Demand, Money and Finance within the New Consensus Macroeconomics: a Critical Appraisal

Transcription

1 Leeds University Business School 17 th Conference of the Research Network Macroeconomics and Macroeconomic Policies (FMM) Berlin, October 2013 The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/ ) under grant agreement n * Demand, Money and Finance within the New Consensus Macroeconomics: a Critical Appraisal by Giuseppe Fontana and Marco Veronese Passarella * The views expressed in these slides are the sole responsibility of the author. The European Union is not liable for any use that may be made of the information contained therein.

2 Download this presentation and the preliminary draft of the paper at: under Conferences & debates

3 Outline of the Presentation Premise: convergence of views in dominant macroeconomics (Woodford (2009)) Aims: o o Critical analysis of both the basic NCM and some further developments of it (FAM) Few amendments are sufficient to make the model account for different equilibria Contents : o Introduction o Basic NMC Physiology of the basic NCM model o Emended NMC Natural equilibrium and the role of demand in the NCM model o Basic FAM Monetary policy and the role of money in the NCM model Financial markets and financial instability in the NCM model o Emended FAM Finance matters: a further amendment to the NCM model o Remarks

4 A new consensus in macroeconomics? Synthesis between Monetarism and Neo Keynesianism. Agreement on: 1. inter-temporal general-equilibrium foundations (or first principles ) 2. quantitative policy analysis based on econometrically-validated structural models 3. agents expectations explicitly considered (Lucas critique) 4. the main source of business fluctuations are real disturbances 5. monetary policy is effective Analytical components (Taylor (2000)): 1. long-run real GDP trend: Solow growth model with endogenous technology 2. no long-run trade-off between inflation and unemployment 3. short-run trade-off between inflation and unemployment (due to sticky prices) 4. expectations of inflation and of future policy decisions are endogenous 5. monetary-policy decisions are best thought of as rules (or reaction functions)

5 The basic three-equation model (MODEL 1) The basic DSGE-NCM model (Clarida et al. (1999), Romer (2000)): g g g (1) Yt a0 a1y t1 a2e Yt 1 a3 rt E t 1 1 (2) by b b E g t 1 t 2 t1 3 t1 2 Plus a reaction function, replacing the old LM curve: g T (3) r 1 c RR E c Y c c r t 3 t t1 1 t1 2 t1 3 t1 IS curve Phillips curve (or IA) Taylor rule Notes: 1. E(x) = x, except for ε 2. from (1) and (3) we derive the AD(π), whereas the AS is given by equation (2) 3. RR * is the natural real interest rate (Neo Wicksellian approach) 4. since prices are sticky, changes in r affect RR 5. mechanics: Y g > 0 π > π T r Y g = 0, where Y g = Y g (v i,u i,t i )

6 Simulation of the impact of a fiscal stimulus (a 0, +10%, from t = 25, MODEL 1) 60 Output gap 3.5 Inflation rate Output gap Inflation rate after the shock (p_1) Inflation rate before the shock (p_0) 3.9 Nominal interest rate.46 Real interest rate Nominal interest rate after the shock (r_1) Nominal interest rate before the shock (r_0) Real interest rate after the shock (r_1-p_1) Real interest rate before the shock (r_0-p_0)

7 Pros and cons of the NCM basic model Pros: o Includes some features of real world (price stickiness, imperfect information, etc.) o Money supply as residual (endogenous?) variable o Scope for stabilization policies in the short run o Keynesian concern for practical effectiveness and policy Cons: o No fiscal policy reaction function o No impact of current output on potential output (ergodicity and path-independency) o No involuntary or long-run unemployment o No room for financial markets, banking system and money creation o Microeconomic bias, in spite of weak microfoundations Critical reviews: Arestis (2007, 2009), Arestis and Sawyer (2004, 2006, 2008, 2009, 2013), Fontana (2009), Fontana and Palacio-Vera (2002, 2007), Lavoie 2006, León-Ledesma and Thirwall (2002), McCombie and Pike (2013), Setterfield (2002)

8 Demand matters o Real world economies are non-ergodic and path-dependent systems. Sample moments do not converge on their true values over time (Hanngsen (2006)). o Economic variables do not progress steadily toward a exogenously-given unique and stable equilibrium. They can reach several (sub-optimal) equilibria, and every equilibrium achieved depends, partly to least, on the dynamic process of getting that position. o Once productive capacity has been wasted, workers have not been trained, and investments have not been undertaken, it is not possible to turn back to the previous potential output, as if nothing happened (Setterfield (2002)). o Within MODEL 1 the achievement of the natural equilibrium is postulated. Yet, to the extent that it is admitted that the potential output is not independent of the current output (i.e. of the short-run effective demand) the NCM tale does not hold.

9 An emended NCM model (MODEL 2) A second model: (4) Y r t 0 1 t1 t1 1 (5) Y Y n t t1 1 t1 t1 2 T n (6) r RR Y E Y t t t 1 t 1 2 t t 1 IS curve Phillips curve (or IA) Taylor rule with: Y = Y n + Y g. Using (4) in (6), with r = r T and Y = Y n, we can derive the equilibrium real interest rate: (7) RR Y n / t 0 t 1 (8) Y Y Y Y n n n t t1 1 t1 t1 3 Equilibrium rate Finally, in the wake of Lavoie (2006), we amend previous model as follows: Hysteresis

10 Simulation of the impact of a fiscal stimulus (a 0, +10%, from t = 25, MODEL 2) Long-run impact of a fiscal stimulus on output (compared to baseline) 1.12 The short-run volume of demand affects the long-run natural (or potential) output Long-run supply forces are dependent on short-run disequilibrium adjustment paths induced by effective demand (Lavoie (2006), Flaschel (2000)) Output growth rate or volume (y)

11 Simulation of the impact of a fiscal cut (a 0, -10%, from t = 25, MODEL 2) Long-run impact of a fiscal cut on output (compared to baseline) 1.02 Two corollaries: discretional fiscal policy is effective also in 0.96 the long run negative shocks to demand (may) lead to long-lasting involuntary unemployment Actual output growth (y)

12 Monetary policy in the NCM model Five elements (Allsopp and Vines (2000)): 1. main purpose is to provide a nominal anchor to inflation expectations 2. this purpose is better pursued by an independent central bank 3. the main instrument is the short-term nominal interest rate in the unsecured money market 4. to be used also for stabilization purposes 5. fiscal policy affects the effectiveness of the monetary policy, therefore only automatic stabilisers admitted Monetarist features Keynesian (and/or Wicksellian) features Money stock is treated as a residual in NCM model and is, therefore, an endogenous variable. But endogeneity is regarded as a historical accident (Fontana and Palacio-Vera (2007), Arestis and Sawyer (2006)). Long run: ineffectiness of fiscal policy and neutrality of money. No distinction between banks and other intermediaries (Sawyer (2013); Passarella (2013)).

13 Bernanke and the financial accelerator literature FAM is a heretical sub-class of New Keynesian theories and models (Bernanke (1981, 1983), Bernanke and Gertler (1989), Bernanke et al. (1996, 1999)). Two basic hypotheses underpinning FAM models: 1. informational asymmetries entail higher costs of external finance, as compared to internal funds, in the form of agency costs (monitoring and bankruptcy risks) 2. the higher the amount of collateralizable net worth of firms, the lower are (expected) agency costs. Two implications follow: 1. since net worth of firms moves pro-cyclically, the premium on external finance rises in recessions and reduces in booms (cyclical persistence) 2. shocks affecting net worth of firms can trigger (or strenghten) real fluctuations An initial shock to demand, however small, is amplified by the change in balance sheets. Corollary: monetary policy affects output not as much because prices are sticky as because the access to external finance has a crucial impact on investment demand.

14 The financial accelerator mechanism (MODEL 3) Microfoundations: production (or investment) technology involving asymmetric information between entrepreneurs (who have direct access to the technology) and lenders (who have not). Lenders incur agency costs in order to observe returns on firms investment. Such costs are assumed a decreasing function of net wealth of firms. Since net worth moves procyclically, agency costs behave counter-cyclically, therefore improving lending conditions in booms and worsening them in recessions. In formal macroeconomic terms, equation (1) is to be replaced by the following: g g g (9) Yt a0 a1y t1 a2e Yt 1 a3 rt E t 1 a4ht 1 1 IS curve and: (10) h h Y g t t1 t 4 Lender s net wealth where h is net wealth of firms, ω is the share of aggregate (retained) profit and capital gains in total output (gap), and a 4 is the sensitivity of total output (gap) to change in creditworthiness of firms, through a change in investment financing.

15 Simulation of the impact of a fiscal stimulus (a 0, +10%, from t = 25, MODEL 3) An initial shock to demand, however small, is likely to be amplified by the change in net wealth of firms and by conditions in finance and credit markets. FAM authors openly abstract [...] from longterm financial relationships (Bernanke and Gertler (1989)). Financial accelerator mechanism (FAM) strengthens the effect of the initial shock to output gap Price flexibility is no longer regarded as the natural or long-run condition of the system, but just as the limiting case (Bernanke et al. (1999)). Sub-optimal equilibria and flight-to-quality : scope for public intervention Output gap in basic NCM(Yg) Output gap in FAM (YgA)

16 A further amendment to the NCM model (MODEL 4) Finally, it is possible to take into account both the hysteresis of output (MODEL 2) and cumulative effect of change in financial asset prices on investment activity (MODEL 3). In formal terms, equation (4) has been replaced by the following: where: (11) Y r h t 0 1 t1 t1 2 t1 1 (12) h h Y Y n t t1 t t 4 Consequently, equation (7) must be replaced by the following: (13) RR Y h / n t 0 t 2 t2 1 Running MODEL 4 and comparing it with models 1 and 2

17 Simulation of the impact of a fiscal stimulus (MODELS 1,2,4) Long-run impact of a fiscal stimulus on output (compared to baseline): a comparison between different models MODEL 4 (green) is a synthesis of models 2 (red) and 3: changes in finance and credit markets amplify real shocks and can also trigger a boom/recession; in addition, longrun levels of output and employment are affected by the current state of effective demand Basic NCM model Augmented NCM model Augmented FAM model

18 Four different declensions of the NCM model No role of finance Role of finance (accelerator) No permanent effect of demand MODEL 1: Basic NMC MODEL 3: Basic FAM Permanent effect of demand (hysteresis) MODEL 2: Emended NCM MODEL 4: Emended FAM

19 Some further consideration on the FAM model On closer inspection, the financial accelerator is none other than a way to introduce a longlasting (though not ever-lasting) hysteresis of output in the basic NCM model. This is implicitly recognized by FAM proponents too. In the absence of information asymmetries they argue investment demand can be safely assumed to be fixed over time. By contrast, when information asymmetries are present, investment demand will vary and be history-dependent (Bernanke and Gertler (1989)). Corollary: the main task of central banks is not the stabilization of inflation expectations (through the steering of the target interest rate), but the strengthening of firms (and banks ) balance-sheets, through the stabilization of financial asset (viz. collateral) markets.

20 Final remarks To sum up: 1) In spite of the new way of treating monetary policy, the NCM still shares some problematic features with Monetarism. o o o Expansionary fiscal policy has no lasting effect on real activity, while it leads to higher inflation and interest rates. Banks and financial markets are not included in the analysis. NCM model relies on the arguable assumptions that the natural level of output is exogenouslygiven and money is neutral in the long run. 2) FAM models explicitly aim to address the issue of the impact of changes in the financial structure on the real economy. Although the distance between FAM proponents and the Post Keynesians is still relevant, FAM literature represents a step forward. 3) Finally, few adjustments in the NCM basic framework (aiming to account for both the hysteresis of output and the role of finance) are sufficient to make the model produce heterodox results.

21 Thank you Download this presentation and the preliminary draft of the paper at: under Conferences & debates