RETHINKING ENDOGENOUS MONEY: A CONSTRUCTIVE INTERPRETATION OF THE DEBATE BETWEEN HORIZONTALISTS AND STRUCTURALISTS

1 Metroeconomica 55:4 (2004) RETHINKING ENDOGENOUS MONEY: A CONSTRUCTIVE INTERPRETATION OF THE DEBATE BETWEEN HORIZONTALISTS AND STRUCTURALISTS Giusep...

Author:  Ashlee Leonard
0 downloads 0 Views 127KB Size

Metroeconomica 55:4 (2004)

367–385

RETHINKING ENDOGENOUS MONEY: A CONSTRUCTIVE INTERPRETATION OF THE DEBATE BETWEEN HORIZONTALISTS AND STRUCTURALISTS Giuseppe Fontana* Leeds University Business School (June 2001; revised November 2002)

ABSTRACT Beyond a widespread agreement on the idea that ‘loans create deposits’ and ‘deposits make reserves’, there is much controversy in the endogenous money literature over the workings of the reserve market, the credit market and the financial markets. In this paper a constructive interpretation of the debate between horizontalists and structuralists is suggested and their arguments are taken forward by showing that these controversial issues can be explained rigorously once a single-period–continuation framework is adopted.

1. INTRODUCTION

One of the main tenets of Post Keynesian economics is that money is endogenous, meaning that the supply of money is the outcome of purposeful interactions between economic agents. More precisely, the endogeneity of money relies on the following causation: the supply of money is determined by the demand for credit (bank loans), and the latter originates within the system to finance the production process or the upsurge of speculative purchases.

* The final version of the paper was written when the author was visiting research scholar at both C-FEPS and the Economics Department, University of Missouri-Kansas City, Kansas City (USA). The author would like to express appreciation to members of those institutions, to John Henry (Visiting Professor in the Economics Department) and to David Foster (English Department) for providing a stimulating and pleasant working environment. The author would also thank Philip Arestis, Victoria Chick, Paul Dalziel, Augusto Graziani, Peter Howells, Marc Lavoie, Basil Moore and Tom Palley for comments and suggestions on previous drafts of the paper. Finally, the author owes a special debt to Malcolm Sawyer, who was never short of praise and encouragement during and after completion of the PhD research on the subject. The usual disclaimer applies. © Blackwell Publishing Ltd 2004, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

368

Giuseppe Fontana

The main implication of this theory, then, is that money is never neutral. Money is needed for, and is the purpose of, financing the core activities of capitalist economies. In a nutshell, money is the wheel of trade and growth of modern economies (Moore (1988a)). While these propositions are now widely accepted by most, if not all, Post Keynesian economists, and more generally heterodox economists, there are still several details in the theory of endogenous money that are contentious.1 The debate between what have now been called horizontalists and structuralists usually centres on the following three arguments. First, there is disagreement over the degree of accommodation by central banks to the demand for reserves of banks. Are central banks always willing to supply the required reserves at the going short-term nominal interest rate? Or could they resist by changing this rate? Second, there is an enduring discussion about the meaning and relevance of the liquidity preference of banks. Is liquidity preference consistent with endogenous money? And, if so, does this mean that there is an upward-sloping curve for bank loans? Third, there is a controversy over the implication of the liquidity preference of the non-bank public sector. Are the preferences of the subsequent recipients of bank deposits (e.g. wage earners) necessarily consistent with the preferences of the first recipients of those deposits (e.g. firms)? And, if not, is there a mechanism that reconciles the different preferences? Since there is widespread agreement that money is endogenous some Post Keynesian economists have played down the relevance of these questions as ‘a tempest in a teapot’ (Moore (1991a, p. 405)). But there is a big issue at stake. After decades of complete lack of attention, monetary policy has now emerged as one of the most critical responsibilities of modern governments. The received view is that monetary policy is the most flexible tool for achieving stabilization objectives. Driven by growing empirical evidence, leading monetary economists like Svensson (1999), Taylor (1999) and Woodford (1996) have started to acknowledge the endogeneity of the level of reserves and credit (Fontana (2002), Goodhart (2002)). Similarly, in recent years central bankers have experienced more responsibilities in the area of macroeconomic policy. But monetary policy is a very difficult art. As Sir John Hicks often argued, ‘monetary theory is ‘in history’: it is influenced by the course of events in the ‘real world’, in a way in which other departments of economic theory are less influenced, or less continually influenced’ (Hicks (1982a, p. xii)). Monetary policy requires a sound analysis of the way the

1

For critical surveys of the post-Keynesian theory of endogenous money, see Cottrell (1994), Dalziel (2001, ch. 3), Fontana (2003b), Hewitson (1995), Howells (1995) and Rochon (1999).

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

369

reserve market, the credit market and the financial markets really work. This has been the dedicated task of Post Keynesian economists in the last three decades, and it would be very unfortunate if the opportunity to constructively engage with leading monetary theorists and central bankers is missed. The objectives of this paper are twofold. The first objective is to discuss the controversial issues debated by horizontalists and structuralists. It is argued that these two approaches to endogenous money provide very insightful perspectives on the way central banks, banks, firms, financial intermediaries and wage earners enter into the money supply process. Drawing on Palley’s model (1994, 1996a), similarities and differences between these approaches are made explicit by simple graphical analysis. The second objective is to show that Hicks’s distinction between single period theory and continuation theory (Hicks (1982b, p. 223)) can be used to explain the limits to the domains of relevance of the horizontalist and structuralist approaches to endogenous money (Fontana (2003a)). Horizontalists like Moore (1988b, Part 1) and Lavoie (1992, ch. 4) have shown that money is integrated within the economy through the core activities of capitalist economies. The creation of money originates in the production of new bank liabilities during the process of income expansion. But this perspective needs to be complemented by a more detailed analysis of the monetary stance of central banks, the liquidity preference of banks and the liquidity preference of the different holders of bank deposits. Structuralists like Arestis and Howells (1996), Dalziel (1996), Dow (1996), Goodhart (1991), Palley (1987) and Wray (1990) have advanced interesting ideas on these issues that can provide the basis for a more general theory of endogenous money. The paper is in three sections. Section 2 presents the controversial issues between horizontalists and structuralists. The discussion is arranged in terms of contentious arguments related to the working of the reserve market, the credit market and the financial markets, respectively. Section 3 takes the arguments forward by showing that, once a single-period–continuation framework is adopted, the issues can be rigorously explained. Section 4 concludes.

2. CONTROVERSIAL ISSUES

The core argument of endogenous money theory is that the supply of money is the outcome of purposeful interactions between economic agents. At the minimum, there are three markets (for monetary reserves, credit, and securities), and four types of economic agents—i.e. commercial banks (banks for short), firms, wage earners and a central bank—that give content to this

© Blackwell Publishing Ltd 2004

370

Giuseppe Fontana

proposition. In the following, the debate between horizontalists and structuralists is therefore presented in terms of the controversial arguments surrounding the behaviour of the economic agents in the reserve market, the credit market and the financial markets, respectively. Section 2.1 debates the way central banks set the short-term nominal interest rate and accommodate the demand for monetary reserves. In section 2.2 I discuss the liquidity preference of banks and its effects on the base rate and the supply of loans. Finally, in section 2.3 I consider the problems arising from the fact that in the standard literature the demand for loans originates with firms while the deposits created are held by wage earners.

2.1

The reserve market

The first controversy between horizontalists and structuralists is over the relationship between banks and the central bank. The standard Post Keynesian view is that central banks set the short-term nominal interest rate (e.g. the federal funds rate in the USA and the repo rate in the UK) and supply monetary reserves on demand. The short-term nominal interest rate, then, is a control instrument used by central banks to affect the lending activity of banks and, in this way, the entire economic process. For instance, changes in the short-term nominal interest rate prompt banks to modify base rates (e.g. the prime rate, the personal loan rates, the mortgage rates) at which they lend to their customers. These rates, then, have a pivotal influence on investment and on the level of effective demand, which in turn affects the volume of output and employment. The differences between the two approaches to endogenous money can be introduced in terms of a short-run reaction function measuring the elasticity of the short-term nominal interest rate with respect to changes in the demand for reserves. Horizontalists argue for an infinitely elastic reaction function in the time period between revisions of the short-term nominal interest rate (e.g. Moore (1991a, 1995)), whereas structuralists defend a less than perfectly elastic function (e.g. Pollin (1991)). Figure 1 shows the contentious description of the monetary reserves market. The focus of the analysis is upon changes in the supply of money and how those changes arise from the new flow of bank loans to borrowers. The four-panel diagram is derived from Palley (1994, 1996a).2 The upper left panel describes the monetary reserves market. The supply of reserves is 2

Besides Palley, the author is indebted to Dow (1996, 1997), Howells (1995), Lavoie (1996), Pollin (1996) and Sawyer (1996).

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

371

Interest rate

R

S

r1 r0

Fixed Mark-up

A

L0S

i1 L0D

i0 Monetary reserves

R1

L1S

B

R0

L0

L1D L1

Bank Loans

D0

D1

LD line

DR line Bank Deposits Figure 1. An encompassing endogenous money analysis of the reserve market.

represented by a step function, with each horizontal segment representing a different interest rate target (e.g. i0, i1). The horizontal parts of the schedule thus show accommodative behaviour on the part of the central bank, while the upward trend (from right to left in the diagram) reflects the structuralist view that central banks have a less than perfectly elastic reaction function. The upper right panel shows the credit market where firms and banks negotiate the supply of and demand for bank loans. Since the debate over the slope of the supply curve of bank loans is postponed to the next section, that curve is here represented by a perfectly elastic schedule at a base rate (e.g. r0) determined as a fixed mark-up over the short-term nominal interest rate (e.g. i0) set by the central bank. The demand for bank loans (e.g. L0D) is a decreasing function of the base rate (r) and, together with the supply of bank loans (e.g. L0S), determines the total volume of credit (e.g. L0). The lower panels are used to describe the two main insights of the endogenous money theory, ‘loans create deposits’ (LD line) and ‘deposits make reserves’ (DR line), respectively. The equilibrium in the credit market determines via the loans–deposits (LD) line the supply of new bank deposits (e.g. D0) in the lower right panel. Note that the LD line represents the balance sheet constraint of banks and, to make the graphical exposition feasible, it is drawn on the assumption that banks hold their liabilities, like time or

© Blackwell Publishing Ltd 2004

372

Giuseppe Fontana

demand deposits, in a fixed proportion. The supply of monetary reserves (e.g. R0) associated with the new supply of bank deposits (e.g. D0) is shown via the deposits–reserves (DR) line in the lower left panel. The DR line represents the total demand for reserves (e.g. Palley (1994, p. 72, eqn 17)). The four-panel diagram illustrates the underlying sequential analysis that characterizes the endogenous money theory (Dalziel (2001, ch. 4)) as well as the controversial issues related to the reaction function of the central bank. Expansionary shifts of the demand for bank loans (e.g. L1D) cause via the LD line and the DR line increases in the level of bank deposits (e.g. D1) and of reserves (e.g. R1), respectively. But, as a result of the new level of reserves, the central bank might decide, though it does not need to, to tighten the reserve market (e.g. moving to an i1 interest rate policy). This change in the policy stance of the central bank, then, is likely to affect the lending policy of banks in the credit market (e.g. L1S). This simple example suggests that central banks have a very active role in the money supply process. By adjusting the short-term nominal interest rate, they may be able to affect lending conditions in the credit market. This power of central banks is in general recognized by both horizontalists (e.g. Lavoie (1992, pp. 186–9)) and structuralists (e.g. Howells (1995, pp. 12–17)). Their main difference lies in the assumptions regarding the state of expectations of central banks during the money supply process. Horizontalists use the supply curve of reserves associated with a constant state of expectations, whereas structuralists allow for the effects of changes in the state of expectations. Therefore, while the former prefer to discriminate between different stances of monetary policy and focus only on the freely managed short-term nominal interest rate stance (Moore (1988b, p. 265, n. 9), (Lavoie (1996, p. 279)), the latter are more inclined to consider complex reaction functions of central banks (Wray (1992, p. 307), Palley (1996a, pp. 592–3)). In terms of figure 1, by the particular time nature of their models, structuralists are prone to consider the overall upward-sloping step function representing the supply of reserves (i.e. RS), whereas horizontalists focus on each single horizontal part of it (i.e. either the i0 or i1 policy line).

2.2

The credit market

A deeper-going argument between horizontalists and structuralists is over the behaviour of banks in the credit market. Whether or not monetary reserves are forthcoming at a constant short-term nominal interest rate, structuralists hold that, as a result of an increase in the lending activity, price and nonprice terms of credit will rise. Price terms are base interest rates like the prime

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

373

rate where non-price terms mainly refer to income and assets collateral requirements (Wolfson (1996, pp. 456–7)). Drawing on Minsky’s analysis of corporate financial behaviour (Minsky (1975, chs 5 and 6)), most structuralists argue that banks raise base interest rates at the peak of the business cycle (Wray (1995, pp. 278–80)).3 As lending grows, banks become increasingly concerned about their own portfolio balance (usually measured by the ratio of loans to equity and the ratio of loans to safe assets) as well as the liquidity level of their customers (usually indicated by the ratio of debt to equity of firms). Similarly, structuralists maintain that, in this case, banks often impose quantitative restrictions on their lending activity. They conclude that if price and non-price terms are properly considered, the loans supply curve is best represented by an upwardsloping curve rather than a horizontal curve as is argued by horizontalists (Dow (1996, pp. 498–504)). On their part, horizontalists argue for a horizontal loans supply curve in the interest–bank loans space. However, they acknowledge that banks may impose quantitative restrictions on their customers (Moore (1988b, p. 24)). Similarly, horizontalists accept that the liquidity ratios of banks and customers play a role in determining the base rates over the business cycle. However, they object to the structuralists’ contention that the supply of loans is necessarily upward sloping in the long run (Lavoie (1996, pp. 286, 289)). Horizontalists prefer to discuss the effects of changing liquidity ratios in terms of initial restrictions on the borrowing activity of customers. Banks, they argue, do not curtail credit by marginal variations of the markup, though they do change over time the requirements for the identification of sound customers (non-price terms for new loans) and the base rates of their credit offer (price terms for new loans). Therefore, at all times banks only accommodate the solvent demand for loans. More importantly, the supply of loans is a truncated horizontal line: ‘beyond some point, the supply simply vanishes’ (Lavoie (1996, p. 288); also Moore (1991b, p. 126)). Changed conditions in the credit market, then, are best represented by a shift in the demand curve and a new horizontal supply curve. Figure 2 shows the differences between the horizontalist and structuralist analyses of the credit market. The four-panel diagram is again adopted. The significant difference from figure 1 is the assumption of a perfectly elastic

3

Recently, some structuralists have accepted that this need not necessarily be the case (Howells (1995, p. 20)). For instance, they acknowledge the point made by Lavoie (1996, pp. 285–90) to the effect that over the business cycle loans are being taken out, profits earned and loans repaid (out of profits and out of borrowing) such that the ratio of loans to profits or to equity need not necessarily rise during the business upswing.

© Blackwell Publishing Ltd 2004

374

Giuseppe Fontana Interest rate

r1 r0

A L0D

i0 Monetary reserves

R1

R0

L0

L0S

B L1D L1

Bank Loans

D0

D1

LD line

DR line Bank Deposits Figure 2. An encompassing endogenous money analysis of the credit market.

schedule for the supply of reserves, meaning that only a single monetary policy stance is considered (e.g. an i0 interest rate policy). More importantly, the loan supply schedule is now a function of the liquidity ratios of banks and their customers. During an economic expansion banks are most likely going to experience a reduction in the level of liquidity. Illiquidity comes from more risky new loans and from outstanding loans being perceived as more risky. As the peak of the cycle is approached, some banks become aware of the objective fragility of the system and anxious about the illiquidity of their balance sheets. They are then likely to tighten the requirements for new credit and to raise their base rates (e.g. r1). Similarly, as customers take on more debt, banks become concerned about the solvency of borrowers. As in the previous case, it is likely that banks will revise their requirements upward and raise the base rates (e.g. r1). Thus, in these circumstances, the supply of bank loans (L0S) is better represented by a step function. Banks set the base rates (e.g. the prime rate) and this determines the height of the loans supply curve (i.e. the relevant horizontal line of the L0S). Their perception of the state of the economy explains the length of the horizontal parts of the curve, i.e. how long banks hold constant the supply price of credit (Fontana (2003b)). In short, the main difference between horizontalists and structuralist lies in the different assumptions about the behaviour of banks in the credit

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

375

market. Horizontalists look at the credit market with the assumption that during the money supply process banks are not affected by changes, if any, in their own liquidity ratios and the liquidity ratios of their customers. Structuralists allow for the possibility that over the business cycle banks revise their non-price and price terms of credit.

2.3

The financial markets

Another controversy between horizontalists and structuralists is related to the relationship between the different recipients of credit. In the standard analysis of endogenous money, the demand for loans mainly originates with firms, while the deposits created by this lending are eventually held by wage earners. Firms are deficit units that are involved in income–expenditure decisions. They negotiate with banks the amount of credit necessary for purchasing capital and labour services, and once collateral requirements are satisfied they receive the resulting deposits. However, those deposits are used to pay the owners of inputs. If transactions between firms are ignored, i.e. if purchasing of capital services is considered an internal transaction of the firms sector, labour services are the only inputs to buy. The supply of new deposits, then, is equal to the flow of payments from firms to wage earners. Wage earners use these deposits to buy commodities in the goods market and securities in the financial markets. In the simple case in which the public sector and the foreign sector are ignored, firms issue all securities available in the financial markets. Therefore, deposits spent in the goods market as well as those spent in the financial markets return to firms and can be used for repaying the initial debts. This is what in the literature has been labelled the Kaldor–Trevithick reflux mechanism (Kaldor and Trevithick (1988)) and is used by horizontalists to explain how ‘excess’ deposits for wage earners are extinguished from the monetary circuit (e.g. Lavoie (1999, pp. 105–8)). Structuralists usually acknowledge the importance of the Kaldor– Trevithick reflux mechanism in the credit supply process (e.g. Arestis (1988, p. 65)). However, they argue that the reflux mechanism does not automatically extinguish all newly created deposits (Cottrell (1986, p. 17), Palley (1991, p. 397), Chick (1992, p. 205), Dalziel (2001, p. 144, n. 2)). Wage earners save part of those deposits out of precautionary or speculative motives. The consequent allocation of deposits between securities and liquid balances is a portfolio choice and cannot be divorced from changes in interest rate differentials (Arestis and Howells (1996, pp. 540–4)). Structuralists would then conclude that these changes are part of a complex process of portfolio

© Blackwell Publishing Ltd 2004

376

Giuseppe Fontana

adjustments, some details of which are important in the transmission of monetary effects. For example, the base rates at which banks supply deposits are affected by these changes in interest rates. The different views on the ultimate impact of an expansion in the supply of loans lies in the particular assumptions about the behaviour of firms and wage earners in the financial markets and on the feedback effects from those markets to the credit market (Palley (1996a)). The working of the credit market, where the supply of deposits and the base rate are decided, is strictly related to the working of the labour market, where the price and quantity of labour services are negotiated. Firms simultaneously contract the level of money wages with wage earners and the amount of credit with banks. The behaviour of firms in the credit market is thus affected by the behaviour of wage earners in the labour market. Certainly, from this perspective wage earners are not passive recipients of the flow of new deposits. The process of credit creation unfolds as firms trade IOUs for deposits and wage earners exchange labour services for deposits. On the other hand, when wage earners decide how to spend this new flow of deposits, they may wish to hold some of it in the form of liquid balances, e.g. stock of deposits. At once, as a result of the income-generation process, banks supply deposits to wage earners and, as a consequence of their portfolio choice, wage earners ‘supply’ deposits to banks. Horizontalists believe that in this case it is misleading to talk of an independent supply–demand relation. The demand for deposits is necessarily equal to the supply of deposits. In fact, the demand for deposits is simply the supply of deposits already in existence (Moore (1988b, p. 298; 1997, pp. 425–7)). Structuralists reject this conclusion. They hold that the demand for deposits must equal the supply of deposits (Palley (1991)). But this equality is brought about by portfolio (and income) adjustments that change both interest rates and the supply of deposits. More precisely, structuralists discriminate between the final demand for deposits and the credit counterpart supply of deposits (Goodhart (1989, pp. 32–3)). The latter is part of the flow of purchasing power created in the credit market, while the former is the residual stock of liquid balances resulting from portfolio adjustments in the financial markets. Of course, ex post the stock of deposits willingly held by wage earners corresponds to the existing supply of deposits. In fact, that amount of deposits measures the outstanding debt of firms to banks and, once new loans are granted, it also measures the increase over time in the stock of money (Kahn (1954, p. 238)). However, structuralists argue that for this ex post condition to hold (1) important changes in interest rate differentials have to take place and (2) these changes have feedback effects on the future level of base rates and the supply of loans (Palley (1991, p. 397), Dalziel (2001, p. 144, n. 2)).

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

377

Financial markets, then, are the place where these changes take place. Interest rate differentials change as a result of wage earners allocating deposits between different securities and liquid balances. These changes are eventually bound to affect future negotiations between banks and firms in the credit market. In particular, the supply of loans and the base rates are affected by the history of repayments for previous debts (Wolfson (1996, pp. 458–61)). Similarly, the demand for loans is determined by the needs of firms to renegotiate loans equal to the net stock of money in addition to any credit necessary for new production (Fontana (2000, pp. 35–6)). In short, horizontalists have examined the two-way relationship between the credit market and the financial markets with the assumption that the ultimate impact of an expansion in the supply of loans has no effect whatsoever on the expectations of wage earners’ allocating savings to financial assets (including a stock of liquid balances) in a certain proportion. Structuralists have considered the possibility of those expectations changing as a result of the supply of new deposits. How current portfolio adjustments in the financial markets affect future conditions in the credit market is of the utmost importance in their understanding of the money supply process. Nevertheless, once the different nature of the assumptions is made explicit, the two perspectives fit in with each other as part of a more general theory of endogenous money. ‘If it is agreed that ex ante discrepancies [between demand for loans and demand for deposits] are resolved by repayment of loans but that this involves a circulation of deposits that triggers relative price signals, then indeed there is nothing between this version of the reflux [by horizontalists] and our position’ (Arestis and Howells (1999, p. 118); also Cottrell (1988, p. 296)).

3. A GENERALIZED THEORY OF ENDOGENOUS MONEY

The foregoing account of the reserve market, the credit market and the financial markets suggests that horizontalists and structuralists have in mind two distinct classes of models of the money supply process. These models share the same methodological and theoretical framework, but they differ in terms of the particular assumptions made about the state of expectations of agents involved in the economic process. The purpose of this section is to give precise meaning to this idea. The Hicksian distinction between a singleperiod and a continuation theory of money is thus used to explain rigorously the limits to the domains of relevance of the horizontalist and structuralist approaches to endogenous money. This argument is offered fully in the spirit of recent contributions by Palley (1996a, 1998). The final aim is thus to

© Blackwell Publishing Ltd 2004

378

Giuseppe Fontana

encompass these approaches into a generalized theory of endogenous money.4 Horizontalists and structuralists concur that the general aim of the endogenous money theory is to propose an organized and orderly method of investigation into the complex questions of creation, circulation and destruction of money. They recognize that calendar time normally elapses between the moment in which central bankers, banks, firms and wage earners make decisions and the ultimate outcome of those decisions. During this time, disappointment or new opportunities play a central role in shaping and constraining rational behaviour. Accordingly, agents continuously revise expectations and plans for the future course of events. Having acknowledged the relevance of calendar time and expectations, horizontalists and structuralists differ in providing alternative specifications about the effects of expectations on the money supply process. In particular, horizontalists and structuralists believe that, although expectations are a necessary component of a sound monetary theory, different assumptions could be made about the constancy of the expectations and their influence on the working of the reserves market, the credit market and the financial markets. From this perspective, horizontalists have proposed what along Hicksian lines could be labelled a single-period theory of endogenous money, whereas structuralists have proposed a continuation theory of endogenous money (Hicks (1982b, p. 223)). A single period is the minimum effective unit of economic time for the analysis of agents involved in the economic process. The length of this period is such that changes in expectations never occur within it but rather at the junctions of one single period to the next. A single-period theory of endogenous money, then, is built on the simple assumption that the state of expectations of central banks, banks, firms and wage earners is given and constant. This assumption allows the specification of simple and stable functional relationships that continuously changing expectations would have made impossible. Furthermore, the idea of assuming that the state of expectations of agents is given and constant is a legitimate extension of the basic methodology used by Keynes in the General Theory (Kregel (1988, pp. 45– 52), Gerrard (1997, pp. 193–7)). It is a realistic attempt to specify fundamental relationships without ignoring the possibility that changes in the state of expectations may shift the behavioural functions of agents. For example, while considering relevant the expectations of the central bank and the 4

For a discussion of the encompassing principle as an appropriate characterization of the postKeynesian way of thought, see Fontana and Gerrard (2002). Recent exemplars of the encompassing principle in practice are Fontana and Palacio-Vera (2002, 2003) and Palley (1996b).

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

379

difference between expected and actual level of inflation in influencing these expectations, horizontalists hold in abeyance their effects in order to explain the nature of the link between the reserve market and the credit market. A similar argument is used for the purpose of defining without ambiguity the sequential process that relates the demand for credit of firms, via the creation of bank deposits, to the portfolio choice of wage earners. In this case, the focus of the analysis is on the nature of the nexus between the credit market and the financial markets. Thus, the single-period analysis proposed by horizontalists is an important contribution to a generalized theory of endogenous money. It provides a framework in which the role of economic agents in the money supply process can be precisely defined. Specifically, the logical point that in modern economies ‘loans create deposits’ and ‘deposits make reserves’ is made absolutely clear. At the same time, horizontalists need to allow plenty of space for qualifications, adjustments and development of their results. Expectations are normally changing over time, and this is the main shortcoming of their approach. More importantly, these changes are the natural outcome of the complicated sequences of events that result in the creation of money. Figures 1 and 2 have shown that the ability of horizontalists to shed light on some monetary issues is limited. The possibility that, over the business cycle, banks may revise price and non-price terms or that central banks may adopt new monetary stances has no place in their models. This should not come as a surprise. The formal features of a single period narrow the issues that can be investigated with such a framework. As explained by Hicks, ‘the end of the one period is to be distinguished from the beginning of the next by the change in expectations which may occur when this same instant of time “puts on its other hat” ’ (Hicks (1985, p. 94)). Once a single-period framework is adopted, the choice is made: little change will occur within it. Expectations can be disappointed, but their effects are not allowed to alter the current course of events. Any shift in the expectational functions of agents has to wait for the next period. As a result of recent criticisms (e.g. Palley, 1996a), horizontalists now acknowledge the limitations deriving from the formal features of their approach. According to Moore, their view of the money supply process ‘is taken to refer only to the immediate current market period’ (Moore (1998, p. 175); also Palley (1996a, p. 585, n. 1)). But there is an evident problem with this view. Having adopted a single-period framework, horizontalists record the effects of changing expectations at the beginning of the following period. How the end of a single period relates to the next one is of the utmost importance if a complete analysis of the endogenous supply of money is to be carried out. The actual path followed by the sequence of activities

© Blackwell Publishing Ltd 2004

380

Giuseppe Fontana

that describes the money supply process is explained by the interactions between what agents plan to do and what they discover they ought to have planned to do. There are interesting things to be learnt when expectations are allowed to affect the course of the process of creation of money. This is truly the primary purpose of the continuation theory of money, which is concerned with the effects of the events of a period upon the expectations that determine the events of the following periods. A continuation theory, in fact, is the analysis of the dynamics of a sequence of single periods. It deals explicitly with the linkages between successive periods, and those linkages are an essential step in moving beyond the boundaries of selfcontained periods. It is here that structuralists have directed their most convincing efforts. The framework of a continuation theory explicitly allows for the fact that the general state of expectations may change in the light of realized results. In a continuation analysis, the interaction between state of expectations and realized results is modelled as an ever-evolving process. Inconsistencies between plans of agents come to the front of the analysis as all sorts of mechanisms to reconcile them. There are likely to be movements along the demand and supply curves at the same time that these curves will themselves be shifting as a result of changes in the state of expectations. For example, if central bankers realize that the actual proceeds of monetary policy are not what they had expected, they will attempt to do something before it is too late (e.g. Palley (1996a, pp. 589–92)). As their expectations interact with the realized level of demand for monetary reserves, the short-term nominal interest rate is likely to change to reflect the new conditions in the economy. The base interest rates would then be affected, as would be the demand for loans and the holding of deposits. Thus, the new aggregate supply of reserves would be continuously adjusting to the conditions in the credit market and the financial markets. Policy reactions from the reserve markets would finally feed back to those markets, creating a complex network of interactions between all agents involved in the money supply process. Those interactions, policy reactions and feedback are an important feature of a continuation analysis and a major difference from a single-period analysis. Keeping with the same example, the latter would show that demand and supply conditions in the credit market affect the initial price and non-price terms in the reserve market. A single period would then continue for a sufficient length of time such that the credit creation works itself out completely. During this period, central banks may be disappointed by the result of their policy, or banks may experience new opportunities or unexpected problems. Nevertheless, the formal features of the single period imply that disappointment or new opportunities would not have an effect on the state of current

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

381

expectations. It is only in the next period that the reserve market and the credit market would record new demand and supply conditions. Compare that with the continuation analysis, in which expectations are constantly shifting as a result of the market conditions. In this case, a new state of expectations is superimposed before the previous change has fully worked itself out. At any time, the money supply process is engaged with a number of overlapping events, the existence of which is due to different states of expectations. The extreme dynamic complexity of such a situation explains why horizontalists have adopted a single-period analysis that allows for a lucid presentation of the endogenous money theory (Moore (1998)). The same complexity also explains the practical need for structuralists to often rely on comparative statics exercises rather than full-blown dynamic experiments. The very concept of the money supply process is difficult to understand in these circumstances. The sequence of events is rather fuzzy because the effects of current events are mixed with the ongoing effects of previous events. In this case, in order to understand the succession of actions that describe the money supply process, the same pattern of actions has to recur a sufficient number of times. The effects of shifting expectations, as opposed to the case of constant expectations, in fact require a bit of time in order to work (Chick (1990, p. 62)). At the same time, the particular assumptions underlying a single-period analysis explain the need for rethinking its results in terms of a continuation analysis. It is only through a continuation analysis that issues like shifting monetary policies, liquidity preference or the credit–deposits nexus can be fully investigated. There is a simple and stark conclusion from the comparison of a singleperiod analysis with a continuation analysis. The horizontalist approach, which relies on the former, and the structuralist approach, which is grounded on the latter, are very different theories when it comes to the constancy of expectations and their influence on the working of the money supply process. But together these approaches provide a more general theory of endogenous money. The crux of the matter is that a single-period analysis needs to be carried out before proceeding to a continuation analysis, for the latter would be most beneficial when the former has laid down the basic functional relationships to be scrutinized. The horizontalist and the structuralist approaches to endogenous money are thus complementary rather than antithetical ways of analysing the money supply process. Finally, these two approaches are a legitimate development of Keynes’s ideas (Fontana (2001)). The different assumptions about the state of expectations do not refer to divergent descriptions of the economic process. From this point of view, horizontalists and structuralists are both providing a realistic description of what happens in modern economies. Using different

© Blackwell Publishing Ltd 2004

382

Giuseppe Fontana

assumptions is a way of capturing the variety of views that agents have about the prevailing features of real-world money supply processes. As explained by Keynes ‘it is not the economy under observation which is moving in the one case and stationary in the other, but our expectations of the future environment which are shifting in the one case and stationary in the other’ (Keynes (1973, p. 511)). Again, horizontalists and structuralists ought to be seen as working on the same project, a generalized theory of endogenous money, though the strategy adopted to carry it out is slightly different.

4. CONCLUSIONS

The core of endogenous money theory is that the supply of money in modern economies is determined by the demand for credit (bank loans) and that this, in turn, responds to the need for financing production or speculative purchases. Beyond a widespread agreement over the idea that ‘loans create deposits’ and ‘deposits make reserves’, there is much controversy. Horizontalists and structuralists have now debated for a long time the key issues related to endogenous money. Do central banks accommodate the demand for reserves at the going short-term nominal interest rate? Does the supply of loans slope upward? Do wage earners make portfolio choices that affect the future availability of credit? In this paper I have suggested a constructive interpretation of the debate between horizontalists and structuralists that takes their arguments forward by showing that there is a time framework in which those approaches to endogenous money can be made compatible. This time framework is general enough to be used for analysing specific institutional settings or specific historical instances. The disagreement between horizontalists and structuralists arises from the particular assumptions made about the general state of expectations of economic agents. Horizontalists rely upon a single-period framework that is built on the assumption that the state of expectations of all agents involved in the money supply process is given and constant. This assumption allows the specification of stable functional relationships that continuously changing expectations would make very laborious. On the other hand, structuralists depend on a continuation framework that explicitly takes account of the fact that the state of expectations of agents may change in the light of realized results. In this way, structuralists are able to tackle controversial issues related to shifting monetary policies, liquidity preference and the credit–deposits nexus that are ignored in the horizontalist approach. The conclusion of this paper is that a single-period analysis is to be carried out before proceeding to a continuation analysis, for the latter is of utmost

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

383

value when it draws on the basic monetary relationships laid down by the former. Thus, the horizontalist approach to endogenous money is not antagonistic to the structuralist one. Together they form a more general theory of endogenous money. REFERENCES Arestis P. (1988): ‘Post Keynesian theory of money, credit and finance’, in Arestis P. (ed.): PostKeynesian Monetary Economics, Edward Elgar, Aldershot. Arestis P., Howells P. (1996): ‘Theoretical reflection on endogenous money: the problem with convenience lending’, Cambridge Journal of Economics, 20, pp. 539–51. Arestis P., Howells P. (1999): ‘The supply of credit money and the demand for deposits: a reply’, Cambridge Journal of Economics, 23, pp. 115–19. Chick V. (1990): ‘On the place of Value and Capital in monetary theory’, Greek Economic Review, 12, Supplement, pp. 53–71. Chick V. (1992, orig. 1986): ‘The evolution of the banking system’, in Arestis P., Dow S. (eds): On Money, Method, and Keynes, Selected Essays of Victoria Chick, Macmillan, London. Cottrell A. (1986): ‘The endogeneity of money and money–income causality’, Scottish Journal of Political Economy, 33, pp. 2–27. Cottrell A. (1988): ‘The endogeneity of money: reply’, Scottish Journal of Political Economy, 35, pp. 294–7. Cottrell A. (1994): ‘Post-Keynesian monetary economics’, Cambridge Journal of Economics, 18, pp. 587–605. Dalziel P. (1996): ‘The Keynesian multiplier, liquidity preference and endogenous money’, Journal of Post Keynesian Economics, 18, pp. 311–31. Dalziel P. (2001): Money, Credit and Price Stability, Routledge, London. Dow S. C. (1996): ‘Horizontalism: a critique’, Cambridge Journal of Economics, 20, pp. 497–508. Dow S. C. (1997): ‘Endogenous money’, in Harcourt G. C., Riach P. A. (eds): A ‘Second Edition’ of The General Theory, Routledge, London. Fontana G. (2000): ‘Post Keynesians and circuitists on money and uncertainty: an attempt at generality’, Journal of Post Keynesian Economics, 23, pp. 27– 48. Fontana G. (2001): ‘Keynes on the ‘nature of economic thinking’: the principle of nonneutrality of choice and the principle of non-neutrality of money’, American Journal of Economics and Sociology, 60, pp. 711–43. Fontana G. (2002): ‘The making of monetary policy in endogenous money theory: an introduction’, Journal of Post Keynesian Economics, 24, pp. 503–9. Fontana G. (2003): ‘Post Keynesian approaches to endogenous money: a time framework explanation’, Review of Political Economy, 2003, pp. 291–314. Fontana G. (2004): ‘Hicks on monetary theory and history: money as endogen money’, Cambridge Journal of Economics, 2004, pp. 73–88. Fontana G., Gerrard B. (2002): ‘The encompassing principle as an emerging methodology for post Keynesian economics’, in Arestis P., Desai M., Dow S. C. (eds): Money, Microeconomics and Keynes, Essays in Honour of Victoria Chick, vol. 2, Routledge, Cheltenham. Fontana G., Palacio-Vera A. (2002): ‘Monetary policy rules: what are we learning?’, Journal of Post Keynesian Economics, 24, pp. 547–68. Fontana G., Palacio-Vera A. (2004): ‘Monetary policy uncovered: theory and practice’, International Review of Applied Economics, pp. 25–41. Gerrard B. (1997): ‘Method and methodology in Keynes’s General Theory’, in Harcourt G., Riach P. (eds): A ‘Second Edition’ of The General Theory, Routledge, London.

© Blackwell Publishing Ltd 2004

384

Giuseppe Fontana

Goodhart C. A. E. (1989): ‘Has Moore become too horizontal?’, Journal of Post Keynesian Economics, 12, pp. 29–34. Goodhart C. A. E. (1991): ‘Is the concept of an equilibrium demand for money meaningful? A reply to “has the demand for money been mislaid?” ’, Journal of Post Keynesian Economics, 14, pp. 134–6. Goodhart C. A. E. (2002): ‘The endogeneity of money’, in Arestis P., Desai M., Dow S. C. (eds): Money, Macroeconomics and Keynes, Essays in Honour of Victoria Chick, vol. 1, Routledge, London. Hewitson G. (1995): ‘Post-Keynesian monetary theory: some issues’, Journal of Economic Surveys, 9, pp. 285–310. Hicks J. R. (1982a): ‘Preface’, in Hicks J.: Money, Interest and Wages, Collected Essays on Economic Theory of John Hicks, vol. 2, Clarendon Press, Oxford. Hicks J. R. (1982b, orig. 1956): ‘Methods of dynamic analysis’, in Hicks J.: Money, Interest and Wages, Collected Essays on Economic Theory of John Hicks, vol. 2, Clarendon Press, Oxford. Hicks J. R. (1985): Methods of Dynamics Economics, Clarendon Press, Oxford. Howells P. G. A. (1995): ‘Endogenous money’, International Papers in Political Economy, 2 (2), pp. 1–41. Kahn R. (1954): ‘Some notes on liquidity preference’, The Manchester School, 22, pp. 229–57. Kaldor N., Trevithick J. (1988, orig. 1981): ‘A Keynesian perspective on money’, in Sawyer M. C. (ed.): Post-Keynesian Economics, Edward Elgar, Aldershot. Keynes J. M. (1973): ‘Variorum of drafts of the General Theory’, in Keynes J. M.: The Collected Writings, The General Theory and After, vol. 14, Macmillan, London. Kregel J. (1988, orig. 1976): ‘Economic methodology in the face of uncertainty: the modelling methods of Keynes and the post-Keynesians’, in Sawyer M. C. (ed.): Post-Keynesian Economics, Edward Elgar, Aldershot. Lavoie M. (1992): Foundations of Post-Keynesian Economics, Edward Elgar, Aldershot. Lavoie M. (1996): ‘Horizontalism, structuralism, liquidity preference and the principle of increasing risk’, Scottish Journal of Political Economy, 43, pp. 275–300. Lavoie M. (1999): ‘The credit-led supply of deposits and the demand for money: Kaldor’s reflux mechanism as previously endorsed by Joan Robinson’, Cambridge Journal of Economics, 23, pp. 103–13. Minsky H. P. (1975): John Maynard Keynes, Macmillan, London. Moore B. J. (1988a, orig. 1983): ‘Unpacking the post Keynesian black box: bank lending and the money supply’, in Sawyer M. C. (ed.): Post-Keynesian Economics, Edward Elgar, Aldershot. Moore B. J. (1988b): Horizontalists and Verticalists: the Macroeconomics of Credit Money, Cambridge University Press, Cambridge. Moore B. J. (1991a): ‘Money supply endogeneity: “reserve price setting” or “reserve quantity setting”?’, Journal of Post Keynesian Economics, 13, pp. 404–13. Moore B. J. (1991b): ‘Has the demand for money been mislaid? A reply to “Has Moore become too horizontal?” ’, Journal of Post Keynesian Economics, 14, pp. 125–33. Moore B. J. (1995): ‘The exogeneity of short-term interest rates: a reply to Wray’, Journal of Economic Issues, 29, pp. 258–66. Moore B. J. (1997): ‘Reconciliation of the supply and demand for endogenous money’, Journal of Post Keynesian Economics, 19, pp. 423–8. Moore B. J. (1998): ‘Accommodation to accommodationism: a note’, Journal of Post Keynesian Economics, 21, pp. 175–8. Palley T. I. (1987): ‘Bank lending, discount window borrowing, and the endogenous money supply: a theoretical framework’, Journal of Post Keynesian Economics, 10, pp. 282–303. Palley T. I. (1991): ‘The endogenous money supply: consensus and disagreement’, Journal of Post Keynesian Economics, 13, pp. 397–403. Palley T. I. (1994): ‘Competing views of the money supply’, Metroeconomica, 45, pp. 67–88.

© Blackwell Publishing Ltd 2004

Rethinking Endogenous Money

385

Palley T. I. (1996a): ‘Accommodationism versus structuralism: time for an accommodation’, Journal of Post Keynesian Economics, 18, pp. 585–94. Palley T. I. (1996b): ‘The emergence of theoretical and institutional coherence in post Keynesian economics’, in Palley T. I.: Post Keynesian Economics: Debt, Distribution and the Macro Economy, Macmillan, London. Palley T. I. (1998): ‘Accommodationism, structuralism, and superstructuralism’, Journal of Post Keynesian Economics, 21, pp. 171–3. Pollin R. (1991): ‘Two theories of money supply endogeneity: some empirical evidence’, Journal of Post Keynesian Economics, 13, pp. 366–96. Pollin R. (1996): ‘Money supply endogeneity: what are the questions and why do they matter?’, in Deleplace G., Nell E. J. (eds): Money in Motion: The Post Keynesian and Circulation Approaches, Macmillan, London. Rochon L. P. (1999). Credit, Money and Production: an Alternative Post Keynesian Approach, Edward Elgar, Cheltenham. Sawyer M. (1996): ‘Money, finance and interest rates’, in Arestis P. (ed.): Keynes, Money and the Open Economy, Essays in Honour of Paul Davidson, vol. 1, Edward Elgar, Cheltenham. Svensson L. E. O. (1999): ‘Price-level targeting versus inflation targeting: a free lunch?’, Journal of Money, Credit and Banking, 31, pp. 277–95. Taylor J. B. (ed.) (1999): Monetary Policy Rules, University of Chicago Press, Chicago, IL. Wolfson M. (1996): ‘A post Keynesian theory of credit rationing’, Journal of Post Keynesian Economics, 18, pp. 443–70. Woodford M. (1996): ‘Loan commitments and optimal monetary policy’, Journal of Monetary Economics, 37, pp. 573–605. Wray R. L. (1990): Money and Credit in Capitalist Economies: the Endogenous Money Approach, Edward Elgar, Aldershot. Wray R. L. (1992): ‘Commercial banks, the central bank, and endogenous money’, Journal of Post Keynesian Economics, 14, pp. 297–310. Wray R. L. (1995): ‘Keynesian monetary theory: liquidity preference or black box horizontalism?’, Journal of Economic Issues, 29, pp. 273–83. Economics Leeds University Business School Maurice Keyworth Building The University of Leeds Leeds LS2 9JT UK E-mail:[email protected]

© Blackwell Publishing Ltd 2004

Life Enjoy

" Life is not a problem to be solved but a reality to be experienced! "

Get in touch

Social

© Copyright 2013 - 2019 DOKUMENTIS.COM - All rights reserved.