Pankinjohtaja Sinikka Salon puhe Maltalla
Puheet ja haastattelut Puhe perjantai 28.3.2003, 0.00
Remarks in the SUERF (The European Money and Finance Forum)
Seminar "Securing Financial Stability: Problems and Prospects for new EU Members", jointly organized with the Central Bank of Malta on 28 March, 2003
First of all let me thank the organisers for this opportunity to talk in front of this distinguished audience. I have attended SUEF meeting more or less regularly since early 1980s and it is a great pleasure to be here again. Large changes are about to happen in the economic scene in Europe over this period and this is very much reflected both by the place and the participants of this seminar. I very much welcome these changes. I understand that I am supposed to tackle the issue of fiscal soundness and monetary stability from the point of view of a practitioner rather than providing any academic insight in the issue. This is no doubt my approach. However, when preparing this presentation I browsed some recent academic literature on the subject and I could not resist the temptation to provide some very "subjective" reflections of how to apply them in practice. Anyhow my approach in the paper that I am going to summarise to you is very "practical" relating to the assessment of fiscal stability in the accession countries.
Relations between fiscal and monetary policies have been and are likely to remain the subject of many discussions in both academics and policy-making circles. Of these discussions, the best known is perhaps the so-called ”policy mix” debate. This is concerned mainly with the design of fiscal and monetary policy responses to different macroeconomic disturbances, or more recently, design of different policy rules to be applied to different probability distributions of disturbances. The relation between fiscal soundness and monetary stability is a closely related but nevertheless distinct issue. It is concerned with disturbances that are of fundamental or extreme nature and it always has a bearing on the possibility of fundamental policy failures.
Of the different aspects of fiscal soundness and monetary stability, this presentation will focus on the sustainability of public finances and the risk of high inflation. High inflation is - at least in its extreme form of hyperinflation - largely a modern phenomenon, related to the need to print paper money to finance large fiscal deficits. These deficits are often but not always caused by exceptional events such as wars, revolutions, collapses of empires, and establishment of new states.
It would be interesting – and in the present economic situation, some would say more timely - to discuss also the other side of the coin, namely the relationship between fiscal stability and deflation, but I must leave this issue out of this presentation. For me, it is a side issue and the main theme is complicated enough even without it.
My presentation is structured the following way. I first discuss two rather different academic approaches that consider the relationship between fiscal soundness and monetary stability. No in-depth survey of the literature will be attempted here; my purpose rather is just to give some glimpses of research that has been done on this subject. (References to some recent literature are provided on my main paper.) Thereafter I summarise in general terms some of the main steps usually adopted in practical assessments of the sustainability of public sector indebtedness.
Finally, in line with the general theme of this seminar, I discuss some of the fiscal challenges the accession countries may encounter in the coming years. Although the level of government indebtedness in most accession countries is now rather low and in any case lower than in most of the present member states, fiscal deficit are high and, in view of the forthcoming pressures, may rise further. The medium-term picture foreseen by the countries themselves in their Pre-Accession Economic Programmes is very favourable, but I see a risk of increasing deficits and of a continuous increase in the level of governments’ indebtedness. Such fiscal policies would place a heavy burden on national monetary policies in the new member states. Furthermore, fiscal pressures in the new member states may intensify political pressures to weaken the rules of fiscal surveillance in the EU. There is thus a risk that the enlargement will both directly and indirectly increase the burden fiscal policies places on monetary policies.
In the "theoretical part" of my paper I summarise two strands of academic discussion relating to relation between fiscal policy and inflation, notably the so-called inflation tax literature and the fiscal theory of the price level. The latter literature distinguishes between monetary dominant (or Ricardian) and fiscally-dominant (or non-Ricardian) equilibria or regimes. In the monetary dominant regime fiscal primary surplus adjusts, given any sequence of prices, to guarantee fiscal solvency, whereas in the fiscally dominant regime the government's inter-temporal budget constraint is satisfied only for some price paths and the price level is assumed to settle itself to a path satisfying the government budget constraint. Interestingly, it has been suggested in a paper published by the EU Commission Services that the EMU is a monetary dominant regime, provided that the Stability and Growth Pact is respected.
In the so-called inflation-tax literature, connection between fiscal policy and inflation is studied in a simple demand-for-and-supply-of-money framework. Inflation is treated as a tax on money balances and fiscal policy is connected with inflation through seigniorage revenues obtained from increasing money supply. The cost of increased money supply is borne by holders of money balances as the purchasing power of money decreases.
A substantial body of empirical evidence confirms that inflation is highly correlated with money growth, the long-run correlation between these two variables being close to one-to-one. In high inflation countries, the connection between money growth and inflation is broadly contemporaneous. In low-inflation countries, the effects of money growth are distributed rather evenly across the current and previous periods, and the short run correlation between contemporaneous and even lagged money growth and inflation appears to be in general rather low.
As to the relationship between public finances and money supply, a plausible and often advanced hypothesis is that money supply and seigniorage income should increase as a function of the fiscal needs, as reflected by the size of the budget deficit. However, empirical evidence of this link is weak. In their recent empirical analysis Fisher et. al. found a negative relationship between budget balance and seigniorage in their cross-sectional data, but the coefficient was small. A ten-percentage-point increase in deficit leads on average to a 1,5 percent increase in segniorage revenue. Furthermore, even this weak relation seems to derive mainly from some exceptional observations with very high levels of seigniorage (over 6 per cent of GDP). These exceptional observations included, by the way, one accession country, namely Malta. For the lower-inflation countries the coefficient is even lower and insignificant.
The basic message of the inflation tax literature for the analysis of fiscal soundness and monetary stability is that in some circumstances, monetary financing may be the only source of finance available for the government and that monetary financing is bound to affect the functioning of the monetary system. These are still valid messages.
There is also some research on inflation tax as an element in an optimal or efficient system of taxation. I am not convinced of the usefulness of this type of analysis and, to save time, I will not deal with this issue in my presentation – I have discussed it briefly in my main paper.
To summarise: There seems to be a widespread conviction that high inflation is almost always caused by fiscal pressures. Against that background, the lack of a firm ex post correlation between deficit and seigniorage may be surprising. However, among other harmful things, high inflation disrupts the budgeting of expenditures and wrecks tax collection systems. Thus, government receipts and disbursements are likely to vary erratically, which may explain the lack of a correlation between fiscal deficits and inflation.
Fiscal Theory of the Price Level
Another, recent strand of academic discussion, called the fiscal theory of the price level, deals explicitly with fiscal and monetary policy and their interactions. The theory focuses on the government’s inter-temporal budget constraint and the sustainability of the fiscal position. There are many ways to state the government’s inter-temporal budget constraint. One useful way to write it is in terms of the change of the (net) debt ratio, or
(1) db/dt = -s + (i – g - p) b + e.
Here b is the government’s net debt and s primary surplus, both as per cent of GDP. db/dt denotes the change in the debt ratio whereas i is the (average) interest rate on debt, g the growth rate of real GDP and p the rate of inflation. The stock-flow adjustment term e captures, as per cent of GDP, all other effects on the change in the debt ratio such as reevaluations of debt items etc.
The fiscal theory of the price level assumes that agents optimise their behaviour over time in anticipation of future economic developments, including the expectation of future policy actions. There is also a shift of emphasis towards the analysis of budgetary and monetary policy rules, rather than single decisions, and their time-consistency.
The inter-temporal and optimising character of analysis makes models complicated and difficult to analyse, even given their strong simplifying assumptions. Nevertheless, the theory has made many interesting contributions to the analysis of fiscal soundness and monetary stability, including the analysis of the institutional framework needed to govern the interactions between monetary and fiscal policy – I mentioned earlier as an example a justification for the Stability and Growth Pact. I will not try to provide even a selective survey of the results of the analysis here; rather, my intention is to draw some more or less subjective lessons from the literature.
Compared to the inflation tax literature, the fiscal theory of the price level extends both the time perspective and the spectrum of liabilities and assets considered in the analysis. Instead of a single instantaneous rate of inflation, the whole time path of the price level is relevant, and instead of the stock of money, the whole portfolio of government liabilities and assets is considered at least in principle.
Results from the theory illustrate the fact that irrespective of the starting position, the burden of irresponsible fiscal policies, if followed persistently over time, becomes excessive in the end even for strongly stability oriented monetary policies. The risk of vicious debt dynamics is rather explicit in equation (1). If primary budget deficit is large (as reflected in a large negative value of s), debt tends to increase. When debt becomes large enough, it starts to grow at an accelerating rate, as nominal interest rate normally exceeds nominal output growth. At this stage, there is not much to be done to increase the real rate of growth (g) and there are also limits to feasible primary surpluses (s). For very high debt ratios, unsustainable debt dynamics can only be reversed if the rate of interest on debt (i) is low enough and the rate of inflation (p) is high enough, i.e. through monetary financing. In rational expectation models, this is of course clearly perceived by the public well before its actual occurrence.
Furthermore, models considered in the literature clearly illustrate the fact that monetary and fiscal policies need to be consistent in order to be able to produce balanced and harmonious policy outcomes. In fact, there is no guarantee in these models that a uniquely determined, well-behaving equilibrium time path for the price level exists, even for policy rules which in themselves seem to be quite sensible but not mutually consistent. For example, inflationary and deflationary spirals, stochastically fluctuating explosive inflation or sunspot equilibria may emerge even if monetary authorities let money stock grow at a constant rate or follow a Taylor rule.
Even though this literature has provided many useful insights, I have some doubts about the extent to which its results are relevant from an empirical or political point of view. By their very nature, analytical models are highly simplified and therefore unrealistic representations of complex reality. At least for the moment, the contributions from this literature have been more conceptual in nature, and the models involved appear to have left little marks on the way debt sustainability or interactions between fiscal and monetary policies are analysed in practice.
Having had a look at the recent strand of theories about the relation between fiscal soundness and monetary stability let me now turn to assessment of government debt sustainability.
Debt sustainability and some reflections on “fiscal dominance”
The concept of fiscal dominance is related to the concept of government debt sustainability. In practical work on debt sustainability by institutions like the IMF, the government’s intertemporal budget equation like (1) above is also the cornerstone of the analysis, but rules of thumb and back-of-the-envelope computations replace the solution of forward-looking rational expectations differential equations. In all likelihood, one is forced to rely on judgement. This is in part related to the fact that the authorities seldom have any well-defined “policy rules” assumed in the theory or, if they have, follow them with the consistency assumed in academic models. While the role of the monetary policy in the sustainability analysis is crucial, in practical work it is usually specified only indirectly through macroeconomic assumptions about inflation, interest rates, exchange rates etc.
The imprecise character of policy rules makes debt sustainability a rather vague concept. In practical terms, a government’s liability position can be regarded as sustainable if, given the cost of financing the government faces in the debt market, the government indebtedness does not grow continuously without major correction in the balance of income and expenditure. The exclusion of major corrections in income or expenditure captures the notion that there are social and political limits to adjustment that determine the willingness to pay.
The first step in a sustainability assessment is to determine the starting position. In theoretical models, this step is trivial but in actual assessments it may be difficult. The compilation and dissemination of basic data such as level of public sector indebtedness, primary balances, interest bills, the real interest rates, etc are complicated by difficulties such as outright lack of data and imprecisely defined off-budget and contingent liabilities.
A major step in the exercise is to project key macroeconomic variables, such as interest rates, rates of economic growth, and exchange rate changes, together with the flows of revenues and expenditures over sufficiently long period. To the extent that these variables are influenced by government policies, this part of the assessment is intrinsically uncertain.
An almost overwhelmingly difficult decision is to determine the point at which the government can no longer be expected to be able to continue servicing its debt without unrealistic corrections in income or expenditure. No single threshold such as a given level of indebtedness can reliably define this point. For some developing countries, a government debt level of 40 per cent of DGP may be unsustainable whereas some other countries have successfully managed a public sector debt in excess of 100 per cent of GDP.
Ultimately, assessments of sustainability can be only probabilistic. An increase in public sector debt is likely to increase the perceived risk of unsustainable public finances and thereby the perceived risk of a future loosening of the monetary policy and ultimately of monetary financing. The size of the perceived risk depends on many things such as the level of government debt, the soundness of the fiscal policy framework as well as the status of and the credibility of the monetary policy strategy adopted by the central bank. An increase in indebtedness gives rise to less concern if it is associated with strong and credible overall policy framework and with strong commitments by the fiscal authorities to restore the debt level to a low level within a realistic timetable.
The composition of debt as well as related “stock-flow” corrections in the level of indebtedness symbolised by e in equation (1) are also relevant. For example, debt denominated in domestic currency may bring with itself a temptation to inflate in order to reduce the real value of the debt. As the risk is recognised by the public, inflationary pressures in the economy may emerge as domestic debt increases even if the government has no intention to take recourse to monetary financing.
Comprehensive medium-to-long-term fiscal programmes can be a key element in both in assessing and in shaping policies to avoid the risk of fiscal unsustainability. For that purpose, the stability and convergence programmes drafted by the EU member states now include a section analysing the sustainability of public finances from a longer run point of view. This work is based on and complemented by in-depth analyses by the Economic Policy Committee on the effects of ageing and the government’s contingent liabilities.
While official fiscal programmes may be useful, their usefulness depends entirely on the assumptions used in programme’s projections. The experience from the convergence and stability programmes suggests that official programmes tend to suffer from optimistic assumptions. Moreover, usually only marginal deviations from the baseline forecast are considered in them and thus they do not provide proper stress tests for analysing the consequences of substantial risks to fiscal soundness.
Fiscal challenges in the accession counties
In view of the main theme of this seminar, I will now venture to reflect on fiscal challenges in the accession countries, keeping an eye on the soundness of public finances. This is a complex and sensitive topic. My intention here is not to try to present an exhaustive discussion of it –that would clearly be beyond my competence – but rather only to restate some of the points I have made earlier on fiscal soundness and monetary stability in somewhat more concrete terms, using accession countries as illustrative material.
As to the starting point, the public sector indebtedness in the 10 accession countries is not alarming, given the relatively high level of economic development achieved by these countries. In 2001, the ratio of government debt to GDP ratio was higher than 60% of GDP only in Malta.
On the other hand, the balance of income and expenditure is less favourable and many accession countries have large budgetary deficits. In 2001, five accession countries, including the three largest countries (Poland, the Czech Republic and Hungary) ran deficits above 3% of GDP. I have no data for 2002, but the deficit estimates presented in the last year’s Pre-Accession Economic Programmes range from near zero in Estonia to between 5% and 7% of GDP in the Czech Republic, Hungary, Malta and Poland. Given the low or moderate level of indebtedness, high deficits indicate, in general, large primary deficits.
In view of the large primary deficits there may be a risk that at least some of the accession countries are in the early phase of a perhaps rapid debt accumulation process. In fact, the debt ratio has already started in some countries to increase very rapidly. The risk of prolonged debt accumulation stems, among other things, from following fiscal challenges now facing the accession countries:
Accession countries are undergoing tremendous structural and institutional changes, including the need to upgrade public infrastructure and to fulfil the commitments entailed in the European Agreement. These will have in future significant budgetary implications.
While accession countries will benefit in net terms from the EU accession, the impact on budgets is thought to be negative. Investment in public infrastructure, complying with EU environmental standards and the implementation of also rest of the EU acquis implied by the accession will be only partly compensated by net transfers and positive budgetary effects of tax harmonisation. In particular, absorbing structural and cohesion funds requires domestic fiscal co-financing.
The financial sector in most accession countries is still at an early stage of development and therefore governments must rely heavily on foreign savings to finance public deficits. This exposes them to balance of payments difficulties. Lack of progress in reining in excessive deficits could undercut market sentiment and economic activity in countries with large external deficits.
Sharp stock-flow corrections may increase public debt. These corrections may include the impact of exchange rate changes on foreign currency-denominated public debt and the public assumption of private debt (for instance as a result of privatisation deals, bailouts, calls of public guarantees etc.). The share of foreign currency denominated debt out of total public debt is significant in many accession countries already now and may increase in the future. In some cases, government guarantees may also be a matter of concern.
The ageing of the population will affect profoundly economic developments in all EU countries, including the future members. It remains to be seen whether its effects are comparable to events like wars or end of empires, which in the past have been the major driving force behind fiscal and monetary instability. Anyway, while many accession countries have implemented comprehensive reforms of their pension systems, I expect that the ageing of the population poses considerable challenges for most if not all of them, especially in view of the consequences of the expected increase in labour mobility after the accession.
A scenario of growing or perhaps even accelerating public sector indebtedness seems to be in contrast with the scenarios presented in the accession countries’ Pre-Accession Economic Programmes. Last year, the programmes forecast progressive fiscal consolidation at general government level, with budgetary positions in 2005 ranging from balanced budgets (Bulgaria, Estonia) to a deficit of 3.1 % and 5.5 % in Malta and Czech Republic, respectively. According to the programmes, the debt ratio would have risen significantly only in the Czech Republic and, to a lesser extent, Latvia and Poland.
While the scenarios in Pre-Accession Programmes certainly were not infeasible at least at the time of their drafting, they fail to be fully convincing. To some extent this is due to a lack of a throughout long-term sustainability analysis in the programmes, but a more important reason is the assumptions used in the programmes. The Pre-Accession Programmes assume high and even accelerating growth rates so that in all programmes the 2005 growth rate is the highest one projected. While the accession countries generally managed to post growth rates of 2-5 per cent in 2002, this was only because robust domestic demand helped to offset the impact of slowing exports. Large current account deficits underscore the risks related to these trends. In order to permanently achieve growth rates assumed for the later years, accession countries should achieve a significant acceleration in the pace of the catching-up process. Only by taking early action to dispel even the slightest doubts about fiscal developments which may over time lead to an erosion of debt sustainability one can be assured that macroeconomic environment is favourable for the achievement of this key priority.
Although the situations may differ, there is perhaps a general need to strengthen mediumterm budgetary frameworks in the accession countries. There are several elements common to all strong medium-term fiscal frameworks. First of all, programmes should be based on realistic or preferably even cautious macroeconomic assumptions. Second, a careful review of government spending priorities and tax cut commitments is needed to secure a momentum of fiscal consolidation and the credibility of medium-term budget targets. In this respect, I note that most Pre-Accession Programmes fail to provide policy commitments to credibly underpin a medium-term path of fiscal consolidation. Nevertheless many of them envisage significant reductions of primary expenditures, in contrasts with the rigidity of primary expenditures recorded in the past.
Finally, there is evidence showing that the institutional design of the budget process has affected fiscal outcomes in accession countries. Budget institutions that appear to be supportive for achieving fiscal discipline are those that strengthen the role of the minister of finance in the budget process and limit the autonomy of spending ministers and individual legislators without responsibility for the budget.
Fiscal soundness is essential for monetary stability. There is much evidence, some of which has been provided rather recently by the accession countries, that it is impossible to maintain price stability without sound public finances. Compared to those examples, the situation in Europe is now comforting. Many EU countries have achieved price stability, and also the accession countries have progressed considerably towards this goal. The fact that risk of high or even moderate inflation is now more or less latent implies that public finances must be on a relatively sound footing. Nevertheless, adopting a longer time perspective, there is no reason to believe that public finances have become for good irrelevant for monetary stability.
I have argued in this presentation that there is no clear line of demarcation separating sound public finances from unsound public finances. Rather, there is a risk of unsound public finances, which more or less continuously depends on a great number of variables such as the level and composition of liabilities, credibility of the fiscal and monetary policy, of the commitments adopted by the government etc. At very low risk levels the probability of a budgetary collapse is virtually zero and the burden the risk places on monetary policy is insignificant. In such a situation a weakening in the government’s budgetary position may have insignificant effects on monetary policy. At very high level of risk, the burden be-comes overwhelming, and a weakening in the government’s budgetary position may again be insignificant, because the collapse is doomed to be inevitable anyway. I would think that the situation relevant for both member and accession countries is somewhere between these two extremes. This means that much depends on the policy adopted by the authorities. An increase in deficit increases the risk of unsound public finances and thereby complicates the task of maintaining monetary stability.
In the accession countries, debt ratios are relatively low. With their higher expected growth rates, accession countries may feel that they are facing favourable debt dynamics, at least in comparison with the present member countries, and that their fiscal policies do not pose any concern for monetary stability. I do not think that this conclusion is warranted. On the contrary, fiscal pressures seem to be emerging at an alarming rate. Successful real convergence is likely to require substantial public expenditure to finance public investment and the expenditure needed to fulfil the obligations of the EU membership. Public expenditure needs also to be stepped up in order to maximise the benefits from the EU membership through structural and cohesion funds. These challenges, some of which have already contributed to a widening of the fiscal deficits in the accession countries, may lead to a further weakening of the budgetary discipline.
A relaxation of budgetary discipline in the new member states would impose a heavy burden on their national monetary policies and could complicate the goal of attaining substantially higher levels of real and nominal convergence. Instead of a weakening of the budgetary discipline, the new member states would benefit from adopting strong and realistic stability-oriented medium- to long-term fiscal frameworks, including where needed changes in budgetary institutions.
Once the accession countries join the EU, the Treaty provisions and all secondary legislation on economic and budgetary policy, in particular the Stability and Growth Pact, will apply to them. This, I believe, is one of the great benefits from the accession. Given the expected state of public finances in the new member states, some of the new member states may have difficulties in complying with the requirements of the pact. In view of the size of the economies, such difficulties would not necessarily directly threaten the monetary stability in the euro area, even in the case these economies would be part of the area. Rather, the main risk to monetary stability is an indirect one, namely that political pressures to loosen the pact may intensify after the accession. In a monetary union, fiscal rules like the Stability and Growth Pact are an essential element of overall policy framework. Therefore, even though one could allow more room for country specific factors in the interpretation of the Stability and Growth Pact in the future, the temptation to weaken it is dangerous and should be firmly resisted.