Sovereign Risk, Fiscal Policy, and Macroeconomic Stability
  • 1 https://isni.org/isni/0000000404811396, International Monetary Fund
  • | 2 https://isni.org/isni/0000000404811396, International Monetary Fund

This paper analyzes the impact of strained government finances on macroeconomic stability and the transmission of fiscal policy. Using a variant of the model by Curdia and Woodford (2009), we study a "sovereign risk channel" through which sovereign default risk raises funding costs in the private sector. If monetary policy is constrained, the sovereign risk channel exacerbates indeterminacy problems: private-sector beliefs of a weakening economy may become self-fulfilling. In addition, sovereign risk amplifies the effects of negative cyclical shocks. Under those conditions, fiscal retrenchment can help curtail the risk of macroeconomic instability and, in extreme cases, even stimulate economic activity.

Abstract

This paper analyzes the impact of strained government finances on macroeconomic stability and the transmission of fiscal policy. Using a variant of the model by Curdia and Woodford (2009), we study a "sovereign risk channel" through which sovereign default risk raises funding costs in the private sector. If monetary policy is constrained, the sovereign risk channel exacerbates indeterminacy problems: private-sector beliefs of a weakening economy may become self-fulfilling. In addition, sovereign risk amplifies the effects of negative cyclical shocks. Under those conditions, fiscal retrenchment can help curtail the risk of macroeconomic instability and, in extreme cases, even stimulate economic activity.

1 Introduction

In the wake of the global financial crisis, sovereign risk premia have risen sharply in several countries, prompting policymakers to start fiscal tightening even as private demand remains weak. What are the likely consequences for economic activity? The present paper assesses this question quantitatively, starting from the observation that sovereign funding strains tend to spill over into private credit markets.1 Through this sovereign risk channel, higher public indebtedness may adversely affect economic activity by raising private-sector financing costs. Conversely, upfront fiscal retrenchment may help improve credit conditions in the broader economy, thereby counteracting the direct contractionary effect of lower public spending. Recent developments in Europe provide clear evidence in support of spillovers from sovereign risk to broader financial conditions. The panels in Figure 1 display time-series data on credit default swap (CDS) spreads for government debt and nonfinancial corporate debt.2 The figure compares two sets of euro area countries: those with relatively low sovereign spreads (left panel) and those with relatively high sovereign spreads (Belgium, Greece, Ireland, Italy, Portugal, and Spain).3 The series display substantial comovement, particularly in countries that face fiscal strain (right panel). For the time period shown, the daily correlation between corporate and sovereign CDS spreads in high-spread countries is 0.71. For the low-spread countries, it is lower, but still significantly positive at 0.36 percent.

Figure 1:
Figure 1:

Sovereign and nonfinancial corporate CDS spreads

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: 5-year CDS spreads in low-spread and high-spread euro area countries, as well as for nonfinancial corporations headquartered there. Low-spread euro area includes Austria (number of firms in our sample: 1), Finland (1), France (24), Germany (18), and Netherlands (8). High-spread euro area includes Belgium (number of firms: 1), Greece (1), Ireland (0), Italy (4), Portugal (2), and Spain (4). The corporations in our sample are the constituents of the Itraxx Europe index. The same relative weights are adopted for the sovereign and corporate index series. For example, of the 52 firms in the low-spread euro area sample, 24 are headquartered in France. As a result, in the sovereign low-spread euro area series, France has a weight of 24/52. Data sources: Bloomberg; Markit.

In this paper, we formally explore the implications of the sovereign risk channel. We build on the model proposed by Cúrdia and Woodford (2009), in which heterogeneous households engage in borrowing and lending via financial intermediaries. Our variant of the model features two critical innovations. First, we allow for sovereign risk premia that respond to changes in the fiscal outlook of the country. Although the precise numerical relationship is uncertain and likely to vary over time, the basic premise that risk premia are affected by fundamentals should be uncontroversial. Second, private credit spreads rise with sovereign risk because strained public finances raise the cost of financial intermediation. This way of modeling spillovers allows for a tractable representation of the sovereign risk channel within a simple variant of the canonical New Keynesian model. Consequently, we are in a position to supplement our numerical results with analytical solutions for interesting special cases.

The sovereign risk channel amplifies the transmission of shocks to aggregate demand, unless monetary policy manages to offset the spillover from sovereign default risk to private funding costs. Offsetting higher sovereign risk premia would typically require cuts in the policy rate. However, the central bank’s capacity to enact such cuts may be hampered, most notably if the nominal interest rate is already at the zero lower bound (ZLB), as has been the case for several major central banks in recent years. Our model is developed with an explicit reference to this ZLB problem. Yet it is important to stress that the ZLB is but one prominent example of a constraint on central bank action. Monetary policy would be similarly constrained—and our analysis would carry through—under a currency peg or in other situations where political or institutional considerations prevent the central bank from counteracting a rise in sovereign risk premia.

Our analysis of the sovereign risk channel gives rise to two distinct sets of results for an environment where monetary policy is constrained. First, under these circumstances sovereign risk may give rise to indeterminacy, or belief-driven equilibria. Specifically, to the extent that a pessimistic shift in expectations (unrelated to fundamentals) implies an upward revision of the projected government deficit, the risk premium on public debt rises and, through the sovereign risk channel, spills over to private borrowing costs. Higher private funding costs, in turn, slow down activity, validating the initial adverse shift in expectations. Under normal circumstances, this scenario could be averted by the central bank’s commitment to appropriately lower the policy rate. To the extent that monetary policy is constrained, however, expectations may become self-fulfilling, especially when sovereign risk is very high. In this scenario, we also find that the anticipation of a procyclical spending response—that is, fiscal tightening in response to a cyclical fall in tax revenue—can help to ensure determinacy.

Our second set of results suggests that (under equilibrium determinacy) the sign and the size of the government spending multiplier depend critically on the state of the economy. When the central bank is unconstrained, the sovereign risk channel is not operative in our model, as looser monetary policy can offset the impact of higher risk premia. By contrast, when the central bank is constrained, the sovereign risk channel typically reduces the spending multiplier. Fiscal retrenchment, by improving the outlook for public finances, may thus have less adverse effects on economic activity relative to a situation where sovereign risk is absent. These departures from standard macroeconomic results remain fairly modest as long as sovereign risk is contained. Only when public finances are very fragile, sovereign risk is high, and monetary policy is constrained for an extended period, does the sovereign risk channel unfold a sizeable impact on economic outcomes. The more severe the prevailing strain on public finances, the less fiscal retrenchment will weigh on growth. In relatively extreme cases, tighter fiscal policy may even stimulate economic activity.

As a caveat we emphasize that the present paper is not meant to add to the theory of sovereign default. Following Eaton and Gersovitz (1981), a number of authors, including Arellano (2008) and Mendoza and Yue (2010), have recently modeled default as a strategic decision of a sovereign that balances the gains from foregone debt service against the costs of exclusion from international credit markets. In equilibrium this implies that the probability of default increases in the level of debt. In order to maintain the tractability of our model for business cycle analysis, we impose such a relationship without explicitly modeling a strategic default decision. Specifically, we link the sovereign risk premium to the expected path of public debt (or, alternatively, future fiscal deficits). We thereby abstract from a number of other factors that may also affect the market’s assessment of sovereign risk, such as the quality of fiscal institutions or the composition of the investor base for government bonds. Implicit in our approach to modeling sovereign risk is the assumption that there are limits to credible commitment on the part of fiscal policymakers; otherwise, there would be no risk premium in the first place, and policymakers seeking to protect growth would arguably prefer to delay retrenchment until the economy is on a firm recovery path.

The remainder of the paper is structured as follows. Section 2 describes the model economy and presents our calibration. Sections 3 and 4 report analytical results and results from model simulations, respectively. Section 5 concludes.

2 The model

The key motivation for our model is the observation that sovereign risk systematically affects private-sector borrowing conditions. The model, therefore, needs to account for the possibility that private-sector borrowing and lending take place in equilibrium. We rely on the framework developed by Cúrdia and Woodford (2009) (CW, henceforth), which gives rise to an interest rate spread in an otherwise standard New Keynesian model. The spread emerges as a result of heterogeneity among households and because of costly financial intermediation. By assuming asymptotic risk sharing, CW are able to maintain the tractability of the New Keynesian baseline model. We add to their model a slightly richer specification of fiscal policy and allow the state of public finances to affect financial intermediation. In the following we briefly outline the model and stress the instances in which we depart from the original CW formulation.

2.1 Households

The economy is populated by a unit measure of households indexed by i[0,1]. Household i is of one of two types, indexed by superscript τt(i){b,s}. In equilibrium, households of type τt(i) = b will be “borrowers,” and households of type s will be “savers.” Households infrequently change their type. In each period, the probability of redrawing a type is 1 - δ, where δ ∈ (0, 1). Conditional on redrawing, the household will end up being a borrower with probability πb, and a saver with probability πs=1πb. The objective of household i[0,1] is given by

E0Σt=0(etβt)[(ξτ)στ1[Ct(i)]1σ1τ1στ1ψτ1+νht(i)1+ν],

where ct (i) is an aggregate of household expenditures:

ct(i)=[01ct(j,i)θ11θdj]θθ1;θ>1.(1)

Here ct(j, i) is a differentiated output good produced by firm j ∈ [0, 1]. ht(i) denotes hours worked by the household. et is a unit-mean shock to the time-discount factor β ∈ (0, 1), and ξττ, ψτ, and ν are positive parameters.

Households are able to insure against idiosyncratic risk through state-contingent contracts. Yet the resulting transfer payments are assumed to occur infrequently, namely only in those periods in which a household is assigned a new type. Meanwhile, households may borrow or save through financial intermediaries. The beginning-of-period wealth of household i is given by

At(i)=[Bt1(i)]+(1+it1d)+[Bt1(i)](1+it1d)+(1ϑt)Bt1g(i)(1+it1g)+Dtint+Tt(i)+Ttc.(2)

Here [Bt-1(i)]+ denotes deposits at financial intermediaries at the end of the previous period, which earn the deposit rate it1d;[Bt1(i)] denotes debt at financial intermediaries, which charge the borrowing rate it1b. In equilibrium, household i is either borrowing or saving. In the case where it is saving, the household may also hold government debt Bt1g(i)0. We depart from CW by assuming that government debt is not riskless: in any period, the government may honor its debt obligations, in which case ϑt = 0; or it may partially default, in which case ϑt = ϑdef, with ϑdef ∈ [0, 1) indicating the size of the haircut. it1g is the notional interest rate on government debt. Dtint are profits from competitive financial intermediaries that are distributed across households in a lump-sum manner. Tt(i) denotes transfers resulting from state-contingent contracts (which are zero for those households that do not redraw their type and are therefore temporarily without access to the payoff scheme implied by asymptotic risk sharing). Ttc is a lump-sum transfer that, in case of a sovereign default, compensates bond holders for losses associated with the sovereign default. Yet the payment is not proportional to the size of an individual’s holdings of government debt (see Schabert and van Wijnbergen (2008) for a similar setup). This assumption, along with the risk of a haircut, drives a wedge between the risk-free rate, itd, and the interest rate on government debt, itg.

The end-of-period wealth of household i is given by

Bt(i)=At(i)Ptct(i)+(1τtω)Ptωtht(i)+DtTtg.(3)

Pt denotes the consumption price index, τtω is the labor tax rate, and ωt is the economy-wide real wage rate; Dt are profits earned by intermediary goods producers and Ttg are lump-sum transfers by the government.

Assuming identical initial wealth for all households, state-contingent contracts ensure that post-transfer wealth is identical for all households that are selected to redraw their type. It is given by

At=[dt1(1+it1d)+(1ϑt)bt1g(1+it1g)bt1(1+it1b)]Pt1+Dtint+Ttc,(4)

where btg denotes government debt in real terms. dt denotes aggregate savings deposited with intermediaries, and bt denotes aggregate private borrowing, both in real terms. The latter evolves according to

bt=δbt1(1+ωt1)(1+it1d)/Πtπbωtbt+πb[δbt1g(1+it1g)/Πtbtg](5)+πbπs[(ctbcts)(1τtω)(ωthtbωthts)].

Intuitively, the accumulation of debt depends on four terms. The first term is the last period’s private debt level plus interest (for those households that do not redraw their type). The second term, -πbωtbt is the gain accruing to borrowing households from fraudulent loans (discussed below). The third term captures whether sovereign indebtedness (suitably adjusted for the change in household types) falls. In order to reduce sovereign indebtedness, current taxes need to be relatively high, which increases the need for borrowing by borrowers. Put differently, if sovereign indebtedness falls, so that [δbt1g(1+it1g)/Πtbtg]>0, more resources are made available by savers to borrowers since savers resort more to private sources for storing value. The last term captures the difference in consumption levels relative to the difference in wage income across household types.

Turning to the intertemporal consumption decisions, note that, as a result of asymptotic risk sharing, all households of a specific type have a common marginal utility of real income, λtτ, and choose the same level of expenditure:

cts=ξs(λts)σs(6)
ctb=ξb(λtb)σb.(7)

The optimal choices regarding borrowing from and lending to intermediaries, as well as to the government, are then governed by the following Euler equations:

etλts=βEt[et+11+itdΠt+1{(1δ)πbλt+1b+[δ+(1δ)πs]λt+1s}],(8)
etλts=βEt[et+1(1ϑt+1)(1+itg)Πt+1{(1δ)πbλt+1b+[δ+(1δ)πs]λt+1s}],(9)
etλtb=βEt[et+11+itbΠt+1{(1δ)πsλt+1s+[δ+(1δ)πb]λt+1b}].(10)

Optimal labor supply by households, in turn, is given by

hts=(λtsψs(1τtω)ωt)1/ν,(11)
htb=(λtbψb(1τtω)ωt)1/ν,(12)

Across household types, average labor supply, ht=πbhtb+(1πb)hts, is given by

ht=(Λtψ(1τtω)ωt)1/ν,(13)

where

Λt:=ψ[πb(λtbψb)1/ν+πs(λtsψs)1/ν]ν(14)

and ψ1/ν=πbψb1/ν+πsψs1/ν. Finally, for future reference we define

λt=πbλtb+(1πb)λts(15)

as the average marginal utility of real income across types.

2.2 Financial intermediaries

Saving and borrowing across households of different types takes place through perfectly competitive financial intermediaries. As in CW, we assume that an interest rate spread emerges because financial intermediation requires resources, Ξtbt, and because in each period a fraction of loans, Xt, cannot be recovered, irrespective of the characteristics of borrowers (due to, say, fraud). Moreover, deposits, dt, are assumed to be riskless and intermediaries collect the largest quantity of deposits that can be repaid from the proceeds of the loans that they originate, that is, (1+itd)dt=(1+itd)bt. The cash flow in period t of a financial intermediary is thus given by dtbtχtbtΞtbt. Using ωt to define the spread between lending and deposit rates, we have

1+ωt=1+itb1+itd.(16)

Substituting dt = (1 + ωt)bt, and choosing bt to maximize the profits of the intermediary yields the first-order condition for loan origination

ωt=χt+Ξt.(17)

In departing from CW, we assume that either χt or Ξt depends on sovereign risk. This assumption captures the increased difficulties in monitoring and enforcing loan contracts in an economy under fiscal strain. Conceptually related is the notion that in case of a sovereign default, the government diverts funds from the repayments made by borrowers, see Mendoza and Yue (2010).

Costs χtbt and Ξtbt differ in that only the latter are assumed to enter the economy’s resource constraint. For the linearized version of the model, used in Section 3, we let loan origination costs be covered by χt > 0, and set Ξt = 0, which facilitates deriving analytical results. For the dynamic simulations in Section 4 we set χt = 0 and let Ξt > 0. Specifically, we assume that either

χt=χψ[(1+itg)/(1+itd)]αψ1andΞt=0,(18)

or

χt=0andΞt=χψ[(1+itg)/(1+itd)]αψ1,(19)

where parameter χψ > 0 is used to scale the private spread in the steady state, and αψ measures the strength of the spillover from the (log) sovereign risk premium to the (log) private risk premium. Finally, transfers from intermediaries to households include loans that are not recovered by the intermediaries such that Dtint=Pt(ωtbtΞtbt).

2.3 Firms

There is a continuum of firms j ∈ [0, 1], each of which produces a differentiated good on the basis of the following technology

yt(j)=zth(j)1/ø,(20)

where zt is an aggregate productivity shock. In each period only a fraction (1 - α) of firms is able to reoptimize its prices. Firms that do not reoptimize adjust their price by the steady- state rate of inflation, II. Prices are set in period t to maximize expected discounted future profits.4 The resulting first-order condition for a generic firm that adjusts its price, Pt*, is

(Pt*Pt)1+θ(ø1)=KtFt,(21)

with

Kt=λtetμPøωt(ytzt)ø+αβEt[(Πt+1Π)θøKt+1],(22)
Ft=λtetyt+αβEt[(Πt+1Π)(θ1)Ft+1],(23)

where μp = θ/(θ - 1). The law of motion for prices (inflation) is given by

1α(ΠtΠ)θ1=(1α)(pt*Pt)1θ.(24)

For future reference it is also useful to define price dispersion Δt:=01(Pj,tPt)θødj, which evolves as follows

Δt=αΔt1(ΠtΠ)θø+(1α)(1α(Πt/Π)θ11α)θøθ1.(25)

Finally, profits distributed to households are given by Dt=01Pt(j)yt(j)Ptωtht(j)dj; or in equilibrium, dt=Pt(Ytωt(Yt/Zt)φΔt)

2.4 Government

Real government debt evolves as follows:

btg=(1ϑt)bt1g(1+it1g)Πt+gt+TtcPtTtgPtτtωωtht,

where gt denotes government spending. Below we will consider different assumptions regarding the law of motion for government spending. As is customary, we assume throughout that the expenditure share of each particular differentiated good in government spending is the same as the share of that good in private consumption. By assumption, transfers Ttc ensure that a sovereign default is neutral ex post in regard to any distributional consequences and the debt level. That is, under our assumptions, a sovereign default does not automatically ease the degree of fiscal strain. In particular, we set

Ttc=Ptϑtbt1g(1+it1g)Πt.

The consolidated government flow budget constraint is thus given by

btg=bt1g(1+it1g)Πt+gtTtgPtτtωωtht.(26)

Letting

trt=τtωωtht+Ttg/Pt(27)

denote the part of taxes that is related to the business cycle and to stabilization policy, we assume

(trtt)=[øT,y(yty)+øT,bg(bt1gbg)],øT,Y0,øT,bg>0.(28)

Throughout the paper, we assume that øT,bg is large enough so as to eventually stabilize public debt.5

While actual default is neutral in the sense described above, the probability of default is crucial for the pricing of government debt, itg, and for real activity.6 A fully specified model of sovereign default is beyond the scope of the present paper. Instead, we draw on earlier work in this area, which has pursued two distinct approaches. First, following Eaton and Gersovitz (1981), Arellano (2008) and others have modeled default as a strategic decision of the sovereign. Second, and more recently, Bi and Leeper (2010) and Juessen et al. (2011) consider default as the consequence of the government’s inability to raise the funds necessary to honor its debt obligations. Under both approaches, the probability of sovereign default is closely and nonlinearly linked to the level of public debt.

In the current paper we operationalize sovereign default by appealing to the notion of a fiscal limit in a manner similar to Bi and Leeper (2010). Whenever the debt level rises above the fiscal limit, default will occur. The fiscal limit is determined stochastically, capturing the uncertainty that surrounds the political process in the context of sovereign default. Specifically, we assume that in each period the limit will be drawn from a generalized beta distribution with parameters αbg, βbg, and b¯g,max. As a result, the ex ante probability of default, pt, at a certain level of sovereign indebtedness, btg, will be given by the cumulative distribution function of the (generalized) beta distribution:

Pt=Fbeta(btg4y1b¯g,max;αbg,βbg).(29)

Note that b¯g,max denotes the upper end of the support for the debt-to-GDP ratio. Since we keep the size of the haircut in case of default constant, we have

ϑt={ϑdefwithprobabilityPt,0withprobability1Pt.(30)

Turning to monetary policy, we assume throughout that the central bank follows a Taylor-type interest rate rule that also seeks to insulate aggregate economic activity from fluctuations in risk spreads, at least to some degree. In particular, we assume:

log(1+itd,*)=log(1+id)+øΠlog(Πt/Π)øωlog((1+ωt)/(1+ω)).(31)

Here, itd,* marks the target level for the deposit rate itd, and øΠ>1,øω>0. In deep recessions, however, the target level and the actual deposit rate can diverge. The reason is that in implementing rule (31), the central bank relies on steering the riskless nominal interest rate itd, which cannot fall below zero. Therefore, itd=itd* can only be implemented provided that itd0. Otherwise, itd=0. As a result, an increase in the spread ωt cannot be offset if monetary policy is constrained in lowering the policy rate.7

2.5 Market clearing and equilibrium

Good market clearing requires

yt=01ct(i)di+gt+Ξtbt=πbctb+πscts+gt+Ξtbt.(32)

The total supply of output is given by

ytΔt1/ø=ztht1/ø.(33)

In order to describe the equilibrium, we use equations (5)-(15), which characterize the solution to the household problem; equations (16)- (19), which characterize financial intermediation; equations (21)- (25), which characterize optimal price setting behavior; equations (26)-(30), which characterize the behavior of fiscal policy; the assumption about the evolution of labor taxes; the interest rate target rule (31) and the lower-bound constraint; and finally the good market clearing conditions (32) and (33). For given exogenous realizations of {et, gt, zt}, these equations pin down a sequence for the endogenous variables

{bt,btg,ctb,cts,χt,Δt,Ft,ht,htb,hts,itb,itd,itd,*,itg,Kt,λt,λtb,λts,Δt,ωt,Pt,Pt*/Pt,Πt,τtω,Ttg/Pt,trt,ϑt,ωt,Ξt,yt}.

2.6 Calibration

In order to solve the model numerically, we assign parameter values on the basis of observations for U.S. data. The relationship between sovereign risk, private-sector spreads, and the debt level is calibrated based on cross-country evidence. A time period in the model is one quarter.

With respect to monetary policy, we assume an average inflation rate of 2 percent per year. The coefficient on inflation in the Taylor rule is set to a customary value of øΠ = 1.5. We consider different values for parameter øω, the response of monetary policy to the interest rate spread. The specific parameterization will be discussed in the respective sections below. The steady-state level of government spending (consumption and investment) relative to GDP is g/y = 0.2. The level of gross public debt in the steady state is set to 60 percent of annual GDP. These values are broadly in line with U.S. averages over the last 20 years. In the baseline scenario, we set distortionary tax rates to zero and assume that the adjustment of taxes over the business cycle and in response to the debt level is achieved through lump-sum taxes. This assumption allows us to focus on the main channels of transmission in a transparent way while accounting for a feedback from economic activity to the fiscal outlook.8 We assume that taxes react to debt sufficiently strongly (øT,bg large enough) so as to ensure that the debt level remains bounded throughout and that øT,y = 0.34. This value is reasonable for the U.S., but at the lower end of estimates for other OECD countries; compare Girouard and André (2005).

With regard to the preference parameters, we set the curvature of the disutility of work to ν = 1/1.9, in line with the arguments provided by Hall (2009) regarding plausible values for the Frisch elasticity of labor supply. We set an elasticity of demand of θ = 7.6 so as to generate a gross price markup of μp = 1.15, which is in the range of customary values used in the literature. Finally, we assume that the average intertemporal elasticity of substitution σ¯=c/y, where σ¯:=πb(cb/y)σb+πs(cs/y)σs. . If the model had a representative household, this would correspond to the case of log-utility. Further, we assume that aggregate hours worked in the steady state are given by h = 1/3. We choose the relative values of the intertemporal elasticity of substitution for the two types of households (σb and σs) and of the scaling parameters for the disutility of work (ψb and ψs) such that the linearized model can be represented in the canonical three-equation New Keynesian format. This representation allows us to derive a number of analytical results in the next section. Importantly, under this calibration only the current value of the interest rate spread enters the dynamic IS- relationship and the New Keynesian Phillips curve. In addition, the evolution of output and inflation is independent of the level of private debt. Appendix A spells out in detail the conditions under which this representation is valid. Specifically, given the other parameter values, we set σbs = 0.53 and ψbs = 0.82. We explore to sensitivity of our results with respect to these assumptions through numerical simulations in Section 4.

We target a ratio of private debt to annual GDP, b/4y, of 80 percent, in line with Great Moderation averages for the U.S. More precisely, the figure refers to nonfinancial, nonmortgage, nongovernment credit market debt outstanding recorded in the U.S. flow of funds accounts. The same target is used by Cúrdia and Woodford (2009). Along with the market clearing condition, this determines scaling parameters ξb and ξs. Next, as in Cúrdia and Woodford (2009), we assume that households redraw their type on average every 40 quarters, giving δ = 0.975. This implies that the average time during which a specific type is without access to payoff streams from asymptotic risk sharing is 10 years.

A central element in our calibration is the share of borrowers in the economy, πb. This determines the share of economic activity that is affected by an increase in the spread and therefore deserves some discussion. One possible calibration would refer to the (U.S.) Survey of Consumer Finances. Averaging over the latest surveys (1998, 2001, 2004, and 2007), the share of U.S. families that hold some kind of debt is 76 percent; compare Aizcorbe et al. (2003) and Bucks et al. (2009). This suggests a value of πb = 0.76, and of πs = 0.24. However, loans secured by the primary residence make up a large share of that debt. This suggests that such a calibration for πb might overstate the importance of borrowing and the related effect that an increase in borrowing spreads could have on economic activity. Another metric, also from the Survey of Consumer Finances, that is more directly related to the notion of “borrowers” and “savers” in our model is that on average 57 percent of families in the survey report that – over the year preceding each survey date – they have been spending less than their income, that is, they have saved. This suggests a value for πb of 1 - 0.57, or πb = 0.43. That said, both of the aforementioned figures do not explicitly take into account the borrowing by firms in the economy (other than by single-owner firms). To the extent that households in our model own firms and also make the intertemporal decisions for these firms, any purely household-based measure of indebtedness is likely to underestimate the degree of indebtedness and thereby the importance of the borrowing spread. In particular, using the same measure of private borrowing as above (nonfinancial, nonmortgage, nongovernment credit market debt), 50 percent of private borrowing is accounted for by corporations rather than by households. To capture this, we set πb = (1 - 0.17). 0.43 + 0.17. 1 = 0.53 in our baseline calibration. This formula hypothetically divides households into consumption entities, some of which are indebted, and investment entities, all of which have debt. In the calculations, 0.17 is the share of nonresidential private domestic fixed and inventory investment in private domestic economic activity.

In regard to the normal spread between deposit and lending rates, we target a steady-state value of 2.1 percent (annualized), in line with commercial and industrial loan rate spreads in the Federal Reserve’s Survey of Terms of Business Lending. This pins down parameter χψ. The steady-state level for the central bank’s target interest rate, id, is set to 4.5 percent annualized, from which the time discount factor, β, follows.

Turning to the production parameters, we set ø = 1, implying a linear production function. We furthermore target a unit value for steady-state output; this pins down productivity, z. We set parameter α = 0.9 in order to generate a slope of the Phillips curve in line with the empirical evidence.9

Finally, it remains to determine the parameters that govern the spillover from sovereign risk premia to private-sector spreads. Actual haircuts in case of a sovereign default show large variation; see Panizza et al. (2009) and Moody’s Investors Service (2011). ϑdef = 0.5 appears to be a reasonable average value. With respect to the specification of the fiscal limit, we seek to replicate the relationship between the sovereign risk premium and public debt shown in Figure 2. The figure plots CDS spreads of industrialized economies against the level of projected gross debt of the general government (relative to GDP). The projections are taken from the IMF’s World Economic Outlook in April 2011. The blue dots show projections for the end of 2011. For comparison, the figure also plots IMF forecasts for the debt-to-GDP ratio by the end of 2015. For the countries shown in the figure, CDS spreads are systematically higher the higher the level of projected gross public debt.10 In fact, the risk premium appears to rise disproportionately as the debt level rises. We choose parameters αbg = 3.70, βbg = 0.54, and b¯g,max = 2.56 to match this empirical relationship. The black solid line displays the implied steady-state relationship between debt levels and the sovereign risk premium.

Figure 2:
Figure 2:

Sovereign risk premia vs. debt

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The figure shows 5-year sovereign CDS spreads against forecasts for end-2011 gross general government debt/GDP (blue circles) and end-2015 debt/GDP (green triangles). The countries shown are Australia, Austria, Belgium, Czech Republic, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Netherlands, Norway, Portugal, Slovak Republic, Slovenia, Spain, Sweden, United Kingdom, and United States. Excludes Japan. Forecasts are taken from the IMF World Economic Outlook April 2011.

Regarding the spillovers from sovereign to private-sector risk, Figure 1 is suggestive of a sovereign risk channel that runs from sovereign spreads to spreads in the household and corporate sector. Of course, there might be other reasons for the observed comovement, too. In the present paper, we abstract from these and interpret the comovement as caused by sovereign risk, consistent with standard accounts of the sovereign debt crisis in parts of the euro area. In regard to αψ, Harjes (2011) finds for a sample of large, publicly traded euro area companies that a 100-basis-point increase in sovereign spreads raises private firms’ credit spreads by about 50 to 60 basis points. As our baseline, we therefore set αψ, = 0.55. Although this value implies significant spillovers, it may still represent something closer to a lower bound for two reasons. First, it is based on credit spreads of companies that are large, with often sizeable export activities and access to international credit markets. Spillover effects from sovereign risk are likely to be more pronounced for smaller and less international companies that rely on local bank-based financing. Second, Figure 1 suggests that the comovement between spreads is considerably stronger in countries that face intense fiscal strain. Accordingly, the value of αψ, = 0.55 may understate the strength of the sovereign risk channel for highly indebted countries. For these reasons, we also consider higher values as we move through the simulations in the paper.

3 Sovereign risk, fiscal policy, and macroeconomic stability

In this section, we begin to analyze how sovereign risk affects macroeconomic dynamics and stabilization policy in the model outlined above. Our focus is on the stability properties of the economy, notably its vulnerability to self-fulfilling equilibria. In this context, we investigate in detail how the sovereign risk channel alters the transmission of fiscal policy. We show that monetary policy plays a key conditioning role, notably through its capacity to insulate private borrowing costs from fluctuations in sovereign risk premia. To capture this aspect we consider a scenario where monetary policy may be constrained by the ZLB. Our analysis proceeds in two steps. First, in this section, we consider a special case of our model for which we are able to obtain analytical solutions. For this case, we assume that the probability of sovereign default depends on the expected primary deficit, rather than on the level of debt. In the next section, we will turn to simulation-based results for the full model.

3.1 A tractable special case of the model

In this section, we focus on a first-order approximation of the equilibrium conditions around the deterministic steady state. The aggregate equilibrium dynamics of the model can be represented by a variant of the New Keynesian Phillips curve and a dynamic IS relationship. The former relates inflation to expected inflation, output, and government purchases:

Π^t=βEtΠ^t+1+κyy¯tκgg¯t,(34)

where κy=κ(ν+σ¯1) and κg=κσ¯1, with κ=(1α)(1αβ)α.11.11 In terms of notation, y¯t=yty,g¯t=gtg, and Π^t=log(Πt/Π), where variables without a time subscript refer to steady-state values.

The dynamic IS-relationship links output to real government spending and the effective real interest rate through

y¯tg¯t=Ety¯t+1Etg¯t+1σ¯[i^td+(πb+sΩ)ω^tEtΠ^t+1+Γt],(35)

whereb ω^t:=log((1+ωt)/(1+ω)),i^td:=log((1+itd)/(1+id)), and Γt:=Etlog(et+1)log(et). From the IS relationship, it is clear that fluctuations in the private-sector spread can influence economic activity unless they are neutralized by monetary policy. The degree to which the private-sector spread affects economic activity, for a given policy rate, is determined by parameters πb + sΩ. As discussed in Cúrdia and Woodford (2009), parameter sΩ:=πbπs(σbcb/yσscs/y)/σ¯ indicates the extent to which interest rate increases affect the aggregate demand by borrowers more adversely than that of savers. In our calibration, cb > cs and sΩ > 0.

As regards monetary policy, equation (31) implies that during normal times (in deviations from steady state):

i^td=øπΠ^tøωω^t.(36)

For the analytical results, we initially consider the case øω=(πb+sΩ), in which the central bank fully neutralizes the effect of the sovereign risk premium on aggregate economic outcomes, as apparent from the IS equation (35). However, monetary policy may not always be able to do so. In particular, we assume below that monetary policy becomes constrained such that the nominal interest rate is stuck at the lower bound of zero in the initial period of our simulations. This means that the central bank can no longer offset a rise in the interest rate spread. We follow Christiano et al. (2011) and Woodford (2011) in assuming that monetary policy will return to Taylor rule (36) in the next period and be unconstrained thereafter with probability 1 - μ, where μ ∈ (0, 1). Otherwise, the zero interest rate persists into the next period. The same Markov structure applies for the remaining periods. As a result, the expected length of the ZLB episode is given by 1/(1 - μ). Shocks to the time discount factor, Γt, follow the same Markov structure.12 Given that there are no endogenous state variables in the special case of our model considered here, once the ZLB situation ceases to persist, the economy immediately reverts to the steady state.

As indicated above, we make one further simplifying assumption in this section that allows us to derive analytical results. Specifically, we assume that the probability of sovereign default—and thus the sovereign risk premium—depends on the expected primary deficit rather than the level of public debt as in the full model. Consequently, the interest rate spread also depends on the expected deficit. We postulate a linear relationship of the form

ϖ^t=ξEt(g¯t+1øT,yy¯t+1),(37)

where, in order to ease the burden on notation, we have defined the spread that enters the IS relationship over and above the risk-free deposit rate as ϖ^t:=(πb+sΩ)ϖ^t. Parameter ξ ≥ 0 indicates the extent to which fiscal strain—as measured by primary deficits—spills over to private-sector spreads. Parameter øT, y ∈ [0, 1) measures the sensitivity of tax revenue with respect to economic activity.

3.2 The size of the spillover from public to private risk premia

To appreciate our results below, it is useful to discuss the range of plausible values for ξ in equation (37), which—through a sequence of back-of-the-envelope calculations—links to the fundamental parameters of the model. Let ξ′ be the slope of the risk premium with respect to debt at a specific debt level, evaluated in the steady state. Our assumptions regarding the sovereign spread in Section 2, in particular equation (29), imply that

ξ=αψ(πb+sΩ)ϑdef1ϑdefFbeta(bg4y1b¯g,max;αbg,βbg)14y1b¯g,maxfbeta(b4y1b¯g,max;αb,βb).(38)

The first column of Table 1 reports the resulting values for ξ′ at different debt levels using the calibration of the fiscal limit distribution discussed in Subsection 2.6.

Table 1:

Quantifying parameter ξ

article image
Notes: The table presents estimates for the slope of the average private interest rate with respect to the fiscal deficit, ξ, for different average lengths (in quarters) of the ZLB episode and for different debt/GDP ratios. The entries in the columns “ξ by length of ZLB episode (qtrs)” are based on the formula ξ=1+μ(1μ)μ(1μ)ξ that is explained in detail in the main text and in Appendix B.

These values appear to be fairly small. It needs to be borne in mind, however, that the relationship in equation (37) links the interest rate spread to the expected deficit, whereas the full model implies a link between the interest rate spread and the expected level of debt, i.e., the result of a series of accumulated deficits. The values for ξ′ are thus bound to understate the sensitivity of the interest rate spread to a persistent fiscal deficit. More precisely, an appropriate mapping from the slope of the risk premium into the simplified model environment needs to take into account the horizon over which deficits accumulate. The following expression is meant to capture this fact for empirically reasonable values of μ > 0.5 (so the ZLB is expected to be binding for at least two periods):13

ξ=1+μ(1μ)μ(1μ)ξ.(39)

The columns under “ξ by length of ZLB period” in Table 1 report the corresponding values of ξ for different initial debt levels if the ZLB has an expected duration of six, seven, or eight quarters. These calculations suggest that a value of ξ as high as 0.1 cannot be ruled out if the initial debt stock is large and the recessionary shock persists for an extended period. In particular, recall from Figure 1 that the spillovers may be noticeably stronger than usual for countries that face intense fiscal strain. A bigger spillover parameter, αψ, would scale up linearly the entries in Table 1.

3.3 Sovereign risk channel and equilibrium determinacy

Assume, as a baseline scenario, that the level of government spending is exogenously given. For this case, we find that the sovereign risk channel significantly alters the determinacy properties of the model while the ZLB is binding. Specifically, the range of parameters that ensure determinacy shrinks in the presence of sovereign risk. In the following, we establish distinct restrictions on parameters that yield a (locally) determinate equilibrium.14

Proposition 1 In the economy summarized by equations (34) – (37), let the interest rate be equal to zero in the initial period. In each subsequent period, let the interest rate remain at zero with probability μ ∊ (0, 1). Otherwise, let monetary policy be able to permanently return to Taylor rule (36). There is a locally unique bounded equilibrium if and only if

a)a<1/(βμ),andb)(1βμ)(1α)>μσ¯κy,

where a:=μ+μξøT,yσ¯andκy:=κ[ν+1/σ¯]..

Proof. See Appendix C. ▪

In the absence of an endogenous risk premium (i.e., for ξ = 0) as in Christiano et al. (2011) and Woodford (2011), condition a) is always satisfied. Hence there will be a unique bounded equilibrium if and only if condition b) holds. If ξ = 0, condition b) is given by (1βμ)(1μ)>μσ¯κy. The previous literature has shown that the set of “fundamental” parameters for which this condition holds is larger (i) the less persistent the ZLB situation (in our parameterization, the smaller μ); (ii) the lower the interest sensitivity of demand (the smaller σ¯); and (iii) the flatter the Phillips curve (the smaller κy). In addition to these findings, our analysis shows that the range of parameters for which the equilibrium is determinate shrinks in the presence of a sovereign risk channel. Specifically, with ξ > 0, condition a) is violated if either the interest rate spread is sufficiently responsive to the deficit or if tax revenue is sufficiently responsive to output (øT,y is large enough). Note that the same parameters are also key determinants for whether condition b) is satisfied. 15

It is instructive to contrast this baseline result with a situation where government spending adjusts endogenously to output while the economy is at the ZLB. The following proposition summarizes the pertinent conditions for the existence of a unique bounded equilibrium.

Proposition 2 In the economy specified in Proposition 1, let government spending g¯t take on a value of g¯t=φy¯t when the economy is at the ZLB, and g¯t=0 otherwise. Suppose further that ϕ > 1. Define a*:=μ+μξøT,y*σ¯*;κy*=κyφκg;øT,y*:=øT,yφ; and σ¯*=σ¯/(1φ). There exists a locally unique bounded equilibrium if and only if:

1. with a a* > 0

a)a*<1/(βμ),andb)(1βμ)(1a*)>μσ¯*κy*

2. with a* < 0:

a)(1+βμ)(1+a*)>μσ¯*κy*andb)(1βμ)(1a*)>μσ¯*κy*.

Proof. See Appendix C.▪

To appreciate the implications of this proposition, consider first the possibility that there is no sovereign risk channel (ξ = 0). In this case the range of parameters for which the equilibrium is determinate is larger if spending is countercyclical (ϕ < 0). With an endogenous risk premium, however, the opposite may hold. More precisely, if ξ > 0 and if, in addition, the conditions of Item 1 of Proposition 2 hold, then subject to some limits on the elasticity of taxes with respect to output, namely, øT,y<1κν(1βμ)ξ, the range of fundamental parameters for which the equilibrium is determinate is at least as large with a procyclical spending response, ϕ (0, 1), as without any response; indeed the range can even be larger. Note that this case is more likely the lower the elasticity of tax revenue to economic activity (the smaller øT,y), and the more strongly the interest rate spread responds to the deficit (the larger ξ). The main conclusion is straightforward, if unconventional: a procyclical fiscal stance may reduce the risk of equilibrium indeterminacy in the presence of sovereign risk.16

The two propositions above deserve some further discussion. While previous work has focused on the case in which government debt and a rising risk premium imply explosive debt, we consider a situation in which debt ultimately will always be stabilized through one-off tax measures (although sovereign default can still occur). In such an environment, we find that an economy with an endogenous risk premium and a constrained central bank can be prone to belief-driven equilibria. Moreover, we find that systematic spending cuts during recessions (when the ZLB is binding) may actually help to anchor expectations to a unique equilibrium. To see why, assume that during the ZLB period agents expect some drop in output. Lower output means less tax revenue and, in the absence of a fiscal response, higher deficits, which in turn imply a higher interest rate spread. As the widening of the interest rate spread cannot be offset by monetary policy at the ZLB, the real interest rate rises. Consequently, expectations of adverse output developments can become self-fulfilling in high-debt economies where a high and rising interest rate spread weighs heavily on output, thus confirming agents’ beliefs in equilibrium. By contrast, a procyclical fiscal stance may be sufficient to prevent an adverse expectational shock from confirming itself, because public spending cuts would offset the expected decline in tax revenue triggered by a fall in output.

Figure 3 illustrates the results of Propositions 1 and 2 for the baseline parameterization. Each panel of the figure displays results for a different value of μ implying, from left to right, an expected duration of the ZLB episode of 6, 7, and 8 quarters, respectively. We evaluate whether a unique equilibrium exists for different values of the slope of the risk premium, ξ, measured on the horizontal axis, and the response of government spending to output, ϕ, measured on the vertical axis. Grey areas indicate determinacy regions, while white areas indicate equilibrium indeterminacy. In the case of protracted ZLB episodes (right panel) with a steep slope of the risk premium, we find that there may be no countercyclical spending policy—of the simple form analyzed here—that ensures determinacy. However, a policy of procyclical spending cuts is not a panacea either; although it is helpful under the circumstances detailed above in reducing the risk of belief-driven equilibria, it does not ensure determinacy for all values of ξ under consideration.

Figure 3:
Figure 3:

Determinacy regions with endogenous response of government spending

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: Determinacy regions with endogenous response of government spending to economic activity during a deep recession. Grey areas mark parameterizations that imply determinacy. y-axis: response of government spending to output, φ(g¯t=φy¯t). x-axis: response of the interest rate spread to the deficit, ξ. From left to right: ZLB is expected to bind for 6, 7, or 8 quarters (μ = 5/6,6/7, 7/8).

On a final note, we have made the risk premium depend on the expected next-period deficit in this section, as apparent from (37). This conforms to the notion that sovereign spreads reflect a forward-looking assessment of the fiscal position. Quantitatively, however, it turns out that our results would be very similar if we had assumed that the sovereign spread depends on the current period’s deficit.

3.4 Output effects of government spending cuts

In the following we shift our focus to the output effects of exogenous government spending cuts. To that end, we limit our analysis to parameterizations for which a stable and unique equilibrium exists. Our results suggest that sovereign risk is a key determinant for the size, and even sign, of the fiscal multiplier. In the simplified model setup that we consider at the moment, the sovereign risk channel is governed by parameter ξ. However, since monetary policy may neutralize the effects of the sovereign risk channel outside the ZLB episode, parameter μ plays a key role in our analysis as well.

Following Woodford (2011) and Christiano et al. (2011), we focus on an economy at the ZLB and assume that government spending deviates from its steady-state level during the ZLB episode only, i.e., by taking on a value of g¯t=gL. Otherwise government spending is set to its steady-state level. Our key result is summarized by the following proposition.

Proposition 3 Under the conditions spelled out by Proposition 1 (which ensure that a locally unique bounded equilibrium exists), let shock Γt take on a value of Γ or as long as monetary policy is constrained to zero interest rates, and a value of 0 otherwise. Similarly, let government spending take on a value of gL when the economy is at the ZLB, and 0 otherwise. As before, define a=μ+μξøT,yσ¯ and b=μ+μσ¯ξ. Then, while monetary policy is constrained, output is given by

yL=ϑr(log(1+id)Γ)+ϑggL,

where

ϑr=σ¯(1βμ)(1βμ)(1a)μσ¯κy>0(40)

and

ϑg=(1βμ)(1b)μσ¯κg(1βμ)(1a)μσ¯κy.(41)

Proof. See Appendix C. ▪

Note that ϑg provides a measure for the government spending multiplier on output at the ZLB. It is characterized in more detail by Corollary 5 in Appendix D.2. Specifically, under the determinacy conditions established above, equation (41) implies that the multiplier is positive if and only if

(1μ)μσ¯κg1βμ>μξσ¯.(42)

If this condition is satisfied, a spending cut at the ZLB will reduce output, in line with conventional wisdom. If ξ = 0, i.e., in the absence of a sovereign risk channel, this will always be the case. Moreover, the government spending multiplier will be strictly greater than one, consistent with the analysis of Christiano et al. (2011) and Woodford (2011). In contrast, if ξ > 0, the government spending multiplier at the ZLB may actually become negative, such that spending cuts raise output.

The left panel of Figure 4 illustrates this result. It displays the output effect of a government spending cut during the ZLB episode for different levels of spillovers from sovereign risk, as measured by alternative values for ξ, and for different assumptions regarding the expected length of the recession, as measured by alternative values for 1/(1 - μ). The other parameters underlying these computations remain fixed as laid out earlier. In the same figure we also show the response of the budget deficit (middle panel) and the interest rate spread (right panel).

Figure 4:
Figure 4:

Effects of early retrenchment

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The figure shows the effects of a unit cut in government spending for the length of the ZLB episode. Effect on output (left panel), on the deficit (middle panel: negative means the deficit shrinks), and on the interest rate spread (right panel). On the axes: responsiveness of interest spread to expected deficit, ξ, and expected duration of ZLB episode: 1/(1 - μ). Only parameterizations that imply determinacy are shown. For better readability, multipliers and deficits are capped at the maximum level indicated in the charts.

The figure shows that the fiscal multiplier depends crucially on both dimensions under consideration. Consider first the case where ξ = 0, that is, a situation when there is no sovereign risk channel. In this case, a spending cut invariably causes a sizeable decline in output. In fact, for an expected duration of the ZLB episode of eight quarters, the government spending multiplier on output reaches a value as high as 3, a result recently stressed in Christiano et al. (2011). The underlying mechanism is well understood: the deflationary effect of spending cuts cannot be counteracted by a reduction in policy rates, thus causing an increase in the real interest rate which crowds out private demand. The effect is stronger the longer the expected duration of the ZLB episode, as private demand is determined by the expected path of current and future short-term real interest rates.

Turning to the scenario of an active sovereign risk channel, we focus first on a case where the ZLB period is expected to be short, namely four quarters only. In this case, as the interest rate spread becomes more responsive to the fiscal deficit, i.e., as ξ takes on bigger values, the fiscal multiplier tends to decline. In other words, output falls by less in response to a public spending cut. Still, the role of the sovereign risk channel is limited, even for very high values of ξ. This is due to the fact that monetary policy is expected to be able to offset the effect of sovereign risk on private interest rates in the near future.

The picture changes significantly if monetary policy is expected to be constrained for an extended period. Under those circumstances, the sovereign risk channel has a strong bearing on the fiscal transmission mechanism. In fact, if ξ and μ both take on sufficiently high values, the sign of the output multiplier turns negative, implying that a spending cut during the ZLB episode becomes expansionary. To understand this finding, it is useful to consider the responses of the deficit and the risk premium. Note that for almost all parameterizations, a cut in government spending reduces the deficit (see also Erceg and Lindé (2010b)). If fiscal strain is severe at the outset, the lower deficit leads to a considerable decline in the risk premium. This, in turn, reduces the interest rate spread, stimulates private demand, and raises tax revenue. The result is a virtuous cycle of an additional decline in interest rate spreads, increased economic activity, and a further improvement of the fiscal outlook. 17

In sum, our simplified model setup reveals a potentially important role for the sovereign risk channel. Although the model implies standard positive multipliers under normal conditions (i.e., public spending cuts depress short-term output), fiscal retrenchment may become less contractionary, and even turn expansionary, in the presence of severe fiscal strain if monetary policy is unable to cushion the adverse effects of sovereign risk on private-sector borrowing conditions.

As such, our results shed some fresh light on the theoretical possibility of “expansionary fiscal contractions,” which has attracted significant attention following the prominent study by Giavazzi and Pagano (1990), who consider the experience of Denmark and Ireland during the 1980s.18 In order to rationalize expansionary consolidations, previous theoretical accounts have often focused on the “expectations view,” whereby immediate fiscal retrenchment triggers a shift in expectations regarding the long-run level of spending, and thus a downward revision of the anticipated tax burden (Bertola and Drazen 1993, Sutherland 1997, and Perotti 1999). In a similar vein, Blanchard (1990) formalizes the idea that early consolidation may reduce the likelihood of more disruptive future adjustment, thus boosting current private spending. Giavazzi and Pagano themselves stress that monetary policy may also have played an important role in the Danish and Irish episodes. Specifically, both episodes were marked by credible exchange rate pegs, which arguably led to declining country risk premia and lower real interest rates. More recently, Erceg and Lindé (2010a) have analyzed government spending cuts in a model of a currency union where country risk is a function of the state of public finances.19 Our own analysis provides a novel contribution to this literature by highlighting a distinct set of economic conditions—high sovereign risk and a constrained central bank—under which spending cuts may have non-standard effects.20

4 Dynamic analysis

We now turn to a numerical analysis of the full model as outlined in Section 2 above. This allows us to revisit our analytical results while accounting for the possibility that sovereign risk depends on the expected debt level rather than the expected deficit. In order to highlight the role of monetary policy, we focus again on a ZLB scenario. However, we depart from the simplifying assumption that the expected duration of the ZLB episode is constant. Instead, we envisage a scenario in which a) the initial debt level matters for the depth of the recession; and b) fiscal retrenchment may alter the length of the ZLB episode.21

4.1 Deep recessions and sovereign risk

To set the stage for our analysis, we subject the model economy to a large shock that pushes monetary policy to the ZLB. Specifically, we consider a first-order autoregressive process for êt to capture in a stylized manner the output loss in the U.S. during the 2007–2009 recession. The Congressional Budget Office (2011) estimates that the output gap reached 6.7 percent in 2009 and that it will still be at a 1.7 percent level in 2014 and at 0.5 percent in 2015.22 For the simulations, we also assume that taxes stabilize the debt level, if only very gradually.23 While the simulations for the baseline scenario assume that taxes are raised in a lump-sum manner, Subsection 4.3 assesses the extent to which distortionary taxation would affect our conclusions.

Figure 5 shows the evolution of output, the interest rate spread, and the policy rate in response to the recessionary shock. It displays the behavior of the economy in the absence of any discretionary fiscal policy measure for three different initial levels of government debt: 60 percent of GDP (black solid line), 90 percent (blue dashed line), and 115 percent (red dots). The adjustment dynamics differ substantially in that the decline in output and the rise of the interest rate spread are much stronger if initial debt is high. To understand this result, note that the shock induces an increase in the budget deficit, which leads to a build-up of public debt and thus an increase in the sovereign risk premium. Moreover, since the shock pushes the economy to the ZLB, monetary policy is unable to offset the spillover from sovereign risk to private interest rates. Private expenditure thus falls further, compounding the initial decline in output. This effect is stronger the higher the initial debt level, because the relationship between public debt and sovereign risk is fundamentally nonlinear.

Figure 5:
Figure 5:

Deep recession with different initial levels of government debt

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The figure shows responses to the recessionary shock for different initial debt levels. Black solid line: 60 percent debt-to-GDP, blue dashed: 90 percent, red dots: 115 percent. Output is expressed in terms of percent deviations from the steady state, the risk premium in annualized basis points, and the interest rate in annualized percentage rates.

In our simulations, a persistently high risk premium also lengthens the time span over which the ZLB remains binding. This result is apparent from the dynamics of the policy rate shown in the right panel of Figure 5: the ZLB episode is extended by as much as 11 quarters if government debt is high. Hence, our dynamic analysis delivers a first additional insight: not only does the importance of the sovereign risk channel depend on whether the central bank is constrained in steering interest rates, but sovereign risk may itself be an important determinant for how severely monetary policy is constrained in the face of certain shocks.

4.2 Fiscal retrenchment during deep recessions

What are the effects of government spending cuts in the recessionary environment just described? In order to mimic the setup of Section 3, we consider an episode of fiscal restraint that lasts for two years and starts at the onset of the recession, that is, we consider an “immediate retrenchment” scenario. We discuss alternative timing assumptions in Subsection 4.3. During the episode of fiscal restraint, government spending falls below its steady-state level by 2 percent of (steady-state) GDP. Figure 6 depicts the resulting dynamics relative to the baseline scenario. The panels on the left show the response of output (top) and the interest rate spread (bottom). In order to isolate the effect of a binding ZLB, the panels in the middle and on the right show the response of the same variables under the assumption that the exit from the ZLB is not endogenous, but fixed exogenously at 7 and 18 quarters. These lengths correspond, respectively, to the duration of the ZLB episode for debt-to-GDP ratios of 60 percent and 115 percent in Figure 5.24

Figure 6:
Figure 6:

The impact—relative to the baseline—of immediate retrenchment

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The panels show the effects of an immediate retrenchment in government spending by 2 percent of steady-state output for 8 quarters on output (top) and the risk premium (bottom). Solid black line: 60% initial debt-to-GDP ratio, dashed blue line 90%, dotted red line: 115%. Left panel: the effect of the retrenchment when the timing of the exit from the ZLB is endogenously determined according to equation (36). Other panels: for each initial debt level, the depth of the recession without retrenchment is calibrated such that it implies a binding ZLB for 7 quarters (middle) and 18 quarters (right). For the middle and rightmost panels, regardless of the austerity package being implemented, monetary policy is assumed to keep the nominal rate at zero for that time period (7 and 18 quarters, respectively).

We find, first, that for all initial debt levels, the retrenchment package is effective in reducing the deficit (not shown). Thus, the level of government debt and the risk premium (consistent with Figure 4) decline. In addition, monetary policy escapes from the ZLB constraint somewhat earlier for all debt levels (not shown). However, the spending cuts have quite different effects on output across our different scenarios. If initial debt is low, the spending cuts prompt an initial output decline of more than 2 percent, as private expenditure also falls. In contrast, if initial debt is high, the initial output response is actually positive, reflecting a strong increase in private expenditure. To understand this finding, recall that spending cuts affect real interest rates through two channels. On the one hand, the deflationary effect of lower public spending raises, all else equal, real interest rates as it cannot be offset by a reduction in policy rates at the ZLB. On the other hand, the reduction of public debt reduces sovereign risk, thereby lowering private interest rate spreads. The strength of the effect operating through the second channel increases in the initial level of debt, as the postulated relationship between public debt and sovereign risk is nonlinear. In our simulations, this sovereign risk channel turns out to be dominant if we assume an initial debt level of 115 percent of GDP.

As stressed above, the impact of the recessionary shock and the period of time for which monetary policy is constrained depend on the initial level of debt. In addition, the initial level of debt also determines the quantitative importance of the sovereign risk channel. To isolate the latter dimension, the panels in the middle and right columns of Figure 6 show the effect of the austerity packages on output for different debt levels, but fixing the length of the ZLB episode exogenously. If the lower bound is expected to be binding only for a short period of time (middle panels), the effect of the spending cuts hardly varies with the debt level. In other words, the importance of sovereign risk appears limited. However, in line with our earlier results in Section 3, the picture changes quite dramatically if the lower bound is expected to be binding for an extended period of time (right panels). In this case, we find that spending cuts have a positive and lasting effect on economic activity at high initial levels of debt, which under our specification imply elevated sovereign risk premia (dotted red line, right panel). Moreover, the “output costs” of public spending cuts are noticeably lower than usual even for the intermediate, 90 percent, level of debt to GDP (dashed blue line in the rightmost panel). Overall, we thus find our earlier analytical results confirmed by the simulations of the full-fledged model: fiscal strain alters the fiscal transmission mechanism through the sovereign risk channel, provided that monetary policy is constrained for an extended period.

4.3 Further considerations

This section assesses the sensitivity of the previous results with respect to some modifications that may have a bearing on the sovereign risk channel. We start by assessing how an alternative timing of spending cuts affects the impact of fiscal retrenchment. We then explore alternative assumptions on the tax system and on the conduct of monetary policy.

4.3.1 The timing of spending cuts

So far, we have focused on spending cuts that take place immediately, i.e., at the time the recessionary shock hits the economy. This scenario reflects an important premise guiding the analysis in this paper: namely that policymakers facing nonnegligible sovereign risk premia will generally find it difficult to improve investor confidence solely by announcing future austerity. More likely, investors concerned about default risk will expect to see some immediate proof of policymakers’ readiness to cut deficits, thus prompting upfront tightening measures. Nevertheless, it is worth contrasting the effect of such immediate tightening programs with examples of more backloaded deficit reduction. We now assess the role of such timing decisions on the basis of model simulations. The left panels of Figure 7 reproduce the results for an immediate retrenchment that were reported in the left column of Figure 6. The remaining columns report the impact of two alternatively timed (and sized) consolidation packages. The middle panels display the response of the economy to a package of spending cuts of 2 percent of steady-state GDP that starts two years after the initial recessionary impact and that lasts for 10 years (a “medium-term retrenchment”). The right panels, in turn, show results for a combination of the previous two packages (a “persistent retrenchment,” that sees spending cuts start immediately and last for 12 years in total). As before, the figures report the effect of these different packages depending on the initial level of debt (the range of the axes has been rescaled to accommodate the range implied by these simulations).

Figure 7:
Figure 7:

The timing of retrenchment

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The panels show the effect of a retrenchment in government spending by 2 percent of steady-state output on output (top) and the risk premium (bottom) for different assumptions on the timing of the austerity episode. Left panels: immediate retrenchment for first 2 years; middle: medium-term retrenchment (during years 3 through 12), right: persistent retrenchment for 12 years. Solid black line: 60% initial debt-to-GDP ratio, dashed blue line 90%, dotted red line: 115%. For all the panels the timing of the exit from the ZLB is endogenously determined according to equation (36).

Under medium-term retrenchment (middle panels), there are immediate output gains from fiscal forward guidance, in line with the results in Corsetti et al. (2010). Specifically, most or all of the spending cuts are implemented when monetary policy has escaped from the ZLB constraint and regained its ability to contribute to economic stabilization. When the spending cuts occur, they reduce contemporaneous demand and inflation, causing the central bank to lower the real interest rate. The prospect of a lower future real rate already crowds in consumption, and thus output, during the early periods of the recession, i.e., before the government spending cuts occur. This leads to higher tax revenue and implies an immediate reduction in the sovereign risk premium, which further stimulates demand through reduced interest rate spreads facing the private sector. Note that such anticipated medium-term retrenchment unfolds a stronger expansionary effect upfront, the weaker the initial fiscal situation (dotted red line vs dashed blue line in the central panels).

The persistent retrenchment scenario, in turn, is a combination of the two previous scenarios. In the simulations shown in the right column of Figure 7, persistent retrenchment has the strongest effect on the risk premium (red dotted line), reflecting the larger overall cut to spending. However, its output effects are generally somewhat smaller than in the medium-term retrenchment scenario. For the highest level of debt under consideration (i.e., 115 percent of GDP), the persistent retrenchment still crowds in economic activity for most of the time that the economy remains constrained by the ZLB. However, the initial boost to output is more subdued than under medium-term retrenchment, because the impact from the additional reduction in debt—and hence the risk premium—is more than offset by the immediate contractionary effect of upfront spending cuts.

Together, our simulations using alternative timing assumptions underscore the theoretical benefit of delaying fiscal adjustment until after the economy has recovered from the worst of the initial recession. However, the simulations strictly depend on the credibility of pre-announced consolidation plans. If private agents harbor any doubts about the likelihood of implementation, they will expect a less benign path for deficits and debt than has been announced, and the sovereign risk premium will be correspondingly higher. In this sense, the medium-term retrenchment scenario considered here may not actually be available as a credible policy option, notably at times of significant concern about sovereign risk. This notion is an important motivation for the present paper and explains the focus of most of our analysis on fiscal adjustment scenarios that involve immediate spending cuts. That being said, it bears repeating a key result of our earlier work in Corsetti et al. (2010): to the extent that policymakers can credibly commit to future austerity, backloaded consolidation strategies will generally garner superior outcomes in terms of near-term growth, as they avoid tightening fiscal policy at a time when the central bank is unable to offset the contractionary fiscal impulse.

4.3.2 Distortionary taxation

Distortionary taxes generally weigh on economic activity. To the extent that an early retrenchment reduces the need for distortionary taxation in the future, such a policy would appear to be less harmful to growth than we have reported so far. Figure 8 assesses this hypothesis, focusing again on results for three different levels of initial debt.25

Figure 8:
Figure 8:

Effects of retrenchment on output depending on taxation

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The panels show the effect of retrenchment on output for different assumptions about taxation. The left column reproduces the effect of retrenchment with lump-sum taxes (as in the baseline). The middle panel shows the effect of retrenchment when labor taxes are used throughout. The right column shows the effect of retrenchment if taxes are lump-sum for the first 18 quarters but distortionary thereafter. For the simulations in which labor taxes are used the initial shock has been rescaled such that for each debt level the severity of the recession and the length of the ZLB episode without retrenchment are roughly comparable to the baseline responses shown in Figure 5.

The first column reproduces the responses for the baseline case, in which only lump-sum taxes are used. The second column shows how distortionary taxation alters the effect of the spending cuts. In the simulations, distortionary taxes, if used throughout, have little bearing on how fiscal retrenchment affects output. While lower future distortionary taxation stimulates future economic activity, and thus current demand, a second, countervailing effect is present. Without retrenchment, distortionary tax rates rise more strongly, with a view toward stabilizing the rising debt burden. As stressed by Eggertsson (2011), to the extent that higher labor tax rates raise inflation, they may help to stabilize output if the economy is stuck at the ZLB. By contrast, an early retrenchment reduces the labor tax burden and exerts a negative effect on activity. In the simulations shown here, this effect outweighs the beneficial impact arising from the reduced tax burden in later periods.

One may argue, of course, about the likelihood that taxes would actually be raised while the economy remains stuck at the ZLB. Since the timing of taxation is important for the aforementioned finding, the rightmost column of Figure 8 shows the results for a different taxation scenario. The goal is to disentangle the impact of fiscal retrenchment from the specific effects that labor taxes have in a ZLB situation. Thus, the responses in the right column are based on the assumption that lump-sum taxes are used for the first 18 quarters to prevent debt from exploding. Only after these 18 quarters, and hence well after the end of the ZLB episode, do labor taxes take over. For the 60 percent debt level, the difference relative to the case of lump-sum taxation is small, and the effects are very similar to the baseline. For the 90 percent debt level, the output effects of retrenchment turn more favorable. However, this mostly affects the medium-term retrenchment package. Finally, results for the highest debt level shown here, i.e., 115 percent of debt to GDP, illustrate that the sovereign risk channel may be stronger in the case with distortionary taxation, provided that changes in distortionary tax rates materialize only after the ZLB has ceased to bind. Intuitively, for this level of sovereign indebtedness, the heightened risk premium demands relatively strong increases in distortionary taxes to finance the debt burden, which by itself worsens the recession. As a result, an early retrenchment now crowds in output, and much more strongly so than in the case with lump-sum taxes only.

4.3.3 Degree of monetary policy offset for risk premia

Next, we consider the sensitivity of our results with regard to the extent to which the central bank is able to neutralize the effect of sovereign risk on private borrowing rates. Our baseline simulations have assumed that once the economy has exited from the ZLB situation, the central bank can set interest rates in a way that prevents sovereign risk from affecting economic activity in any way.

Figure 9:
Figure 9:

Effects of retrenchment on output when CB cannot fully neutralize risk premia

Citation: IMF Working Papers 2012, 033; 10.5089/9781463933180.001.A001

Notes: The figure shows the effect of the three retrenchment packages (described in the notes to Figure 7) for the case where the central bank does not fully offset the risk premium even once the economy has left the ZLB. Top row shows the baseline responses for 60, 90, and 100 percent initial debt-to-GDP ratio. Bottom row shows the case when the response coefficient ø ω is only ¾ of the size of the benchmark calibration.

We now assess, instead, the effect of spending cuts if the central bank does not, or cannot, fully neutralize the sovereign spread. In particular, we assume that the central bank’s response is only three-quarters the size of the value of ø ω that we used in the previous simulations. To this end, we set ø ω = 0.75(πb + sΠ). The first point to note about this scenario is that the baseline recession is deeper, reflecting the stronger headwinds from persistently high risk premia. The lower panels of Figure 9 show the effects of fiscal retrenchment relative to that new baseline. By comparison, the upper panel revisits the baseline in which the central bank fully offsets higher sovereign risk premia after the ZLB episode. As it turns out, early retrenchment is less detrimental to economic activity at high initial debt levels in the lower panels. Intuitively, the central bank’s limited willingness (or capacity) to neutralize higher spreads in the future raises the benefit of fiscal retrenchment as a way to bring down long-term real interest rates via the sovereign risk channel.

5 Conclusion

The present paper analyzes how the “sovereign risk channel” affects macroeconomic dynamics and stabilization policy. Through this channel, rising sovereign risk drives up private-sector borrowing costs, unless higher risk premia are offset by looser monetary policy. If the central bank is constrained in counteracting higher risk premia, sovereign risk becomes a critical determinant of macroeconomic outcomes. Its implications for stabilization policy have not been fully appreciated in earlier formal analyses, although they are likely to be of great importance for many advanced economies currently facing intense fiscal strain.

Building on the model proposed by Cúrdia and Woodford (2009), we show that the sovereign risk channel makes the economy (more) vulnerable to problems of indeterminacy. In particular, private-sector beliefs about a weakening economy can become self-fulfilling, driving up risk premia and choking off demand. In this environment, a procyclical fiscal stance—that is, tighter fiscal policy during economic downturns–can help to ensure determinacy.

Further, we find that sovereign risk tends to exacerbate the effects of negative cyclical shocks: recessionary episodes will be deeper the stronger the sovereign risk channel, which in our specification is a nonlinear function of public-sector indebtedness. Moreover, in deep recessions that force the central bank down to the zero lower bound (ZLB) for nominal interest rates, sovereign risk delays the exit from the ZLB, hence prolonging macroeconomic distress. The sovereign risk channel also has a significant bearing on fiscal multipliers. Specifically, the effect of government spending on aggregate output hinges on (i) the responsiveness of private-sector risk premia to indicators of fiscal strain; and (ii) the length of time during which monetary policy is expected to be constrained. Our analysis suggests that upfront fiscal retrenchment is less detrimental to economic activity (i.e., multipliers are smaller) in the presence of significant sovereign risk, as lower public deficits improve private-sector financing conditions. In relatively extreme cases where fiscal strains are severe and monetary policy is constrained for an extended period, fiscal tightening may even exert an expansionary effect. That being said, fiscal retrenchment is no miracle cure. Indeed, all our simulations feature a deep recession even if tighter fiscal policy, under the aforementioned conditions, may stimulate economic activity relative to an even bleaker baseline.

As an additional caveat, we note that our analysis has focused on fiscal multipliers under a go-it-alone policy that does not involve external financial support at below-market rates. Availability of such support could allow countries to stretch out the necessary fiscal adjustment as they benefit from lower funding costs and, possibly, positive credibility effects. Indeed, if and where announcements of future fiscal adjustment are credible, delaying some of the planned spending cuts remains a superior strategy in terms of protecting short-term growth. How countries end up dealing with the challenges summarized here may prove to be a defining feature of global economic developments over the coming years.

Sovereign Risk, Fiscal Policy, and Macroeconomic Stability
Author: Mr. Keith Kuester, Mr. Gernot J. Mueller, Giancarlo Corsetti, and Mr. Andre Meier