates should perhaps be reassessed in this light. On the more theoretical side, one major problem seems to lie in an overly simplistic conception of the macroeconomy in relation to the (mal)functioning of capital markets and their role in the business cycle. This is a major stumbling block whether one advocates better monetary policy or better regulation policy of financial markets. It is amazing that the debate is going on as if the two issues were independent or alternative. Clearly, we need reforms on both sides, and to this effect we need better, integrated, understanding of micro-sources and macro-consequences of capital market failures. Less anecdotal reconsideration of Wicksell’s (and, on these grounds, of Keynes’s) insights about saving-investment imbalances might help.