Government Debt After The Crisis

Transcription

1 Government Debt After The Crisis Rome, July 24, 2009

2 I Motivation I I I I I Conclusion

3

4 I Automatic stabilisers and discretionary policy led to huge increase in scal de cits I Equity stakes in banking sector may lead to further increases I Below trend growth expected for several years I Trend growth may have been adversely a ected I Elasticity of revenue wrt GDP growth may have fallen I Ageing population

5 Red ink for as far as the eye can see

6 THREE FISCAL FEARS 1. Government debt rising unsustainably and scal policy needing to show imminent correction 2. Large issuance of debt leading to higher interest rates, crowding out and unstable debt dynamics 3. Governments responding to large debt by engineering high in ation

7 TALK OF EXITS 1. What is optimal level of debt? 2. How quickly to restrain debt dynamics? 3. What type of debt to issue?

8 Don t look at issue of whether de cits stimulate the economy - only whether level of debt should constrain de cits

9 b t (1 + r t g t ) b t 1 + g t y t y t 1 y t τ t y t So for debt to be stable need to run primary surplus t t y t g t y t = (r t g t ) b t 1 y t 1

10 r g Primary Balance 2010 Gross Debt Canada France Germany Italy Japan United Kingdo United States

11 Seems likely r g has increased since 2007 I g has fallen I r increases due to banking crisis and capital scarcity I r increases due to increase in risk premium I r increases due to concerns over devaluations and default I r falls due to increase in private sector savings If r g has increased then two parts to adjustment i) returning primary surplus to level that keeps debt sustainable given 2007 values of r g ii) increasing primary surplus to achieve sustainability given changed values of r g

12 I Only for Germany, Italy and Japan is 2010 projected primary balance featured in matrix I If r g > 1% then Germany and Japan need tightening I UK and US largest need for adjustment I For most countries scale of adjustment is within what has historically been achieved (Canada, Italy, UK and US 5-7%) I Changes in r g secondary to need to change policy I Given scale of adjustment debt is likely to rise substantially beyond levels projected in 2010 I If both r and g tend to zero (as in Japan) then level of debt becomes less important as constraint on policy

13 How high should debt go? How long should adjustment take? Unlikely countries can swiftly adjust their de cits. Debt liable to show substantial and persistent increases How high should debt go? Should governments try to return debt to pre crisis levels? How long should adjustment take?

14 How high should debt go? How long should adjustment take? Theory says little on this matter. This is why scal rules always so di cult Key restriction is debt equals net present value of all future primary surpluses Only real restraint has been imposed by capital markets and for G7 this has been a relaxed constraint

15 How high should debt go? How long should adjustment take? UK Government Debt US Government Debt

16 How high should debt go? How long should adjustment take? I Is war time data relevant? A political economy argument about sharing costs of scal burden I Note that between 1929 and 1934 UK debt rose from 160% to 205% GDP, between 1970 and 1998 US debt doubled I Historically two facts - level of debt seen in past exceeds that currently projected for G7 - reductions in debt take several decades

17 How high should debt go? How long should adjustment take? Intertemporal budget constraint says debt equals net present value future surpluses b t 1 (s t ) = PVD(s t ) Consider case of large and persistent (but not permanent) increase in government expenditure Under tax smoothing government raises taxes permanently but runs de cits in near term which are matched by higher future surpluses once expenditure increase diminishes and higher taxes remain Result is that debt initially increases and then declines Debt acts as a bu er and shows long term swings in response to scal shocks

18 How high should debt go? How long should adjustment take? Can be more precise I Debt is not mean reverting but contains a large unit root I Debt shows much greater persistence than any other variable including tax revenue and government expenditure I Debt will appear to be unsustainable along an optimal path - "explosive dynamics" I If g and t a ected for several years then debt is a ected for several decades

19 Previous section showed historically and theoretically that debt displays huge long term swings in the face of adverse expenditure shocks May appear explosive dynamics but eventually sustainability achieved How is sustainability achieved? Primary surpluses? In ation? Look at period 1960 to 2005 for G7 (less France)

20 I Period shows large uctuations in debt broadly in range expected for G7 countries in next decade I Essentially a period of peace - debt increases not due to military factors I Contains period of debt increasing and decreasing

21 Based around a period by period budget constraint G t T t = B t Υ t 1 Π t Q t B t 1 + H t 1 Π t Q t H t 1, Gives an intertemporal constraint α b b t 1 + α h h t 1 + α g g t + α τ τ t ' α 0 E t δ j 0 d t+j + α 1 E t δ j 0 h t+j j=1 j=0 T 1 +α 2 E t δ j 0 rt+j 1 + δ 1 π t+j + (1 + δ 1 ) γ t+j. j=0 Imbalances between stock of debt, monetary base and g and t (LHS) are corrected by future changes in primary de cits, monetary base, interest rates, in ation or growth (RHS)

22 I Around 85% or more of imbalances is corrected by changes in future primary surpluses I In ation provides an at best marginal contribution I Fiscal imbalances have no role to play in forecasting in ation at any future horizons I Italy a good example - saw its primary surplus shift from -8% to 6%

23 Governments will be issuing lots of debt, funding costs will rise and yields may shift adversely "Optimal " debt management is issue long and hold short term assets Not optimal if risk of default, transaction costs paramount Expect sell lot of longer indexed bonds

24 I Serious issues over debt and de cits - major policy concern I Risks overstated I Debt should be used as bu er and show long term swings that last decades and look unsustainable - not mean reverting I During times of war and peace seen huge uctuations in debt and adjustments of similar magnitudes I Post 1960 adjustment from primary surpluses, very little from in ation I Suggests if government wants can keep in ation low (Political consensus?)

25 I What about ageing? I Wasn t taken seriously in 2007 so why now? I Crisis e ect "small" relative to ageing so why stress about it? I Markets are key. But implicitly must have been prepared to nance huge age related de cits so why not crisis? I G7 can issue 10,20, 30 and 50 year debt and then rollover. So long as that remains expect persistently very high debt

26