Chapter 13: Fiscal policy
Aims of the chapter
This chapter looks at fiscal policy under three different, but related, profiles. First, we revise the simple arithmetic of the government budget constraint and its powerful implications for the conduct of fiscal policy. Second, we discuss in the context of fiscal policy the arguments in favour and against the use of full discretion in the implementation of active demand management policies, and how this has led most OECD countries
to constrain their fiscal policy within rules that guarantee long-run solvency. Finally, we highlight the role of seignorage as a means to finance public spending, and explain why budget deficit crises are the root of most hyperinflation or high inflation episodes which have occurred over the last 100 years.
• discuss the simple arithmetic of the government budget constraint
• describe the implications of the Ricardian equivalence proposition for the conduct of fiscal policy
• revise the theoretical underpinning of fiscal policy and the debate between the active or passive use of fiscal policy
• assess how the discretionary use of fiscal policy has evolved over time, and why most OECD countries currently determine the conduct of their policy according to specific fiscal rules
• explain the determinants of seignorage, and the links between the budget deficit and inflation
• explain why budget deficits may be a source of high inflation and even hyperinflation.
This chapter discusses several issues related to the conduct of fiscal policy and its effectiveness as a means of stabilising the economy. The chapter is divided into three parts. The initial discussion focuses on the arithmetic of the government budget constraint and an assessment of fiscal policy sustainability. We highlight the fact that the simple arithmetic of the government budget constant has very important implications for the conduct of fiscal policy itself. In particular, we show when permanent and temporary debt-financed increases in government spending are unsustainable, in that they yield an explosive debt dynamic, and require a fiscal correction (i.e. an increase in taxes) some time in the future. This is, ultimately, a restatement of the Ricardian equivalence proposition that we encountered in Chapter 9, but from the perspective of the government sector.
The second part of the chapter focuses on the use of fiscal policy as a tool to stabilise the economy over the business cycle. We review the theoretical underpinnings of fiscal policy, in particular comparing the policy prescriptions arising from the Keynesian and the monetarist approaches. Particular attention is given to the distinction between automatic stabilisers and discretionary fiscal policy interventions, and we highlight how the use of the latter as a tool to achieve macroeconomic stability has been changing over time in most OECD countries. We then highlight how this has coincided with a growing view that modern fiscal policy has to be conducted within specific rules that should guarantee the sustainability of policy over the medium and the long run. However, there is no international agreement over the definition of fiscal rules and, in practice, approaches vary considerably across countries.
We conclude the chapter by pointing out the link between the budget deficit and inflation, and show why budget crises (i.e. the inability of the government to finance public spending through taxes) are the principal sources of episodes of high (two-digit) inflation or even hyperinflation. Note that because inflation is often regarded as a monetary policy issue, we could have discussed the problem of hyperinflation in the previous chapter of the subject guide, which is dedicated to the conduct of monetary policy. However we discuss hyperinflation here, as this problem is usually considered to be the outcome of a bad conduct of fiscal policy, which is the central topic of this chapter of the subject guide.
The arithmetic of the government budget constraint
This section takes you through the basic arithmetic of the **Government Budget Constraint (GBC)** and shows you how to simply derive from the nominal GBC an expression for the real GBC measured relative to GDP, which is the definition of GBC most commonly employed in fiscal policy analyses. Note that in the definition of the GBC we make two important assumptions. First, the government can finance spending either through taxes or borrowing, and cannot resort to money creation. The implications of this third financing channel are explored later in this chapter. Second, we assume that the government can only borrow from the domestic bond market, so we ignore in this analysis the effect that exchange rate policy may have on foreign-currency-denominated debt. The nominal GBC is given by
P t G t − P t T t + i t P t − 1 B t − 1 = P t B t − P t − 1 B t − 1 , (13.1)
where P t indicates the price level in period t, G t is real government spending in goods and services during period t, T t equals real government revenue (taxes minus transfers) during period t, B t–1 is the real value of the outstanding debt at the end of period t–1 or, equivalently, at the
beginning of period t, and i t P t–1 B t–1 measures nominal interest payments on government debt in period t. The constraint states that any funding
shortfall the government faces must be met by increases in the value of its debt.
The real GBC is obtained by moving the term P t–1 B t–1 on the left hand side of equation (13.1) and then dividing each side of the resulting equation by P t , which yields:
G t − T t + ( 1 + i t ) (Pt − 1/Pt) (B t-1) = Bt
If we define the inflation rate as PIt = ( Pt - Pt-1/Pt-1), the term Pt-1/Pt can be written as P t − 1/Pt = 1/1+PI and the real GBC becomes:
G t − T t + ( 1 + r t ) B t −1 = B (13.2)
where 1 + r t = (1+it/1+PIt) and r t is the real interest rate on government debt.
The real GBC measured relative to GDP can be obtained by dividing both sides of equation (13.2) by real GDP Y t , which yields:
d t + ( 1 + r t ) (Bt-1/Yt) = bt
where d t = g t – t t is the real primary deficit-to-GDP ratio, with g t giving real government spending as a proportion of real GDP in period t, and t t real government revenue as a proportion of GDP in period t; b t is the real stock of outstanding debt as a proportion of GDP at the end of period t. Since the term Bt − 1/Y is equivalent to t − 1 t − 1 , the real GBC as a proportion (Bt-1/Yt-1)(Yt-1/Yt) the real GBC as a proportion of GDP can also be written as: d t + ( 1 + r t ) b t −1 (Yt-1/Yt) = bt (Y t −1/Y t) = bt
If we define the growth rate of real GDP as g_Yt = (Y_t − Y_t-1 / Y_t-1) the term Y_t-1/Y_t can be written as Y_t-1/Y_t = (1 / (1 + g_yt)) and the real GBC as a proportion of GDP becomes:
d t + ((1 + r t) /(1 + g Yt)) (b_t-1) = b_t
The term ((1 + r t) /(1 + g Yt)) is equivalent to ((1 + i_t)/(1 + pi_t)(1 + g_yt)) which can approximated as 1 + i_t - pi_t - g_yt, whenever pi_t i_t and g_yt are close to zero.
We can define the variable ρ_t as:
ρ_t = i_t − pi_t − g_Yt
which is the real interest rate on government debt adjusted for output growth, so that the real GBC as a proportion of GDP can be written as:
d t + ( 1 + ρ t ) b t −1 = b t
or, equivalently, as: d t + ρ t b t − 1 = b t − b t − 1 = ∆ b t . (13.3)
Equation (13.3) states that changes in the current debt-GDP ratio are determined by two terms. The first term is the primary deficit-GDP ratio. The second term is the product between the real interest rate adjusted by the growth rate of GDP and the outstanding debt-GDP ratio. Assuming outstanding debt is positive, the debt-to-GDP ratio increases if the government runs a primary deficit (for any given ρ t ), or if there is primary balance but ρ t > 0 – so that the nominal interest rate exceeds the sum of the growth rate of output and the inflation rate. The debt-to-GDP ratio
also rises over time if the government runs a primary surplus but this is more than offset by the net cost of borrowing, d t
The Barro–Ricardo equivalence proposition
The simple arithmetic of the GBC has very powerful implications for the conduct of fiscal policy, in that it imposes a feasibility constraint on fiscal policy actions.
To understand this point recall that under the IS–LM model if an economy is in recession the government can increase government spending (or cut taxes) to speed up the recovery of output to its natural level. Alternatively, because the balanced budget multiplier is positive, the government can also increase output during a recession by financing higher public spending with an equal increase in taxation. Since the public spending multiplier is greater than the balanced budget multiplier, the traditional Keynesian approach suggests that debt-financed fiscal expansions are more powerful than balanced budget fiscal expansions.
This view is in sharp contrast with that arising from the Barro–Ricardo (or Ricardian) equivalence proposition, which states that in financing public spending there is no difference between taxes and the accumulation of debt, since the latter is ultimately equivalent to future taxes. The main policy implication of the Barro–Ricardo equivalence proposition is that tax cuts funded by public borrowing are ineffective policy measures in stimulating aggregate demand.
Note that the Barro–Ricardo equivalence proposition is a restatement of the Ricardian equivalence proposition that we saw in Chapter 9 of the subject guide, but we will examine its implications from the perspective of the government rather than that of consumers. Table 13.1 illustrates the effect on the debt-to-GDP ratio of a permanent increase in public spending (or, equivalently, a permanent tax cut) which is entirely financed by an increase in government borrowing. To simplify the analysis assume that the output-adjusted real interest rate is constant over time, ρ t = ρ , and before the change in spending, in period t–1, the government has no outstanding debt, and has been running a constant balanced budget, d = 0. In period t the government increases spending by ∆ g t , and this increase is financed entirely by borrowing. Hence the debt-to- GDP ratio at the end of period t will be Δg t . In period t + 1 spending remains elevated, and the government must additionally borrow to fund outstanding interest payments on period – t debt. Hence b t+1 = Δg t + (1 + ρ)b t = [1 + (1+ ρ)] Δg t . In period t + 2 debt increases further, to [1 + (1+ ρ) + (1+ ρ) 2 ] Δg t .
More generally, by period t + n the permanent spending increase will have ... resulted in outstanding debt, b t+n , equal to ∑ (1+ρ) i ∆ g t which is growing i=0 exponentially over time.
t–1 b t–1 = 0
t b t = Δg t
t+1 b t+1 = Δg t + (1 + ρ)b t
b t+1 = [1+ (1 + ρ)] Δg t
b t+2 = Δg t + (1 + ρ)b t+1
b t+2 = [1+ (1 + ρ) + (1 + ρ) 2 ] Δg t
bt+n b t+n = ∑ (1+ρ) i ∆ g
Table 13.1: Permanent increase in public spending.
Table 13.2 explores the effect on the debt-to-GDP ratio of a temporary increase in public spending (or, equivalently, a temporary tax cut) which is entirely financed by an increase in borrowing. As in the previous table, it is assumed that before the change in spending, period t–1, the outstanding debt-to-GDP ratio is zero, and the government is running a constant balanced budget; the output-adjusted real interest rate is again assumed to be constant over time. In period t the government increases spending by ∆ g t and this increase is financed by borrowing. Because the increase in spending is temporary, in period t+1 the primary deficit returns to zero, but the debt-to-GDP ratio does not: it is equal to (1 + ρ)b t , or (1+ ρ) Δg t .
Similarly, in t + 2 debt will equal (1 + ρ)b t+1 , or (1 + ρ) 2 Δg t . In period t + n the debt-to-GDP ratio will have grown to (1 + ρ) n Δg t . As for a permanent increase in public spending, a temporary debt-financed increase in government spending also leads to an accumulation of debt over time at an exponential rate.
t–1 b t–1 = 0
t b t = Δg t
t+1 b t+1 = (1 + ρ)b t
t+2 b t+1 = (1 + ρ)] Δg t
b t+2 = (1 + ρ)b t+1
b t+2 = (1 + ρ) 2 Δg t
b t+n = (1 + ρ) n Δg t
Table 13.2: Temporary increase in public spending.
Tables 13.3 and 13.4 look at the financial implications of balanced budget expansions. Both tables show that since balanced budget expansions lead to increases in government spending exactly matched by an increase in taxation, they do not raise any sustainability issue.
t g + ∆ g t = t + ∆ t t ; b t = 0
t+1 g + ∆ g t = t + ∆ t t ; b t+1 = 0
g = t; b t–1 = 0
g + ∆ g t = t + ∆ t t ; b t+n = 0
Table 13.3: Permanent balanced budget increase.
t–1 g = t; b t–1 = 0
t g + ∆ g t = t + ∆ t t ; b t = 0
t+1 g = t; b t+1 = 0
t+n g = t; b t+n = 0
Table 13.4: Temporary balanced budget increase.
The main lesson of this analysis is that a debt-financed fiscal expansion must at some point in time be matched by an increase in taxation. If it is
not, then public debt will be on an unsustainable path, ultimately lending to default. Knowing this, rational consumers will anticipate future tax rises to match any current tax cuts, and tax cuts therefore become an ineffective means for stimulating the economy.
In summary, the Barro–Ricardo Proposition implies that:
1. Aggregate demand is unaffected by the way in which government spending is financed (i.e. either debt or taxes).
2. Permanent and temporary debt-financed increases in government spending are unsustainable without future tax rises, as debt grows at
an exponential rate and will explode in the long run.
3. Thus, both permanent and temporary debt-financed increases in government spending must lead to tax increases in the future. In the example in Table 13.2, if the government pays back its debt in period n n
it has to raise taxes by ( 1 + ρ ) ∆ g t in that period.
4. The longer the government waits to redeem its debt, the higher the increase of taxes in the future.
5. Fiscal policy can be employed to redistribute the tax burden across generations, in that debt-financed tax cuts increase current
generations’ income at the expense of future generations that will have to pay higher tax bills.
The theoretical underpinning of fiscal policy
We are now in a position to revisit the fiscal policy prescriptions arising from the main theories encountered in this course, in light of the feasibility constraint imposed on fiscal policy by the arithmetic of the GBC. The standard framework used to analyse the impact of fiscal policy as a stabilising mechanism is the IS–LM model, which assumes, among other things, a fixed prices, a fixed money supply, and a closed economy. Within this framework, a fiscal expansion increases both output and the real interest rate. Output increases because the fiscal expansion raises aggregate demand in the goods market. Higher output raises money demand for transactions and the fixed money supply implies that the interest rate must rise to maintain equilibrium in the money market. Clearly the extent to which a fiscal expansion stimulates output depends on the degree to which investment falls as a consequence of the higher interest rate. In principle, the lower the interest sensitivity of investment is, the less investment falls, and the greater the effect of the fiscal stimulus on output.
There are several elements that you need to bear in mind when evaluating the fiscal policy prescriptions arising from Keynesian theory.
First, the IS–LM model relies upon the assumption of short-run price and wage stickiness. The greater the degree of price and wage flexibility, the less effective fiscal (and monetary) policy are likely to be in stabilising the economy. But greater flexibility, in turn, reduces the importance of stabilisation in the face of demand and supply shocks.
Second, the arithmetic of the GBC, illustrated in the previous section, shows that any debt-financed fiscal expansion leads to large budget deficits, and to an undesirable accumulation of public debt in the long run unless taxes are increased. Any government carrying out a debt-financed fiscal expansion is likely to have to implement a fiscal contraction in the medium run, in order to restore fiscal sustainability. This may offset the
initial effects of the expansion on output.
Third, the policy prescriptions of the IS–LM model are also undermined by the insights arising from the life cycle and permanent-income hypotheses. These two theories point out that forward-looking consumers are little affected by temporary changes in income taxes unless they face liquidity constraints, in which case their consumption tends to be more influenced by current income. 1 Forward-looking consumers prefer to smooth their consumption over time, in the sense that they try to iron out temporary shocks to their current income, including tax cuts. But this forward-looking behaviour should also mitigate the effect of shocks on output and inflation, thus reducing the importance of fiscal stabilisation in the face of economic disturbances.
Finally, the IS–LM model predicts that debt-financed increases in public spending have more-than-proportional effects on output, since the government spending multiplier is greater than one, whereas balanced-budget expansions (tax-financed increases in public spending) increase output by the same proportion, since the balanced budget multiplier equals one. The Ricardian equivalence proposition, in contrast, suggests that both debt- and tax-financed increases in government spending raise output by the same proportion as one another. This is because forward- looking consumers are aware of the implications of the GBC, and therefore save more when the government pursues debt-financed policies. The exact structure of debt and taxation is irrelevant to them.
The Ricardian equivalence proposition rests upon very strict assumptions, such as infinitely long-living consumers and the absence of liquidity constraints. In reality some consumers have to cut spending as a result of tax increases, since they may be unable to borrow as much as their future expected income would require, so that their spending depends on their current income. With a finite horizon, any debt-financed tax cut might be offset only by a tax increase in the distant future. In this case Ricardian equivalence holds only as long as individuals care about future generations (their children). To the extent that this does not hold, or does not hold completely (such as for individuals with no children), the offsetting impact from future tax increases is less than complete.
The empirical evidence on consumers’ wealth indicates that a sizeable portion of the population have few financial means – bank accounts, bonds, and stocks – to smooth short-term consumption. In addition, low-income households, which normally have little or no saving and lack financial wealth or other collateral, have more limited credit-market opportunities, since they represent risky borrowers. For this reason, it seems plausible to assume that budget deficits have larger short-run effects on output and demand than those predicted by Ricardian equivalence, even though they tend to reduce output in the long run.
Active versus passive fiscal policy
As already discussed in the previous chapter, the traditional Keynesian view is in favour of an active use of demand-management policies. Keynes had emphasised that fiscal, rather than monetary, policy was the key macroeconomic instrument to help economies recovering from recession, in the belief that the actual IS curve was quite steep. Therefore monetary policy was presumed to be rather ineffective in altering demand and output, unlike fiscal policy which affects aggregate demand directly, and can fight recessions faster and more reliably.
This argument was strongly challenged at the beginning of the 1960s by the monetarist school, led by Milton Friedman, who argued against the active use of fiscal policy as a means to fine-tune the economy, on the grounds that fiscal actions are subject to prolonged inside and outside lags, which not only seriously undermine the success of stabilisation policies, but could even be the main source of macroeconomic instability.
In order to evaluate the effectiveness of fiscal policy actions, it is important to make a preliminary distinction between automatic and discretionary fiscal policy actions. Broadly speaking, the automatic stabilisers are the built-in components of the tax and spending system that respond to economic shocks without requiring any deliberate policy intervention. The term ‘automatic’ refers to the fact that these counter-cyclical fiscal effects occur as a result of business-cycle fluctuations, and do not require a specific decision of the government. During economic expansions, tax revenue rises as a result of increased labour income, profits and consumption, which increase the receipts resulting from income, corporate and consumption taxes respectively. Contemporaneously, the fall in unemployment automatically reduces social security spending, hence contributing to the overall reduction of the budget deficit during an economic expansion. The increase in tax receipts and the fall in government spending serve to reduce aggregate demand, and thus to contain output growth during expansions. Conversely, the simultaneous reduction in tax receipts and increase in public spending lead to an automatic increase in the budget deficit during a recession, hence stimulating aggregate demand and output.
**Discretionary fiscal policy interventions** refer to changes in government revenue and spending which are not automatically linked to the business cycle, but are the outcome of a conscious change in policy. The introduction of a new tax or the reduction of a tax rate are two examples of fiscal policy actions classified as discretionary fiscal changes. Automatic stabilisers have, by definition, zero inside lags. Discretionary fiscal policy is mainly subject to long decision and implementation lags, since policy changes normally have to pass through a potentially complex legislative process, even after they have been formally proposed by the relevant government department or ministry.
Both automatic stabilisers and discretionary fiscal policy interventions have relatively long and uncertain outside lags, even though these are likely to be shorter than the outside lags of monetary policy. Once implemented, tax and, especially, government spending directly affect aggregate demand, so they should in principle have a more rapid effect on output than monetary policy actions. Aside from inside and outside lags, the effectiveness of fiscal policy is also undermined by a lack of knowledge about the true structure and behaviour of the economy. As for monetary policy, this implies that discretionary fiscal policy must necessarily rely upon estimates and forecasts, and is thus subject to a high degree of uncertainty.
Finally, a further argument against the active use of fiscal policy as a tool of demand management is that the traditional Keynesian approach of policy evaluation does not take into account the impact of policy actions on agents’ expectations. As stressed by the Lucas critique, it is potentially misleading to use models estimated under one policy regime to predict the impact on the economy of policy changes, since the estimated parameters of the economic models may change when policy actions are implemented, as consumers and businesses revise their expectations in response.
Rules versus discretion
As for monetary policy, there is an ongoing debate as to whether governments should be allowed full discretion or whether they should commit to specific rules in their conduct of fiscal policy. To understand the current conduct of fiscal policy in the major OECD countries it is useful to take an historical perspective and briefly review how the approach to, and understanding of, fiscal policy have evolved over the last 60 years. Discretionary fiscal policy was the most prevalent tool employed for the conduct of demand management policies during the 1950s and 1960s. The international context in those two decades was characterised by the Bretton Woods regime, which pegged the exchange rates between the major world currencies to the US dollar, and by governments having as a main, if not only, objective the maintenance of full employment. In addition, in that period the predominant macroeconomic framework was the Mundell–Fleming model, which predicts fiscal policy to be more effective than monetary policy under fixed exchange rates.
Recall that the Mundell–Fleming model suggests that the effectiveness of fiscal policy depends upon the exchange rate regime and the degree of capital mobility. Under a fixed exchange rate regime the increase in the interest rate resulting from a fiscal expansion causes a capital inflow into the country. To keep the central parity, the domestic central bank is then forced to purchase foreign currency in exchange for domestic currency, and thus expand the money supply until the interest rate falls back to its level before the fiscal expansion. In turn, this reduction in the interest rate prevents a reduction in domestic investment, which makes fiscal policy even more effective than in a closed economy setting. In a flexible exchange rate regime the increase in the interest rate following a fiscal expansion leads to a capital inflow into the domestic economy and an appreciation of the exchange rate. The exchange rate appreciation crowds out net exports, thus offsetting part of the increase in output. Under perfect capital mobility the increase in output from the fiscal expansion is entirely crowded out by the reduction in net exports, and fiscal policy can only affect the composition of demand in the economy.
Given this analysis, discretionary fiscal policy was regarded as the best instrument to stimulate domestic demand during recessions in order to meet the goal of full employment, as well as to avoid balance of payment deficits and currency crises during expansions, when output and prices started to rise. However, the conduct of demand-management policies in those two decades had two main shortcomings. First, governments tended to act asymmetrically, in that they were keen to ease fiscal policy during recessions, but more reluctant to dampen demand during expansions.
Second, in many countries demand management was largely conducted on the revenue side, through tax changes while government spending tended to move pro-cyclically. These two factors, as well as the shortcomings discussed in the previous sections, contributed to limit the effectiveness of fiscal policy as a means of stabilisation, and possibly even to make fiscal policy an important source of internal instability.
The role of discretionary fiscal policy as a stabilisation tool diminished during the 1970s because of two important structural changes. First, the breakdown of the Phillips curve relationship resulted in a period of stagflation – high unemployment and inflation – which made demand- management policies even more destabilising, since the government’s attempt to maintain full employment increased prices even more. Second, the 1973 breakdown of the Bretton Woods regime resulted in a switch from fixed to floating exchange rates, in turn reducing the effectiveness of fiscal policy interventions.
A number of industrialised countries continued to use discretionary fiscal policy pro-cyclically throughout the 1980s. This resulted in large budget deficits and in the substantial accumulation of public debt, which also contributed to higher inflation during this decade.
Contemporaneously, macroeconomists pointed out the existence of a fundamental trade-off between the outcomes of short- and long-run fiscal policy actions, arising because policy-makers are not necessarily benevolent and may attempt to employ fiscal policy to influence their chances of re-election. For example, before an election the government may be tempted to cut taxes, in order to stimulate demand and promote growth, even though the economy is not in recession. Tax cuts generate deficits and debt accumulation, which must be cleared by higher taxes in the future. If higher tax rates have a negative impact on output, a tax cut may imply high growth in the short run but lower growth in the long run, when taxes will be increased to redeem the existing debt. If voters are short-sighted (i.e. they do not take into account the long-run consequences of policy actions), then policy-makers can successfully employ fiscal policy to win elections, resulting in higher growth before elections and low growth after elections. This is known as a political business cycle, in the sense that the economic business cycle is determined by the political cycle.
For these reasons, from the first half of the 1990s the main instrument for the conduct of stabilisation policy became monetary policy, and discretionary fiscal policy continued to play only a minor role as a stabilisation instrument, particularly as action was needed to restore fiscal sustainability.
More recently the principal role of fiscal policy in most OECD countries has been to ensure the sustainability of public finances over the medium term and, only secondarily, to support monetary policy in smoothing the path of the economy over the business cycle through the operation of the automatic stabilisers and, when appropriate, discretionary fiscal policy actions.
To guarantee the sustainability of the public finances over the long run and limit the negative consequences of discretionary fiscal actions, many governments have established specific rules for the conduct of their fiscal policies.
Fiscal rules are, however, very different across countries. For example, in the USA the 1990 Budget Enforcement Act enforced spending caps for discretionary spending on goods and services (and not transfer payments to individuals), for the subsequent five years only. It also enforced the so-called pay-as-you-go rule, which allowed the government to finance new transfer schemes only through an increase in tax revenue or by reductions in existing transfer programmes.
In Europe, the 1991 Maastricht Treaty established the following two rules for the conduct of fiscal policy:
1. The deficit-to-GDP rule states that the annual deficit-to-GDP ratio cannot exceed 3 per cent, unless either the ratio has declined substantially and reached a level close to 3 per cent or, alternatively, the excess over 3 per cent is only exceptional and temporary.
2. The debt-to-GDP rule states that the debt-to-GDP ratio should not exceed 60 per cent, unless the ratio is sufficiently diminished and is approaching 60 per cent at a satisfactory pace.
Finally, in the United Kingdom the 1998 Code for Fiscal Stability set out two rules against which the performance of the government is judged:
1. The ‘golden rule’, which says that the government can borrow over the business cycle only to invest and not to fund current spending.
2. The ‘sustainable investment rule’, which states that the average debt- to-GDP ratio over the business cycle cannot exceed 40 per cent.
But the difficulty with rules like these is that governments can always decide subsequently to break them. Indeed, in the wake of the 2008 financial crisis the debt-to-GDP ratios of a number of European countries – including France, Spain, Ireland and the UK – have climbed well above the formal 60 per cent limit set out in the Maastricht Treaty. In the case of the UK, this is of course a higher ceiling than the 40 per cent limit introduced in 1998, which has therefore also been violated. The question of how best to ensure that public finances remain sustainable in the long term remains an important area of debate among macroeconomists.
Seignorage and tax revenue
In this section we highlight the important link existing between inflation and fiscal policy. To this end, we need to extend the definition of the GBC
in equation (13.1) to include the possible use of the money supply among the sources of financing the budget deficit:
P t G t − P t T t + i t P t − 1 B t − 1 = P t B t − P t − 1 B t − 1 + M t − M t − 1 .
Equation (13.4) is also referred to as the consolidated GBC. Its last term represents the change in the supply of money, which is a third source
from which the government can finance public spending, as well as taxes and debt issues.
By dividing both sides of equation (13.4) by nominal GDP, P t Y t , and rearranging, the consolidated GBC can be written as:
g t + ρ t b t − 1 = t t + b t − b t − 1 + m t − ( m t -1 / (1 + PIt)(1 + gYt)
where m t = M t / ( P t Y t ) denotes the money supply as a proportion of nominal GDP. In particular, the last two terms in equation (13.5) represent
so-called seignorage revenue (i.e. the revenue collected by the government from money creation). Labelling seignorage (relative to GDP) with the symbol s t , the last two terms in equation (13.5) can be written as:
s t = m t − m t − 1 − ((m t-1 / (1 + PIt) ( 1 + g Yt )) + m t - 1
This is equivalent to
s t = ∆ m t + ((pi t + g Y)/ (( 1 + pi t )( 1 + g Yt ))) mt-1
Equation (13.6) shows that seignorage revenue arises from two sources. The first is simple increases in the money supply as a proportion of nominal income, ∆ m t . The second term is the outstanding stock of money interacted with inflation and real output growth,
(pi t + g Yt/ ( 1 + pi t ) ( 1+g Yt ))/ mt-1
This term is the main focus of most analysis of seignorage revenue, since it is nonzero even when the supply of money is unchanged.
Assuming that ∆ m t = 0 and ignoring output growth, so g Yt = 0, seignorage revenue can be written as:
s t = ( pi t / ( 1 + pi t )) mt-1
For small values of the inflation rate, the term pi t / ( 1 + pi t ) is approximately equal to the inflation rate pi t . As a result, seignorage revenue can be thought of as the product of the inflation rate pi t , equivalent to a tax rate, and the real stock of money supply m t–1 , equivalent to a tax base. From a policy perspective, equation (13.7) suggests that the government can finance public spending by raising the growth rate of the money supply, hence increasing inflation. This is because, unlike when borrowing by issuing bonds, when the government issues money it does not have to pay any interest on that issuance. The higher is the rate of inflation, the higher is the nominal interest rate, and thus the amount that the government saves relative to debt finance. In this respect, inflation becomes a tax on real balances and represents an important source of revenue for governments that are unable to entirely finance their spending either through taxes or by issuing debt.
There is, however, a limit to the extent to which seignorage can yield revenue for the government. This limit arises from the fact that the tax base of seignorage, the real supply of money, tends to shrink as inflation rises, because real money demand is negatively related to the nominal interest rate, which increases as inflation rises. In other words, the demand for real money balances falls as inflation increases. If indicating the demand for real money balances relative to GDP as m t = L | r t + pi t | , the ratio of seignorage to GDP can be written as:
s_t = (pi_t / (1 + pi_t)) L ( r_t-1 +/- pi_t-1) (13.8)
which shows that money growth tends to increase seignorage as it increases the inflation rate, but tends to reduce seignorage as it reduces money demand. But the empirical evidence suggests that the latter effect only begins to outweigh the former effect when the annual inflation rate is higher than about 200 per cent
Nevertheless, several countries have experienced in the past, for short periods of time, inflation rates well above those consistent with the maximum seignorage revenue, and even periods of hyperinflation, defined as a situation in which the monthly inflation rate exceeds 50 per cent. Blanchard and Johnson list seven famous examples of hyperinflations that occurred in European countries in the 1920s and 1940s. The main cause of most of these episodes of high and hyperinflation was a so-called budget crisis (i.e. a government inability to finance public spending through taxation or borrowing from the private sector). In this situation the government has to issue bonds to finance its deficit, but the only available buyer of these bonds is the domestic central bank. The central bank prints money to purchase the bonds, and the government uses this money to finance its deficit. This process is called debt monetisation.
There are several reasons why such crises occur: for example, a war or social unrest that destroys the ability of the government to raise revenue through taxes, or an adverse economic shock such as a fall in the price of raw materials exported by a country. In any of these cases governments suddenly become unable to collect revenue through taxes or the private sector refuses to purchase government bonds, and the only source of financing for public spending is through debt monetisation.
Phillip Cagan developed in 1956 a famous theory to explain why governments sometimes allow the inflation rate to rise even above the rate at which maximum seignorage can be collected. The logic lies in the dynamic evolution of the two components of equation (13.8). A fast increase in the growth rate of the money supply immediately raises the inflation rate. However, it takes time for the private sector to adjust its demand for money as a result of an increase in inflation, because money demand reacts slowly to changes in the inflation rate. Therefore in the short run seignorage revenue increases even if the inflation rate is above the rate at which maximum seignorage can be collected.
Note that because hyperinflations are the outcome of a bad conduct of fiscal policy, they normally end once governments implement appropriate fiscal reforms, mainly aiming at reducing public spending and raising more revenue through the tax system. This reduces the need to rely on seignorage revenue and allows a reduction in money growth.
Blanchard, O., Johnson, D.R., (2013) Macroeconomics (sixth edition), Pearson
Chapter 23: Fiscal Policy: A Summing Up
23-1 What We Have Learned
Let’s review what we have learned about fiscal policy:
* In Chapter 3 we looked at how government spending and taxes affected demand and output in the short run. We saw how, in the short run, a fiscal expansion—increases in government spending, or decreases in taxes—increases output.
* In Chapter 5 we looked at the short-run effects of fiscal policy on output and on the interest rate. We saw how a fiscal expansion leads to an increase in output and an increase in the interest rate. We also saw how fiscal policy and monetary policy can be used together to affect both the level and the composition of output.
* In Chapter 7 we looked at the effects of fiscal policy in the short run and in the medium run.
We saw that, in the medium run (that is, taking the capital stock as given), a fiscal expansion has no effect on output but is reflected in a different composition of spending. The interest rate is higher, and investment spending is lower.
* In Chapter 9 we looked at the case when the interest rate is equal to zero and the economy is in a liquidity trap.
We saw that, in this case, conventional monetary policy cannot be used to increase output, and thus fiscal policy has an even more important role to play.
* In Chapter 11 we looked at how saving, both private and public, affects the level of capital accumulation and the level of output in the long run.
We saw how, once capital accumulation is taken into account, a larger budget deficit, and, by implication, a lower national saving rate, decreases capital accumulation, leading to a lower level of output in the long run.
* In Chapter 17 we returned to the short-run effects of fiscal policy, taking into account not only fiscal policy’s direct effects through taxes and government spend- ing, but also its effects on expectations.
We saw how the effects of fiscal policy depend on expectations of future fiscal and monetary policy. In particular, we saw how a deficit reduction may, in some circumstances, lead to an increase in output, even in the short run.
* In Chapter 19 we looked at the effects of fiscal policy when the economy is open in the goods market.
We saw how fiscal policy affects both output and the trade balance, and we examined the relation between the budget deficit and the trade deficit. We saw how fiscal policy and exchange rate adjustments can be used together to affect both the level of output and its composition.
* In Chapter 20 we looked at the role of fiscal policy in an economy open in both goods markets and financial markets.
We saw how, when capital is mobile, the effects of fiscal policy depend on the exchange rate regime. Fiscal policy has a stronger effect on output under fixed exchange rates than under flexible exchange rates.
* In Chapter 22 we looked at the problems facing policy makers in general, from uncertainty about the effects of policy to issues of time consistency and credibility. These issues arise in the analysis of fiscal policy as well as monetary policy. We looked at the pros and cons of putting restraints on the conduct of fiscal policy, from spending caps to a constitutional amendment to balance the
23-2 The Government Budget Constraint: Deficits, Debt, Spending, and Taxes
To answer these questions, we must begin with a definition of the budget deficit. We can write the budget deficit in year t as deficit t = r B t - 1 + G t - T t
If the government runs a deficit, government debt increases. If the government runs a surplus, government debt decreases.
Using the definition of the deficit (equation (23.1)), we can rewrite the government budget constraint as
B t - B t - 1 = r B t - 1 + G t - T t
It is often convenient to decompose the deficit into the sum of two terms:
Interest payments on the debt, r B t - 1 .
The difference between spending and taxes, G t - T t . This term is called the primary deficit (equivalently, T t - G t is called the primary surplus).
* If government spending is unchanged, a decrease in taxes must eventually be offset by an increase in taxes in the future.
* The longer the government waits to increase taxes, or the higher the real interest rate is, the higher the eventual increase in taxes must be.
* The legacy of past deficits is higher government debt.
* To stabilize the debt, the government must eliminate the deficit.
* To eliminate the deficit, the government must run a primary surplus equal to the interest payments on the existing debt. This requires higher taxes forever.
The change in the debt ratio over time (the left side of the equation) is equal to the sum of two terms:
* The first term is the difference between the real interest rate and the growth rate times the initial debt ratio.
* The second term is the ratio of the primary deficit to GDP.
The increase in the ratio of debt to GDP will be larger:
* the higher the real interest rate,
* the lower the growth rate of output,
* the higher the initial debt ratio,
* the higher the ratio of the primary deficit to GDP.
Building on this relation, the Focus box “How Countries Decreased Their Debt Ratios after World War II” shows how governments that inherited very high debt ratios at the end of the war steadily decreased them through a combination of low real interest rates, high growth rates, and primary surpluses.
23-3 Ricardian Equivalence, Cyclical Adjusted Deficits, and War Finance
How does taking into account the government budget constraint affect the way we should think of the effects of deficits on output? One extreme view is that once the government budget constraint is taken into account, neither deficits nor debt have an effect on economic activity! This argument is known as the Ricardian equivalence proposition. While Ricardo stated the logic of the argument, he also argued there were many reasons why it would not hold in practice. In contrast, Barro argued that not only the argument was logically correct, but it was also a good description of reality.
Under the Ricardian equivalence proposition, a long sequence of deficits and the associated increase in government debt are no cause for worry. As the government is dissaving, the argument goes, people are saving more in anticipation of the higher taxes to come. The decrease in public saving is offset by an equal increase in private saving. Total saving is therefore unaffected, and so is investment. The economy has the same capital stock today that it would have had if there had been no increase in debt. High debt is no cause for concern.
23-4 The Dangers of High Debt
Most of the time, fiscal and monetary policies proceed independently. The government finances its deficit through borrowing. The central bank chooses the supply of money so as to achieve its objective (for example, low inflation). But, when the fiscal situation is bad, either because deficits are large or debt is high, and the interest rate faced by the government is high, it becomes increasingly tempting for the government to want to finance itself through money finance. Fiscal policy then determines the behavior of the money supply, a case known as fiscal dominance.
When the central bank finds itself in the fiscal dominance case, the central bank must do what the government tells it to do. The government issues new bonds and tells the central bank to buy them. The central bank then pays the government with the money it creates, and the government uses that money to finance its deficit. This process is called debt monetization - and can lead to hyperinflation.
Problems with hyperinflation
The transaction system works less and less well. One famous example of inefficient exchange occurred in Germany at the end of its hyperinflation: People actually had to use wheelbarrows to cart around the huge amounts of currency they needed for their daily transactions.
Price signals become less and less useful: Because prices change so often, it is difficult for consumers and producers to assess the relative prices of goods and to make informed decisions. The evidence shows that the higher the rate of inflation, the higher the variation in the relative prices of different goods. Thus the price system, which is crucial to the functioning of a market economy, also becomes less and less efficient.
Swings in the inflation rate become larger. It becomes harder to predict what in- flation will be in the near future, whether it will be, say, 500% or 1,000% over the next year. Borrowing at a given nominal interest rate becomes more and more of a gamble. If we borrow at, say, 1,000% for a year, we may end up paying a real inter- est rate of 500% or 0%: a large difference! The result is that borrowing and lending typically come to a stop in the final months of hyperinflation, leading to a large decline in investment.
Chapter 15: Stabilization Policy
15-1 Should Policy Be Active or Passive?
Making economic policy, however, is less like driving a car than it is like piloting a large ship.
Like a ship’s pilot, economic policymakers face the problem of long lags. Indeed, the problem for policymakers is even more difficult, because the lengths of the lags are hard to predict.
The inside lag is the time between a shock to the economy and the policy action responding to that shock. This lag arises because it takes time for policymakers first to recognize that a shock has occurred and then to put appropriate policies into effect. The outside lag is the time between a policy action and its influence on the economy. This lag arises because policies do not immediately influence spending, income, and employment.
Monetary policy has a much shorter inside lag than fiscal policy, because a central bank can decide on and implement a policy change in less than a day, but monetary policy has a substantial outside lag. Monetary policy works by changing the money supply and interest rates, which in turn influence investment and aggregate demand. Many firms make investment plans far in advance, however, so a change in monetary policy is thought not to affect economic activity until about six months after it is made.
Some policies, called automatic stabilizers, are designed to reduce the lags associated with stabilization policy. Automatic stabilizers are policies that stimulate or depress the economy when necessary without any deliberate policy change. For example, the system of income taxes automatically reduces taxes when the economy goes into a recession, without any change in the tax laws, because individuals and corporations pay less tax when their incomes fall.
One way forecasters try to look ahead is with leading indicators. Another way forecasters look ahead is with macroeconometric models, which have been developed both by government agencies and by private firms for forecasting and policy analysis.
In his writings on macroeconomic policymaking, Lucas has emphasized that economists need to pay more attention to the issue of how people form expectations of the future. Expectations play a crucial role in the economy because they influence all sorts of behavior. Lucas has argued that traditional methods of policy evaluation—such as those that rely on standard macroeconometric models—do not adequately take into account the impact of policy on expectations. This criticism of traditional policy evaluation is known as the Lucas critique.
15-2 Should Policy Be Conducted by Rule or by Discretion?
Policy is conducted by rule if policy-makers announce in advance how policy will respond to various situations and commit themselves to following through on this announcement. Policy is con-ducted by discretion if policymakers are free to size up events as they occur and choose whatever policy they consider appropriate at the time.
We begin this section by discussing why policy might be improved by a commitment to a policy rule. We then examine several possible policy rules.
Distrust of Policymakers and the Political Process
The Time Inconsistency of Discretionary Policy. In some situations policymakers may want to announce in advance the policy they will follow to influence the expectations of private decisionmakers. But later, after the private decisionmakers have acted on the basis of their expectations, these policymakers may be tempted to renege on their announcement. Understanding that policymakers may be inconsistent over time, private decisionmakers are led to distrust policy announcements.
Rules for Monetary Policy
Even if we are convinced that policy rules are superior to discretion, the debate over macroeconomic policy is not over. If the Fed were to commit to a rule for monetary policy, what rule should it choose? Let’s discuss briefly three policy rules that various economists advocate.
15-3 Conclusion: Making Policy in an Uncertain World
1. Advocates of active policy view the economy as subject to frequent shocks that will lead to unnecessary fluctuations in output and employment unless monetary or fiscal policy responds. Many believe that economic policy has been successful in stabilizing the economy.
2. Advocates of passive policy argue that because monetary and fiscal policies work with long and variable lags, attempts to stabilize the economy are
likely to end up being destabilizing. In addition, they believe that our present understanding of the economy is too limited to be useful in formu-
lating successful stabilization policy and that inept policy is a frequent source of economic fluctuations.
3. Advocates of discretionary policy argue that discretion gives more flexibility to policymakers in responding to various unforeseen situations.
4. Advocates of policy rules argue that the political process cannot be trusted. They believe that politicians make frequent mistakes in conducting economic policy and sometimes use economic policy for their own political ends. In addition, advocates of policy rules argue that a commitment to a fixed policy rule is necessary to solve the problem of time inconsistency.