Apostolis Philippopoulos What we know about monetary policy The government may have a potentially stabilizing policy instrument in its hands. But is it effective? In other words, is the relevant policy multiplier nonzero? 1. Short run (given prices and expectations): A loose monetary policy leads to an increase in real money balances, a decrease in nominal interest rates and an increase in real economic activity. Open economy: even in the short run, there can be offsetting mechanisms that shift the AD back partly or fully like in the case of fixed exchange rates and high capital mobility. 2. Medium and long run: Money is neutral. Monetary policy is irrelevant to the real allocation. The AS is vertical since all monetary variables change proportionally. Disinflation is costless. 3. Transition from the short run to the medium run: It all depends on how fast inflationary expectations and prices adjust; see shifts of the short run AS. The main results are: (i) Monetary policy is effective (i.e. it can affect the real economy) only if it is unexpected. This gives the incentive for monetary surprises (see below). (ii) There is only a temporary trade-off between higher inflation and lower unemployment (the modern Phillips curve). 4. If the question is Does the choice of the monetary policy regime matter to the real economy, the answer is: (i) If there is full price flexibility and complete current information, the choice does not matter to the real economy. Of course, it matters to prices. (ii) If expectations and prices are given, the choice of the monetary and exchange rate policy regime matters to the real economy. Regime desirability depends on the type of the shock. See e.g. the Poole model where there is the dilemma between money stock rules and interest rate rules (this is an IS-LM model) or the Mundell-Fleming model in an open economy. (iii) If there are temporary nominal fixities and incomplete current information on the part of markets like in 3 above, and policymakers can follow optimally chosen feedback or active or 1
state-contingent policy rules being fully informed about the current state of the economy and optimally react to it, the choice of the monetary policy regime does not matter to the real economy. If they are not fully informed, the same message as in (ii) above, i.e. the choice of the regime or the type of policy rule matters to the real economy. 5. The above was about effectiveness. But monetary policy may not only be ineffective, but it can also do more harm than good in the hands of policymakers. A game between the central bank and wage setters. This is a prisoner s dilemma problem. Rules versus discretion. Ways of improving credibility: the role of institutions (like the ECB and the independent central banker) and the role of reputation. See Chari and Kehoe (Journal of Economic Perspectives, 2006, vol. 20. no. 4, pp. 3-28). 6. General lesson: monetary policy should be conducted by rules that try to keep nominal interest rates and inflation low. 2
What we know about fiscal policy 1. Effectiveness of fiscal policy instruments In the short run, expansionary fiscal policy affects AD and output. But there are adverse side-effects from loose fiscal policy that crowd out private activity. They work via traditional channels like interest rates, exchange rates, inflation, etc. Open economy: under flexible exchange rates and high capital mobility, fiscal policy can be fully ineffective in the short run. In the medium- and long-run, output is back to its natural level with lower private consumption and investment. Specifically, in dynamic setups, individuals react to government deficits and public debts by increasing their own savings in anticipation of higher taxes in the future to repay the debt (Ricardian equivalence) so that they offset the potential stimulating affects of fiscal expansions. 2. Stabilization Governments attempt to stabilize the economy from shocks (oil price shocks, financial shocks, etc). Stabilization takes the form of (i) automatic stabilizers (ii) counter-cyclical, or active, or feedback, policy rules according to which the government follows pre-announced policy rules that react to the state of the economy and (iii) discretionary policy according to which the rules change depending on the state of the economy. Don t forget that shocks are not bad per se: they are bad only when there are market failures that do not allow the economy to cope with these shocks in an efficient way. To study the effects and desirability of automatic stabilizers and active policy, we use dynamic stochastic general equilibrium (DSGE) models. Active policy has been subjected to criticism (see e.g. Tanzi, CESifo Forum, 3/2005). This is because (i) Fiscal policy operates with time lags so that it is hard to time it over the cycle. (ii) Fiscal action fixes one thing but may destabilize another. Actually, feedback policy rules can destabilize an otherwise stable economy. (iii) It is hard, in practice, to distinguish between rule-like counter-cyclical behaviour and discretionary behaviour driven by electoral cycles, etc. (iv) As said, individuals react to government deficits and public debts by increasing their own savings in anticipation of higher taxes in the future to repay the debt (Ricardian equivalence) so that they offset the potential affects of fiscal actions. 3
It is thus better to rely more on automatic stabilizers (which are passive policy rules). Automatic stabilizers work better, especially in countries with relatively large public sectors (less so in the US and the UK). But again there is no strong evidence that larger public sectors are associated with smaller macroeconomic volatility (recall the trade-off between market and policy failures). Pro-cyclical fiscal policy generates instability and is usually associated with corruption. 3. Resource allocation Fiscal policy can improve resource allocation (by providing infrastructure and public goods and correcting for externalities). For a simple model in which there is a trade-off between public infrastructure and distorting taxes required to finance it, see Barro (1990). Effects on long-term growth and welfare, and effects in the short- and medium-run (see Turnovsky, 1995). What is the optimal allocation of scarce tax revenues among different types of public spending? 4. Redistribution Fiscal policy can be used to redistribute income. This is on the grounds of resource allocation (there is the implicit assumption that markets fail because of borrowing constraints, etc) and/or on the grounds of equality (ethics). A trade-off between equality and growth. [If decision-making is via majority voting, the tax-transfer program that is adopted reflects the preferences of the median voter. The lower is his income relative to the mean, the higher will be the tax rate and the associated level of transfers and size of the transfer program (thus, here inequality is bad for growth because it implies higher taxes)]. Problems include: (i) Moral hazard. (ii) Rent seeking and lobbying by interest groups. The idea is that each interest group applies pressure to increase its subsidies and reduce its taxes. Empirically, there is evidence that the number of organized interest groups in a country has a positive effect of the relative size of the government sector. See Mueller, 2003, chapters 13 and 17. 5. Are budget deficits always bad? Deficits are not bad per se (see tax smoothing justification). But permanent deficits are bad. Permanent deficits result from short-sighted behaviour of policymakers as well as voters (e.g. systematic attempts to raise real activity at the cost of future generations, electoral or opportunistic motives of the 4
incumbent parties, strategic behaviour to hurt the next party in power, the common pool problem as interest groups compete for government resources and ignore the effects of their actions on overall public finances (also known as lobbying and pressure from interest groups for spending-tax favors), etc). Note that common pool problems become worse in good times (pro-cyclical policy) because of the so-called voracity effect which means that anti-social, extractive behaviour becomes worse when the pie gets larger. See Public Finances in the EMU, 2006, p. 207. All these problems become worse in a MU where (i) under loose fiscal policy, there are smaller interest rate rises and crowding-out problems than under flexible exchange rates (ii) there are no country or exchange rate risks. Hence national fiscal authorities have an incentive to overspend and overborrow. See Public Finances in the EMU, 2006, p. 143. Hence the arguments for fiscal rules like the Stability and Growth Pact, especially in a monetary union with a single currency. Other popular national fiscal rules include the so-called golden rule of fiscal policy (public net borrowing should be limited to the amount of net public investment) and structurally balanced budget (net borrowing should be allowed only for cyclical reasons so there must be additional saving in case of a cyclical upswing; this permits automatic stabilizers in the short run but implies a balanced budget in the medium run). 6. Spending-tax reforms (fiscal consolidations) Fiscal consolidations via tax increases are contractionary (i.e. they reduce growth and increase unemployment). By contrast, fiscal consolidations via spending cuts are expansionary (i.e. they raise growth and reduce unemployment), especially in countries with poor public finances. By spending cuts we mean cuts in government consumption (see government wages), government investment and transfers/subsidies. See Public Finances in EMU, 2007, pp. 171, 189, 194. Why are (socially beneficial) reforms delayed? For several explanations, see Saint-Paul (2004, Journal of Economic Perspectives, vol. 18, no. 4, pp. 49-68), Public Finances in the EMU, 2005, and Drazen (2000, p. ). 7. Optimal tax policy and credibility 5
What is the optimal path of different tax rates over time? Under commitment, some policy lessons are: (i) tax rates on labor income and consumption should be roughly constant over time (ii) capital tax rates should be low so as encourage capital accumulation (low capital taxes increase the capital-labor ratio, which in turn increases output per worker (iii) returns to public dent and taxes on assets should fluctuate to provide insurance against adverse shocks and balance the budget in a present value sense. The time inconsistency problem. Mechanisms for commitment (fiscal rules and reputation). See Chari and Kehoe (Journal of Economic Perspectives, 2006, vol. 20. no. 4, pp. 3-28). 8. Bureaucracy, corruption, rent seeking policymakers, etc. 9. Size versus efficiency of the public sector Measures of public sector efficiency. 6