«Abstract This chapter on macro aspects of taxation begins with a discussion of the contribution of taxation to stabilization policies. The budget ...»
Since frequently the government is the major debtor, the budgetary consequences of inflation cannot be neglected. Inflation is thus an alternative for traditional taxation. Furthermore, tax systems are most of the time not indexed so, in general, the direct impact of inflation is a more than proportional rise in tax revenues. This is also the case for the taxation of the inflation premium included in interest rates (Darby,1975), Feldstein,1976, and Feldstein and Summers, 1978; see also Tanzi, 1984). In this approach inflation is viewed as exogenous. In the optimal inflation literature, inflation is explicitly considered as a policy instrument to generate seignorage revenues for the government. Without considering the numerous country applications, we mention Bailey (1956), Friedman (1971), Tower (1971), Barro (1972), Marty (1973), Phelps (1973), Auernheimer (1974), Cathart (1974), Kahn and Knight (1982), Barro (1983), Calvo and Peel (1983), Cox (1983), Remolona (1983), Brock (1984), Mankiw (1987), Weil (1987), Marty and Chaloupka (1988), Brock (1989), Kigel (1989), Vegh (1989), Bohanon, McClure and Van Cott (1990), Bruno and Fischer (1990), Calvo and Leiderman (1992), Calvo and Guidotti (1993), Guidotti and Vegh (1993a, 1993b), Steindl (1993), Banaian, McClure and Willett (1994), Braun (1994a), Chang (1994), Marty (1994), Mourmoras and Tijerina (1994) and Vegh (1995).
The traditional assumption is that no alternative payment instruments to domestic money are available. Hercowitz and Sadka (1987) allow for currency substitution. As this limits the revenues from the inflation tax, governments will broaden financial regulations to minimize the domestic use of foreign currencies.
Taxation and inflation are also, although less directly, linked by the budget constraint of the government. Note that taxation will directly affect the price level but since, in general, changes in tax rates do not occur continuously, they will not explain inflation. It is therefore more the decision not to increase taxation by incurring budget deficits that could be important for the inflation process. In general two main channels can be distinguished that link the budget deficit to inflation. A first link results from a monetization of deficits; a second, less direct link, is the consequence of crowding out. Indeed, when higher government expenditures substitute for investment, potential production declines which, when the stock of money remains constant, implies a price increase. Note that this is not the case when the deficit results from a decline in taxes since, following the previous argument, the price level will decline.
24 Taxation: Macro Aspects 6010 Modigliani (1987), however, stresses reverse causation effects running from inflation to higher interest rates and so to deficits; furthermore, restrictive monetary policy to fight inflation lowers income which leads to increased deficits.
Concerning the monetization of the deficit, references mentioned above about the budget constraint are also relevant here. The link between budget deficits, whatever the origin, and inflation was forcefully formulated by Sargent and Wallace (1981) in their ‘Some Unpleasant Monetarist Arithmetic’: interest payments will, when the government debt exceeds some upper limit, have to be financed by money creation due to a shortage of savings. The ‘unpleasant arithmetic’ stresses thus that a bond-financed deficit will eventually have to be financed by money creation and therefore lead to a higher inflation rate. Darby (1984) (reply by Miller and Sargent, 1984) argues that the situation described by Sargent and Wallace is unlikely to occur since it requires that the real bond rate exceeds the real rate of growth for a considerable time (see also McCallum, 1984). King and Plosser (1985) discuss, test and reject (for the US as well as for several other developed countries) the link between seignorage revenues and different policy variables. Sargent and Wallace assume that monetary policy accomodates fiscal policy (‘fiscal dominance’); Buiter (1987) and Buti (1990) consider different policy regimes (cooperation or leadership between monetary and fiscal authorities). Drazen and Helpman (1990) stress the importance of the policies expected to be used to reduce the deficit (see also Niskanen, 1978;
Blinder, 1983; Miller, 1983, and Congdon, 1987).
7. The Positive Approach to Taxation
In democratic societies, tax rates are fixed by politicians and not by technicians who aim for an optimal taxation. The traditional approach holds that taxation or government expenditures are instruments that policymakers modify to realize goals such as a stabilization of the economy, a reallocation of resources or a modification in the distribution of income and wealth. This fine-tuning policy faces several problems and risks. For example, the government can misjudge both the size and timing of interventions, many conflicting objectives need to be pursued, the effects of instruments are not known with certainty, data contain errors, the theoretical structure of the model, the parameters, the value of exogenous variables, the nature of exogenous disturbances are not perfectly known, time lags are involved (recognition, decision, execution and response lag), frequent changes in policy instruments affects policy credibility, and so on. See, for example, Friedman (1960), Okun (1972), Kydland and Prescott (1977), Lucas (1980), Cooley, Leroy and Raymon (1984); and see also
the recent literature on taxation as a source of business cycle fluctuations:
Christiano and Eichenbaum (1992), Braun (1994b), McGrattan (1994) and 6010 Taxation: Macro Aspects 25 Johnsson and Klein (1996). The previous arguments (partly) explain the preference of several authors for policy rules. See the chapter on optimal taxation for more detail. In the public choice literature a different approach is taken. A typical formulation is Brennan and Buchannan (1980a) where a government is seen as a revenue-maximizing leviathan and taxes result out of a struggle between the government and citizens. Taxation is thus essentially viewed as an instrument to satisfy the wellbeing of the policymakers, not of the community and taxes are required to finance government expenditures. In this way taxes are more directly linked to government expenditures than in the ‘traditional’ approach. We refer to Downs (1957), Meltzer and Richard (1978), Anderson, Wallace and Warner (1986), Blackley (1986), Manage and Marlow (1986), Von Furstenberg, Green and Jeong (1986), Hibbs (1987), Ram (1988), Ahriakpor and Amirkhalkhali (1989), Mueller (1989), Blackley (1990), Holmes and Hutton (1990), Miller and Russek (1990), Cukierman and Meltzer (1991), Hoover and Sheffrin (1992) and Bella and Quinteri (1995). Consider also the collection of papers in Buchanan, Rowley and Tollison (1986). Several articles and books apply public choice arguments to a discussion of taxation or tax reform. It is impossibble to survey and list them. A pronounced view is Rose and Karran (1987).
Citizens bear, however, also some responsibility: the aggregate demand for budgetary programs will be excessive since any taxpayer bases his demand on the asymmetry between his small share in the cost of the program he supports and the perceived benefits. Furthermore, they suffer from fiscal illusion (see Dollery and Worthington, 1996, for a recent survey of this literature) that is reinforced by debt finance. As a result, the perceived price of public goods is lower than the effective price resulting in a rise in the demand for these goods.
On the other hand, there is little incentive to combat tax increases since the individual share is negligible. As a result the upward pressure on the supply of government programs is stronger than the incentive to hold down the budget so government expenditures tend to rise and be financed by tax increases.
The literature on the ‘political business cycle’ should also be mentioned here. It concerns the creation of favorable economic conditions at election time so as to increase the re-election chances of the incumbent politicians. Obviously taxation is an instrument that could be used to generate these cycles. We refer to Alesina and Roubini (1992), Borooah and Van der Ploeg (1983), Nordhaus (1989) and Schneider and Frey (1988) for a general review of this literature.
Unfortunately not much research has been performed on the use of taxes to generate a political business cycle (see, however, Tufte, 1978; Edwards and Tabellini, 1991; Grilli, Masciandaro and Tabellini, 1991, and Alesina and Rosenthal, 1995).
Econometric tax functions, most of the time for separate taxes and, eventually, specified as reaction functions, have been estimated in many econometric models. More relevant here but much less numerous, are general 26 Taxation: Macro Aspects 6010 tax equations that contain political determinants. We refer, however, to Frey and Schneider (1978), Alt and Chrystal (1981), Pommerehne and Schneider (1983), Renaud and Van Winden (1987), Van Velthoven (1989) and Ohlsson and Vredin (1996).
8. Open Economies The literature on fiscal policies is too vast to cover here in a few paragraphs so we limit ourselves to some general points (we refer to Chapter 6080 for international taxation). The point of departure for the analysis of the impact of taxation-fiscal policy in open economies is the Mundell-Fleming model (see Kenen, 1995, and Marston, 1995, for surveys). The magnitude and size of the multipliers depend on different parameters such as the effect on domestic interest rates (the assumption about the country size and capital mobility is relevant here) and the exchange rate regime. For example, a fiscal expansion in a ‘large’ country will raise world interest rates and appreciate the exchange rate; in a ‘small’ country a depreciation can be expected since the trade effects will tend to dominate. In the limiting case of perfect capital mobility, fiscal policy will not affect income: any tendency for interest rates to rise will be matched by a currency appreciation leaving total demand unchanged. In other words, exports are completely crowded out. Textbook references are relevant here.
Extensions by McKinnon (1973), Floyd (1980), Dixit (1985), Persson (1985) and Frenkel and Razin (1987) make use of an intertemporal approach covering aspects ranging from the impact of capital mobility and overlapping generations to different specific tax instruments. One general result is that deficits increase world interest rates and lower consumption but the effects depend on whether a country experiences a surplus or deficit. McKibbin and Sachs (1988) and Kole (1988) consider the importance of country size. In the traditional view financial crowding out will not occur in small open economies since, assuming fixed exchange rates, interest rates are given on the world market. Adjustments to imbalances between saving and investment occur through foreign borrowing by way of current account deficits. As a consequence, fiscal policies in all countries determine crowding out, also in countries with balanced budget. In a Ricardo world, this no longer holds.
The previous results must, however, be modified when non-traded goods are introduced. Some models focus more explicitly on the dynamics and on the budget constraint of the government. Work by Frenkel and Razin (1987) and Knight and Masson (1988) demonstrates that, if bonds are net wealth, fiscal expansions will appreciate the exchange rate so as to finance the government deficit. Buiter (1987) takes the impact of interest rates on investment into account so deficits increase interest rates and as a result lower future domestic 6010 Taxation: Macro Aspects 27 and foreign income. For a given level of government expenditures this constrains future taxes (see also Blanchard et al., 1990; Frisch, 1995).
An interesting collection of additional work on the international aspects of fiscal policy is Frenkel (1988). The survey by Dixit (1985), although limited to tax policy in open economies, should also be mentioned.
Concerning the Ricardian equivalence theorem in an open economy, (see Morris, 1988, for a discussion of the Barro-Ricardo equivalence in open economies) we recall earlier comments about the burden of external versus internal debt.
Note that a huge literature exists about the international coordination of fiscal policy and about fiscal policy in a monetary union. We also stress the important literature on the Maastricht convergence criteria for the European monetary union.