Introduction and Overview

  • Jan Marc Berk
Chapter
Part of the Financial and Monetary Policy Studies book series (FMPS, volume 35)

Abstract

For a central bank to achieve its ultimate objective(s), it would be preferable to know in detail how monetary policy affects non-financial activity, that is inflation and real output. In practice, however, the central bank only has imperfect knowledge of the exact interactions that take place within various chains of the monetary transmission mechanism, but has (subjective) probability distributions in mind. The leitmotiv of this book is that this uncertainty should have implications for the preparation of monetary policy. Our goal is to translate the intricate interactions between various endogenous and exogenous variables (policy instruments) into simple relationships. These could then serve as guidelines for and legitimation of specific forms of monetary policy. Or, to put it differently, the degree of transparency of monetary policy and the successful use of rules in conducting monetary policy may be said to, inter alia, depend on the ability of the central bank to establish such relationships empirically. We study some of these relationships in subsequent chapters. The purpose of this chapter is to clarify some important theoretical concepts in order to illustrate the interrelationships between the following chapters.

Keywords

Interest Rate Monetary Policy Price Stability Monetary Authority Market Interest Rate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Goodhart and Vinals (1994); Okina (1995); Fuhrer (1997).Google Scholar
  2. 2.
    See Feldstein (1996) for a calculation of the economic costs of inflation in the US implied by the interaction of the tax system and inflation.Google Scholar
  3. 3.
    In contrast, price level stability is a device to stabilise economies with coordination failures. More specifically, downward wage rigidity provides a break against runaway deflation. It is thus a feature of labour markets that stabilise the economy against extreme outcomes by reducing deflationary expectations and permitting real interest rates to fall (Akerlof, Dickens and Perry, 1996, pp. 51/52).Google Scholar
  4. 4.
    See Romer and Romer (1996) for a discussion of the institutional settings of the central bank most conducive to realising monetary stability.Google Scholar
  5. 5.
    Fair (1996), working with data from thirty countries, finds that functional forms for price and wage equations that imply the possibility of a long-run Phillips curve tradeoff on the whole perform better than those implying no tradeoff. For recent research on the contrary, see Bullard and Keating (1995).Google Scholar
  6. 6.
    Early theoretical explanations of the existence of a short-run Phillips curve include expectational errors (assuming adaptive expectations; Phelps,1967; Friedman, 1968). Under rational expectations, Lucas (1973) argued that a short-run trade-off between inflation and output was the result of signal processing errors of economic agents (confusion by buyers and sellers about the meaning of changes in nominal prices). It could not be exploited for policy purposes. In contrast to this new classical approach, other explanations argue that monetary policy influences real activity because of the existence of sticky wages and prices. Recent research has produced models that explain these nominal rigidities as the outcome of optimizing behaviour, see Ball, Mankiw and Romer (1988); Mankiw and Romer (1991). This line of inquiry also provides evidence that the curvature of the Phillips curve depends on the level of inflation rates (Fischer,1996; Akerlof, Dickens and Perry,1996).Google Scholar
  7. 7.
    Once one of the two is selected as the primary actual operational target, the other cannot be controlled precisely. Only in a non-stochastic system is the setting of price variables the dual of fixing quantities and vice versa (Papademos and Modigliani, 1990; Bryant, Hooper and Mann, 1993).Google Scholar
  8. 8.
    There exists a lively and as yet unresolved debate about the advisability of targeting intermediate variables, see for example M. Friedman (1959, 1968, 1982); Kareken, Muench and Wallace (1973); Waud (1973); B. Friedman (1975, 1977, 1990, 1993, 1994); Bryant (1980, 1983); McCallum (1985, 1990).Google Scholar
  9. 9.
    Thereby implictly assuming that the choice of intermediate target is already made. This choice will be discussed shortly.Google Scholar
  10. 10.
    In the second stage of this procedure, the policy maker aligns his operational target to the desired outcome of the intermediate target. This is realised with the instruments the central bank has at its disposition.Google Scholar
  11. 11.
    See Papademos and Modigliani (1990) and Bryant, Hooper and Mann (1993) for a theoretical discussion and an empirical evaluation, respectively. See also the study by the European Monetary Institute regarding the monetary policy strategy of the ESCB (EMI, 1997).Google Scholar
  12. 12.
    In an inflation-targeting strategy, the monetary policymaker announces a target or a target range for future inflation. A change in the current policy stance is indicated if projected inflation over a medium-term horizon falls outside the announced range. Thus, in contrast to its name, direct inflation targeting is in fact an intermediate targeting procedure (Svensson, 1996).Google Scholar
  13. 13.
    For a discussion of uncertainty in a monetary economy, see Hoogduin (1991).Google Scholar
  14. 14.
    Of course, revision of the strategy is not the first action to be taken by the policy maker. Confronted with persistent differences between projections and actual outcomes, the policy maker will first change his policy instruments. If this is not succesful in aligning projection and outcome, revision of stategy becomes an option.Google Scholar
  15. 15.
    The transmission mechanism is a dynamic process, in that it represents a sequence of events taking place at successive moments in time. The ordering below reflects the timing of this sequence.Google Scholar
  16. 16.
    This issue could have important policy implications. If real rigidities do not vanish in the long run, and monetary policy consequently is non-neutral in the long run, this is a new element in the discussion regarding central bank independence. As this discussion is beyond the scope of this book, we will not take it up here.Google Scholar
  17. 17.
    For an extensive discussion of this channel see, for example, B. Friedman (1983); Bernanke (1983, 1993a,b); Kashyap and Stein (1994); Hubbard (1994b); Bernanke, Gertler and Gilchrist (1996); Friedman and Kuttner (1993); Bernanke and Gertler (1995).Google Scholar
  18. 18.
    This is sometimes denoted as the financial accelerator. It means that because of asymmetric information endogenous cyclical changes in the quality of the balance-sheet structure of borrowers (households and firms) can reinforce themselves, by influencing the development of net-worth and cash-flows of the borrower. See Bernanke, Gertler and Gilchrist (1996).Google Scholar
  19. 19.
    We define money as the financial asset which is used primarily for transactions purposes, and has a rate of return that is exogenously fixed in nominal terms. For convenience, this nominal rate of return is set equal to zero. See Papademos and Modigliani (1990) for details.Google Scholar
  20. 20.
    In other words, arbitrage ensures that, in each period of time, the expected return on the financial instrument that has n periods to maturity equals the return on rolling over (n-1 times) the financial instrument with 1 period remaining to maturity. See Tzavalis and Wickens (1996).Google Scholar
  21. 21.
    Besides changing the relationship between short-term and long-term interest rates, allowing for imperfect substitutability between non-money financial assets requires a multiasset ‘portfolio balance’ specification of asset equilibrium (Tobin, 1961, 1969; M. Friedman, 1959, 1970; Friedman and Schwartz, 1963; Brunner, 1971; Foley and Sidrauski, 1971; Sargent, 1979). This equilibrium determines not a single interest rate, but a vector of rates, representing the yields on bonds, equity and other assets. The transmission of changes in the instrument variables to the ultimate targets operating through this array of imperfect substitutes for money balances is sometimes denoted as the asset price channel (Mishkin, 1995, 1996).Google Scholar
  22. 22.
    In general, pegging the exchange rate is not a substitute for monetary stability and credibility at home. In fact, a peg is only sustainable when it is credible, and credibility is partly determined by domestic policies.Google Scholar
  23. 23.
    Seminal work in this field includes Kydland and Prescott (1977) and Barro and Gordon (1983).Google Scholar
  24. 24.
    New Keynesian economics has evolved from providing microeconomic justifications for the models of the 1960’s to models that explain how small barriers to full and immediate nominal adjustment and the real rigidities in the markets for labour, goods, and credit that cause those small barriers generate large movements in output and employment.Google Scholar
  25. 25.
    For a recent discussion of rigidities in the labour market, see Akerlof, Dickens and Perry (1996).Google Scholar
  26. 26.
    Another manifestation of the imperfect knowledge of the policy maker regarding the transmission mechanism is the use of indicator variables, as discussed in the second section.Google Scholar
  27. 27.
    In addition to these factors, Blinder (1997) argues that the fact that monetary policy decisions in many countries are made by a committee instead of a single person, also contributes to the lags in the effects of monetary policy.Google Scholar
  28. 28.
    The distinction, due to Keynes (1936), between the formation of expectations and the incompleteness of the information on which these expectations are based, is of importance here (see also Hicks, 1974). The latter (a manifestation of fundamental uncertainty) determines the confidence economic agents attach to the former. The lower this confidence, the higher the uncertainty with which agents are confronted. In the face of uncertainty, economic agents may prefer not to commit themselves for too long a period (that is, make irreversible spending decisions), and delay their decisions until the moment they are sufficiently confident to rely on their knowledge (Hoogduin, 1991, p. 65). The existence of uncertainty thus creates a need for liquidity, and money is the most liquid of assets. Note the similarity in reasoning with the recent literature on investment behaviour which stresses the irreversible character of investment expenditure, in combination with the fact that these decisions can be delayed (Pindyck, 1991; Dixit and Pindyck, 1994; Dixit, 1995; Abel, Dixit, Eberly and Pindyck, 1996). According to this line of reasoning, the investment decision has option-like characteristics, and methods developed for pricing options in financial markets can be applied to the theory of investment behaviour. With respect to our study, the implication is that uncertainty regarding future sales and sales price prospects is an important determinant of the investment decision, and also contributes to the variability of lags in the transmission mechanism.Google Scholar
  29. 29.
    Other disadvantages of the structural modelling approach include the incredibility of the identifying restrictions used to obtain equation-by-equation interpretations of structural models (Sims, 1980). Moreover, it may not always be possible to construct these models, due to lack of data or prohibitive costs. Structural models are relatively expensive to construct and to maintain, and the estimation may entail computational inefficiencies (Van Loo, 1984; Knot, 1996). Another problem with large structural models is that they are so complex, and there are so many specific interactions, that the study of such models does not contribute much to our understanding of specific economic mechanisms.Google Scholar
  30. 30.
    However, see Basmann (1972) and Kloek (1988) for a critical assessment of the role played by theory in macroeconomic structural models.Google Scholar
  31. 31.
    Examples of the use of the structural modelling approach in the context of the transmission mechanism include Mauskopf (1990), Bryant, Hooper and Mann (1983), Smets (1995) and Ball (1997).Google Scholar
  32. 32.
    We acknowledge that we do not provide a complete discussion of all the chains in the transmission mechanism. Our analysis is therefore partial by definition.Google Scholar
  33. 33.
    Recent work along similar lines include Aizenman and Frenkel (1986); Frankel and Chinn (1991) and Asako and Wagner (1992).Google Scholar

Copyright information

© Springer Science+Business Media Dordrecht 2001

Authors and Affiliations

  • Jan Marc Berk
    • 1
    • 2
  1. 1.De Nederlandsche BankAmsterdamThe Netherlands
  2. 2.Free UniversityAmsterdamThe Netherlands

Personalised recommendations