Economic theory and policy: a coherent post-Keynesian approach

Arestis, Philip Article Economic theory and policy: a coherent post-Keynesian approach European Journal of Economics and Economic Policies: Intervention (EJEEP) Provided in Cooperation with: Edward Elgar Publishing Suggested Citation: Arestis, Philip (2013) : Economic theory and policy: a coherent post-Keynesian approach, European Journal of Economics and Economic Policies: Intervention (EJEEP), ISSN 2052-7772, Edward Elgar Publishing, Cheltenham, Vol. 10, Iss. 2, pp. 243-255, https://doi.org/10.4337/ejeep.2013.02.08 This Version is available at: https://hdl.handle.net/10419/277272 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Economic theory and policy: a coherent post-Keynesian approach Philip Arestis* University of Cambridge, UK, and University of the Basque Country, Spain This contribution focuses on a coherent new way of thinking about the macroeconomy in terms of both economic theory and economic policies. The central bank should focus on financial stability; for fiscal policy in the short term and in the long term to address demand issues is very important. Interest rate policy should be such that the real rate of interest is in line with trend rate of growth in the economy. Fiscal and monetary policies, though, should be coordinated closely. Major central bank cooperation and intervention in the foreign exchange market is necessary to control the exchange rate. Regional and industrial policies to create the required capacity are important, along with incomes policies, to contain inflationary/deflationary pressures. Distribution of income and wealth is another important policy dimension in view of its importance in terms of the great recession. Keywords: economic theory, economic policies, post-Keynesian economics JEL codes: D30, E12, E42, E43, E44, E64 1 INTRODUCTION This contribution proposes a coherent new way of thinking about the macroeconomy in terms of both theory and economic policy. We discuss first the theoretical framework that underpins the relevant economic policies before we turn our attention to the economic policies themselves. In terms of the latter we argue that distributional effects and financial stability are two important policy dimensions, which have been ignored in the past, and should be seriously taken on board. 1 After this short introduction, we turn our attention in Section 2 to our theoretical background, followed in Section 3 by the discussion of the objectives and instruments of economic policies that emerge from the theoretical framework. In the final part of this contribution, Section 4, we summarise and conclude. * This contribution is a significant extension on our previous relevant publications, for example Arestis (1989; 2010) and Arestis/Sawyer (2010a; 2010b). I am very grateful to Malcolm Sawyer, Eckhard Hein and the participants to the FMM conference for helpful comments. The usual disclaimer applies. 1. Interest in both distributional effects and financial stability have surfaced recently in view of the ‘great recession’. Arestis and Karakitsos (2011), for example, discuss both dimensions. Atkinson et al. (2011) and Van Treeck (2012) concentrate more on the distributional effects dimension. European Journal of Economics and Economic Policies: Intervention, Vol. 10 No. 2, 2013, pp. 243–255 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd The Lypiatts, 15 Lansdown Road, Cheltenham, Glos GL50 2JA, UK and The William Pratt House, 9 Dewey Court, Northampton MA 01060-3815, USA 2 THEORETICAL FRAMEWORK The constituent elements of the model we put forward are discussed below. The model is divided into five blocks as follows. 2.1 Block I: aggregate demand and supply This block is based on the proposition that the level of national income is determined by the level of aggregate demand. The latter is the sum of consumer demand, investment demand and government expenditure plus net exports, as in Equation (2.1): 2 Y¼CþIþGþðX−QÞ(2.1) where Yis national income, Cis consumption, Iis investment, Gis government expenditure, Xis exports and Qis imports, and thus (X−Q) is net exports (NE). Aggregate demand is important in both the short run and the long run for the evolution of the economy. There is no market-based mechanism whereby market forces would propel the level of aggregate demand to any specific level of output and/or supply-side determined equilibrium. We examine the components of aggregate demand next, beginning with consumption, as in Equation (2.2): C¼C½ðWEð1–twÞ;Пð1–tπÞ;R;ΔBLPh þþ−þ(2.2) where Wis wages, Etotal employment, so WE is the wage bill, tw is the tax rate on wages, Πis total profits, tπis the tax rate on profits, Ris the rate of interest on loans, and ΔBLP h is changes in bank lending to households. The availability of credit to households influences consumption since expenditure has to be financed and some households are credit constrained. The sign under the variables indicates the partial derivative of Cwith respect to the relevant independent variable. The independent variables are presumed to have a positive effect on the dependent variable with the exception of R, which has a negative effect. We treat tw and tπas exogenous and the rest of the variables are endogenised as explained below in the relevant blocks. Investment expenditure is taken to be: I¼IðП=K;Y=Ya;R;ΔBLPfÞ þþ−þ(2.3) where the symbols are as above with the exception of K, which is capital stock, ΔBLP f that stands for changes in bank lending to firms, and Ya, which is a measure of capacity output (further discussed below). All the variables affect investment positively, with the exception of R. For simplicity, the rate of interest on loans to households and to firms are taken as closely related, such that the use of a single rate of interest on loans in Equations (2.2) and (2.3) is thereby justified. Government 2. In what follows, capital letters indicate the level of the relevant variable while lower-case letters stand for the rate of change of the relevant variable. 244 European Journal of Economics and Economic Policies: Intervention, Vol. 10 No. 2 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd expenditure (G), as in Equation (2.1), is treated as a policy variable and further examined in Block IV. Net exports (X−Q) are as in Equation (2.4): X−Q¼XðWT;RERÞ−QðY;RERÞ¼NE þ−þþ (2.4) where WT is world trade, RER is real exchange rate, and the rest of the symbols are as above with all these variables in real terms. With the exception of the negative impact of the real exchange rate on exports (X), the rest of the independent variables have a positive impact on net exports, denoted as NE in Equation (2.4) and further examined in Block V. The foreign sector is, therefore, viewed as a significant aspect for the aggregate demand in terms of the imports and exports. The latter are included in the aggregate demand equation to also account for the effects on demand (and hence employment) of variations in the exchange rate. There is also the supply side of the economy, which is reflected in a range of ways. The level of economic activity (as set by the level of demand) relative to the supply potential of the economy impacts on investment, as in Equation (2.3) above, and pricing decisions, as in Equation (2.14) below. The supply potential evolves over time as investment occurs; and it depends on the size and structure of the capital stock and the labour force. The relationship between wages and prices, and the inflationary forces in the economy, depend on the nature of the supply side of the economy and how it interacts with demand. The potential level of output at the level of the firm depends on, in a production function manner, the inputs of labour, capital, etc., which can be deployed by the firm and the state of technology. At the aggregate level, it is assumed that there is a comparable relationship, but it must be recognised that there are severe issues of aggregation and that the structure of the capital stock will also influence the supply potential. This aggregate supply output (Ys), which could be produced, depends on employment of labour, the capital stock and the state of technology, as in Equation (2.5): Ys ¼YsðE;K;STÞ þþ þ (2.5) where Eis employment, Kis some measure of capital stock, and ST is state of technology. Actual output produced is taken to be demand-determined, and the production equation can be inverted to give employment for a given level of (demand determined) output as in Equation (2.6): E¼EðY;K;STÞ: þþ þ (2.6) In this formulation, changes in demand (Y), capital stock and technology have a positive effect on employment. A benchmark level of output from a supply perspective is defined (Ya), which is a capacity measure corresponding to the ‘desired’level of operation (by firms), so that: Ya ¼YaðK;STÞ þþ (2.7) Economic theory and policy: a coherent post-Keynesian approach 245 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd and Ya would change over time in the same direction as capital stock and technology change. There is a level of employment corresponding to Ya (Ea), which would be as in Equation (2.8): Ea ¼EðYa;K;STÞ: þþþ (2.8) Ya is taken as a benchmark for firms’investment decisions: when Y>Ya firms feel themselves to be working at over-capacity and encouraged to invest; but when Y<Ya they are thereby discouraged from investment. For a given level of Kand ST, it is envisaged that the relationship between productivity (Y/E) and Eis such that Y/Einitially rises as Erises, then flattens out, and eventually declines. It is clear from this analysis that aggregate demand and aggregate supply are not necessarily moved towards each other in the long run by the market mechanism. Indeed, how aggregate demand and supply evolve over time depends on the parameters involved. Aggregate demand is influenced by a number of factors over time, such as wage and price changes along with the consequent changes in the distribution of income between wages and profits. Aggregate supply also evolves over time and depends on the parameters as shown in Equation (2.6), in a way that is completely independent of those of aggregate demand. Clearly, in this analysis distributional aspects are relevant and important. We turn to this aspect next. 2.2 Block II: distributional aspects and the inflationary process The interaction of the determinants of the profit rate, price and wage rates produces the distributional aspects of our theoretical framework. We begin with the profit rate (π) function as in Equation (2.9): π¼π½ðP−ULCÞ;Y;R;DRf þþ−− (2.9) where the profit rate is a function of the difference between the price level (P) and the unit labour cost (ULC), both in logarithmic form of their levels, which affects the rate of profit positively; output, also affecting the rate of profit positively, and the rate of interest, which affects πnegatively. This reinstates the significant role of commercial banks in the investment/saving process, and is further explored in Block III. The debt ratio of firms (DR f ), defined as total debt to total assets, also affects πnegatively. The ULC is defined next, as in Equation (2.10), where again the variables are in logarithmic form of their levels: ULC ¼W−PR (2.10) so that ULC is defined as the difference between wages (W) and productivity (PR). We assume that productivity is exogenously determined with the wage level endogenously determined. Equation (2.11) endogenises wages through its rate of change (w): w¼wf½ðW=PÞd−ðW=PÞ;ðY−YaÞ;p;U;π;weg: þþþ−þþ (2.11) 246 European Journal of Economics and Economic Policies: Intervention, Vol. 10 No. 2 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd The rate of change of wages in Equation (2.11) is based on the bargaining position of workers as their desired real wage [(W/P) d ] deviates from the actual real wage (W/P). It is also influenced by the difference between actual and ‘desired’output (Y−Ya), inflation (p), unemployment (U), where unemployment is taken as a percentage of the labour force, by profitability (π), and wage expectations (w e ). The main hypothesis underpinning Equation (2.11) is that workers, unionised or non-unionised, bargain for a ‘desired real wage’. The next step is the determination of unemployment (U) as in Equation (2.12). U¼U½ðY–YaÞ;PR: −− (2.12) Unemployment in Equation (2.12) is related to the difference between actual and ‘desired’output (Y−Ya) and to productivity (PR). Turning to the inflationary process, we emphasise that inflation is multi-causal and the sources of inflationary pressure vary over time and economy. The inflation process is predicated on the premise that the price of output is a mark-up on the remuneration of the variable cost of production, which is labour, as shown in Equation (2.13): p¼p½w;ðY−YaÞ;pr;er;prm;pe þþ −−þþ (2.13) where inflation (p) is positively related to the rate of change of the nominal wage rate, w, to the difference between actual and ‘desired’output (Y−Ya), and it is negatively related to the rate of change of productivity (pr) and to the rate of change of the nominal exchange rate (er). It is also positively related to the growth rate of the prices of raw materials (p m ) and to price expectations (p e ). Finally, for this block, the growth rate of the prices of raw materials is determined as in Equation (2.14): prm ¼prmðer;WTÞ: −þ(2.14) Equation (2.14) makes the growth rate of the prices of raw materials a function of the growth rate of the exchange rate (er), affecting it negatively, and depend on world trade (WT) in a positive way. The latter two variables that are relevant to the open economy case are examined further below in Section 2.6. 2.3 Block III: money and credit The money, credit and finance sector is an essential and important part of the macroeconomic framework under discussion. Money is endogenously determined within the private sector with the liquidity preference of banks providing a crucial role in the determination of the money stock. In the process, loans provided by banks themselves generate bank deposits. The central bank sets the key policy interest rate, which governs the terms upon which the central bank provides the ‘base’money to the banking system. The expansion of the stock of money is driven by the demand for loans, which leads to the expansion of bank deposits in so far as the demand for loans is met by the Economic theory and policy: a coherent post-Keynesian approach 247 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd banking sector. Changes in the money supply can then come about through changes in bank deposits to the government and to the public. We can, thus, have in Equation (2.15): ΔM¼ΔBDGC þΔBDP (2.15) where changes in the money supply (ΔM) is the sum of changes in bank deposits to the government including currency (ΔBDGC) and changes in bank deposits to the public (ΔBDP). ΔBDGC is a small fraction of the total money supply and we treat it as the residual in the identity described by Equation (2.16): ΔBDGC ¼ΔBLP þΔBLG þΔBLES −ΔBDP:(2.16) Equation (2.16) then defines ΔBDGC as the sum of changes in bank lending to the public (ΔBLP) 3 and to the government (ΔBLG), as well as of changes in bank lending to the external sector including other non-bank lending (ΔBLES), minus ΔBDP. ΔBLES is treated as an exogenous variable and with ΔBLG endogenised in Block IV (see Equation (2.20)), the rest of the variables of Equation (2.16) are endogenised as explained immediately below. We begin with ΔBLP as in Equation (2.17): ΔBLP ¼ΔBLPðΔY;ΔR;ΔCRRÞ: þ−− (2.17) ΔBLP is hypothesised to depend on changes in the level of income (ΔY) reflecting requirements for funding as the level of economic activity changes; the cost of borrowing is captured by changes in the rate of interest (ΔR). The variable ΔCR R in Equation (2.17) stands for policy variables, which can affect credit directly. The best example for this variable is credit-rationing by the authorities. This possibility can occur when the demand for bank lending exceeds the available supply, or when commercial banks refuse to lend to more-risky customers who are judged as imprudent in terms of their borrowing behaviour. This is a variable that could be used to proxy financial stability influences as discussed in Section 3. The other variable in Equation (2.16) that we endogenise is ΔBDP, as in Equation (2.18): ΔBDP ¼ΔBDPðΔY;ΔR;XÞ þ−− (2.18) where ΔBDP depends on changes in the level of income (ΔY), reflecting the flow of funds into the banking sector as a result of changes in the level of economic activity. It also depends on changes in the rate of interest on loans (ΔR), which is hypothesised to account for possible portfolio effects: it proxies the relative attractiveness of alternative financial assets available to depositors as the market rate of interest changes. Xis an exogenous variable, which provides another possible avenue for financial stability variables as discussed in Section 3. 3. Clearly, ΔBLP aggregates the bank lending needs to industry (ΔBLP f ), consumers and other (non-commercial bank) financial institutions (ΔBLP h ). 248 European Journal of Economics and Economic Policies: Intervention, Vol. 10 No. 2 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd The market interest rate (R) is determined as in Equation (2.19). ΔR¼ΔRðΔBR;ΔEF;ΔPDCÞ: þþ þ (2.19) Changes in the bank rate (ΔBR), as set by the monetary authorities, influence directly changes in the market interest rate (ΔR) via a mark-up. Two further variables are present: ΔEF to account for changes in external flows that can have an impact on market interest rates (ΔR); and ΔPDC, which stands for sales of public debt to the nonbank public including currency. The variable is included to capture the influence of open-market operations on market interest rates; as such it is treated as exogenous. We examine next the government sector in Block IV. 2.4 Block IV: government sector We begin our discussion of this block with the ΔBLG variable, which is determined through the government budget identity as shown in Equation (2.20): ΔBLG ¼PSBR þΔEF −ΔNBC (2.20) where PSBR stands for the public sector borrowing requirements, ΔEF stands for changes in external financing, and ΔNBC stands for sales of public debt to non-bank public including currency. Equation (2.21) reflects institutional arrangements, where normally that part of the government’s borrowing requirement, which is not financed by the sale of debt outside the banking system, is met by the sale of debt –in particular Treasury bills –to the banking system. The latter acts, thereby, as the residual source of borrowing forthegovernment.WetreatΔEF and ΔNBC as exogenous and discuss PSBR next. PSBR, as portrayed in Equation (2.21), is simply defined as the difference between government expenditure (G) and tax revenues (T): PSBR ¼G−T:(2.21) We hypothesise Gand Tto be determined as shown below in Equations (2.23) and (2.25) respectively. Equations (2.20) and (2.21) are important ingredients of Block IV, which integrate the banking system with the government sector. The existence of a banking system implies that money is provided to the private sector through credit creation. The government supplies currency to the private sector by ‘financing’government expenditure. It is the case that the banking sector cannot provide currency independently from the government. Indeed, government expenditure (G) creates deposits at the central bank; and payment of taxes (T) reduces them. Any change in (G–T) in Equation (2.21) changes the PSBR, which affects ΔBLG via Equation (2.20), and via Equation (2.16) with ΔBDGC changing, the money supply is affected. But by accepting bank deposit money, the government agrees de facto to advance credit; in effect it is currency that discharges the difference (G–T) to banks via the central bank. We examine next Gand T. Gis determined as in Equation (2.22): G¼PGQQþWEGþUUBþID (2.22) where the symbols are as above with the exception of P G , the prices paid by the government for goods and services bought by the government (Q Q ), E G that stands for the Economic theory and policy: a coherent post-Keynesian approach 249 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd number of employees in the government sector, U B is unemployment benefits, and ID that stands for interest payments on government debt. E G is defined as in Equation (2.23): EG¼E−EP−U(2.23) where Eis the total working population, as defined above for the purposes of Equation (2.2), and E P is employment in the private sector. Clearly, E G +E P =E,thatis total employment. We may also endogenise Tsimply as in Equation (2.24): T¼TðYÞ þ(2.24) where Equation (2.24) accounts for tax rates and therefore represents tax policy. 2.5 Block V: open economy aspects We begin with the variable ΔEF –that is, changes in external financing as in Equation (2.25), which provides essentially the main ingredients to Block V: ΔEF ¼CB þΔKM −OEF (2.25) where ΔEF is equal to the current balance of international payments (CB) plus changes in capital movements (ΔKM) minus other external financing (OEF); the latter variable includes external lending to the public sector plus domestic bank lending to the public sector in foreign currencies. We treat OEF as exogenous and endogenise CB and ΔKM. We begin with CB as in Equation (2.26): CB ¼NE þOCB ¼XðWT;RERÞ−QðY;RERÞþOCB þ−þþ (2.26) where the symbols are as above, with the exception of OCB, which stands for other earnings on foreign investments minus payments made to foreign investors and cash transfers; this latter variable is treated as exogenous. Next, we endogenise ΔKM as in Equation (2.27): ΔKM ¼ΔKM½ðR=RWÞ;EðerÞ: þ− (2.27) The ratio of domestic interest rates (R) to world interest rates (R W ) is included on the assumption that capital flows are sensitive to returns available internationally. These returns, however, are not captured simply by interest rates but, perhaps more importantly, by expected exchange rate movements, hence the inclusion of the expected rate of change of the nominal exchange rate E(er). Finally, the real exchange rate (RER) is endogenised as in Equation (2.28): RER ¼RER½ðR=RWÞ;Y;WT;EðerÞ þ−þþ (2.28) where the variables are all as above. Clearly, with the exception of domestic income Y, all the other variables have a positive impact on the real exchange rate. 250 European Journal of Economics and Economic Policies: Intervention, Vol. 10 No. 2 © 2013 The Author Journal compilation © 2013 Edward Elgar Publishing Ltd