Italian Public Debt
Why worry?

Italy presents an ideal case study of the effects of public debt: government indebtedness has already grown to 100% of national income and shows few signs of falling. Italy does not, however, appear to have experienced the adverse effects of high levels of public debt predicted by economic theory. How has the Italian economy adapted to the rapidly increasing levels of public debt? This question was the focus for a conference held at Castelgandolfo, Italy on 15/16 June, organized by CEPR Research Fellow Francesco Giavazzi (University of Venice) and Luigi Spaventa (University of Rome).

In his opening presentation Luigi Spaventa outlined developments in the Italian economy since 1970 and their relationship to the growth of public debt, in order to provide a background to the papers presented at the conference. The debt grew initially as a result of larger primary deficits; in the 1980s high real interest rates compounded the deficits. The resulting explosion of debt, particularly since 1980, is remarkable both in comparison with other countries and with Italy's own history. Nevertheless, macroeconomic performance in Italy has not apparently suffered: real growth has on the whole slightly exceeded the European average, Spaventa noted. Real interest rates, however, have been even higher than the growth rate of GDP, and Spaventa noted that this difference had if anything increased during the 1980s. In the 1970s monetary policy was largely a response to government financing requirements. Changes in the relationship between the Banca d'Italia and the Treasury and the adoption of monetary targets meant that in order to finance continued large deficits the authorities were obliged to make government securities more attractive to the private sector. This was achieved, but the result has been to shorten the average maturity of the debt and to increase the real cost of debt service.

Spaventa then sketched what he believed to be the most important theoretical and policy issues underlying the analysis of public debt in Italy. He argued that the high levels of Italian public debt would prove to be a problem when the private sector began to lose confidence in the ability of the state to service its debt. If the rate of real growth exceeded real interest rates then growth would solve the debt service problem, but Spaventa thought this unlikely. As a result, continued confidence rested on the belief that budget surpluses would be run at some time in the future, but announcements of planned deficit reductions lacked credibility; there was considerable opposition to tax increases and little evidence that expenditure targets would be adhered to.

Economic analyses of public debt often rely on theoretical models based on the assumption of a 'representative household'. In his paper 'The End of Large Public Debts', Alberto Alesina (Carnegie- Mellon University and CEPR) argued that this approach was misconceived. The behaviour of public debt is the outcome of a redistributive struggle within the generation currently alive and, to a lesser extent, between current and future generations. Redistribution is intrinsically a political question, as Alesina emphasized in his analysis of the experiences of countries which had issued large quantities of public debt, such as Germany, France, Italy and the UK after the First World War and the United States after the Second World War.

Alesina identified three economic groups which have conflicting attitudes towards public debt: 'rentiers', the holders of the debt; 'businessmen', who hold physical capital and earn profits; and 'workers', who possess human capital and earn wages. The political power held by these groups affects the stability of the political environment which in turn affects the behaviour of public debt.

In an 'unstable' political situation, each of these groups has enough power to 'block' the imposition of taxes whose burden falls on its own members, but none possesses enough political power to impose explicit taxes on other groups. Tax revenues cannot be increased, public debt grows, and eventually an inflationary spiral is created as the government is forced to monetize the debt. Alesina argued that the Weimar Republic in Germany, France in the early 1920s and Italy in the years 1919-22 were examples of instability. In all three cases the socialist 'workers' were strong enough to prevent the rentiers and businessmen from imposing harsh measures on the working class for fear of insurrections. The resulting fiscal deadlocks forced monetization of the debts.

In a stable political situation, by contrast, either one party has a clear majority or more than one party can combine to implement measures to deal with the debt. For example, in England during the interwar period debt reduction was carried out by a dominant coalition of 'rentiers' and 'businessmen'. The conservative governments followed deflationary policies, particularly in the 1920s, rather than resorting to inflation, default or taxes on capital. Most of the burden of reducing the war debt was borne by tax payers, while debt holders enjoyed high rates of return. In the United States after 1945, by contrast, political stability and rapid economic growth meant that war debt could be reduced without severe distributional and political struggles.

What lessons could be drawn from this historical analysis for the current Italian public debt? Alesina argued that a burst of inflation would not work as it had in France or Germany in the interwar period. Italian public debt, like that of other industrialized economies, is of such short maturity that it would not be easily depreciated by inflation. With debt repudiation, the government could start afresh without the need for taxation to service the debt. But Alesina noted that the Italian debt had not been accumulated in exceptional circumstances, such as war, and the current political regime is the same one that issued the debt. So the private sector may not be inclined to view this debt as an exceptional incident never to be repeated, and it would be unlikely to purchase further debt unless the government could credibly commit itself not to default again. The ten-year hiatus after Italy's conversione forzosa of 1926 is a clear example of this, Alesina noted. Debt repudiation would therefore work only if it were accompanied by a credible reduction in the primary budget deficit. A capital levy (a once-off tax on wealth to service and redeem the debt) would also be counterproductive unless the government could commit itself to a fiscal plan which would assure the public that in the foreseeable future no other 'surprise' taxes on wealth would be introduced. Otherwise, such a policy would compromise investors' confidence and generate capital flights.

The debt could also be reduced by budgetary surpluses, but Alesina noted that a lack of political stability makes a such fiscal adjustment particularly difficult in Italy: the credibility of the commitments of a coalition government are doubtful. In this sense, Alesina concluded, the public debt is ultimately a political problem. The Italian economy is therefore likely to experience a long period of very high debt/GNP ratios and a high debt service ratio.

Alesina's historical analysis of capital levies highlighted the importance of investors' confidence and the monetary and exchange rate turbulence generated by capital flight. This emphasis was shared by Alberto Giovannini (Columbia University and CEPR) in his paper, 'Capital Controls and Public Finance: The Experience in Italy'. Giovannini analysed the events surrounding the major changes in capital controls, focusing on the interwar period and the 1970s. In March 1919, the lira exchange rate collapsed dramatically and remained weak until the second half of 1920. Why did this collapse occur? At the end of World War I Italy suffered from high inflation and a large government debt. The war had reduced government's capacity to tax its citizens, as well as its ability to finance deficits in the domestic bond markets. Government spending was kept high by agricultural and industrial subsidies. The need to improve public finances prompted the authorities to impose a capital levy at rates ranging from 4.45 to 50%. Giovannini argued that wealth holders had correctly anticipated the imposition of the capital levy and had moved funds abroad: this explained the dramatic collapse of the lire. The authorities responded to capital flight by imposing restrictions on capital movements.

Italy also experienced large capital outflows in the 1930s. The state of public finances was rapidly deteriorating and larger budget deficits were expected when in 1934 the government clearly hinted its colonial ambitions. Subsequent events justified the fears of wealth owners: the government imposed strict controls on foreign exchange transactions and levied a series of income and wealth taxes to finance the colonial wars in Africa. Fears of escalating public debt and the desire to evade future tax increases were essential, Giovannini concluded, to any explanation of capital flight in the interwar years, although the overvaluation of the lira also contributed to these speculative outflows.

Capital controls in Italy can therefore be viewed as a response to difficulties in financing government spending. Giovannini supplemented his historical analysis with a theoretical model of the macroeconomic implications of capital flight and tax evasion. Giovannini used a model of government debt due to Diamond, extended to allow domestic residents to trade in goods and assets with the rest of the world. The government can tax only financial assets. If it can tax at the same rate domestic residents' income on all assets, domestic and foreign, there is no incentive to substitute foreign for domestic assets. If income from foreign assets is tax-free, however, either by law or because the law cannot be enforced, any tax increase on income from domestic assets will lead domestic residents to shift towards foreign securities. Capital flight continues until the after-tax rates of return on domestic and foreign investments are equalized. The increase in government debt therefore crowds out the domestic capital stock through tax evasion; this in turn affects the real wage, saving and the current account.

Once all the effects of capital flight for the purpose of tax evasion are taken into account, the case for or against capital controls is not clear-cut in Giovannini's model. In the presence of tax evasion, increases in tax revenue bring about distortions on the production side of the economy; on the other hand, when controls prevent tax evasion through capital flight, changes in tax revenue are associated with distortions of the intertemporal rate of substitution. Which type of distortion is less desirable remains an empirical question, but there was some evidence to indicate that the latter were less important.

There have recently been moves towards liberalization of Italian capital controls. Although inflation has fallen, Italy still relies heavily on the inflation tax (the product of the inflation rate times the monetary base). Doubts about future inflation are intensified by the state of public finances: even under the most optimistic scenarios, Italy will not be able to grow out of its government debt. This implies either higher taxes or increased monetization of government deficits in the future. The prospects of renewed inflation, combined with freer international capital flows meant that keeping the lira within the EMS parities would become much harder. The Italian authorities, Giovannini concluded, will be forced to choose the degree of capital account deregulation jointly with the exchange rate regime.

During the 1980s, the Bank of Italy has increased its autonomy from the Treasury and has pursued its monetary objectives with greater determination. In his paper, 'Monetary and Fiscal Policy Coordination with a High Public Debt', Guido Tabellini (University of California at Los Angles and CEPR) asked whether there was a contradiction between the goal of stabilizing public debt and the decentralization of the monetary and fiscal responsibility.

Tabellini first examined the relationship between the fiscal and monetary authorities using a simple two-period model of an open economy under flexible exchange rates. There were three players: the private sector, the fiscal authority and the central bank. The fiscal authority is assumed to care only about the level of public expenditure, which it attempts to maximize over the two periods, whereas the central bank only cares about the private sector's welfare, which depends on consumption and real money balances. The actions of the two policy-makers are linked by the intertemporal government budget constraint: institutional arrangements determine which authority bears the residual burden of satisfying this constraint. Tabellini's theoretical analysis predicted an inverse relationship between the size of fiscal deficits and the extent of debt monetization by the central bank.

Had the recent Italian institutional changes brought about a fiscal policy more disciplined by the intertemporal budget constraint? If the monetary reforms of the 1980s reduced the degree of fiscal dominance in Italy, then fiscal deficits should be a better predictor of inflation before the monetary reforms than afterwards. Tabellini tested this by regressing inflation on its own lagged values and on lagged values of the deficit and other variables. The results strongly contradicted the conjecture that the monetary reforms had reduced the degree of fiscal dominance: deficits were even more inflationary under the new regime.

Tabellini maintained nevertheless that the degree of fiscal dominance had probably diminished; this was not apparent in the data because the institutional reforms were implemented gradually throughout the 1980s and because private sector expectations may have reacted slowly to the reforms. He also argued that the process of reform was still largely incomplete and so may lack credibility. The Treasury can still create liquidity through its overdraft account at the Bank of Italy, the ceiling on which is 14% of public expenditure in the current period. In addition the Treasury effectively sets a ceiling on interest rates in the Italian money market by choosing the minimum base price at the public debt auctions. If the reforms are not carried further, Tabellini concluded, monetary and fiscal policies in Italy might resemble the equilibrium outcome of a game of chicken: a tight monetary policy motivated by the central bank's desire to establish credibility, accompanied by large fiscal deficits based on the expectation of future debt monetization by an acquirescent central bank.

Public debt is now roughly equal to national income in Italy and and interest payments are around 10% of income. Before 1984 interest income on government securities was exempt from personal income tax, although corporations were taxed indirectly. These arrangements were modified in 1984 and again in 1986 and now both households and corporations are now taxed on their income from government securities. The tax treatment of asset returns clearly affects the private sector's asset demands and hence the rates of return on securities. In their paper 'Taxation and the Distributive Effects of Public Debt', Francesco Frasca (Banca d'Italia) and Ruggero Paladini (University of Rome) explored the effects of these tax changes on asset demands, interest rates and the behaviour of the public borrowing requirement, using simulations of the financial sector of the Bank of Italy's econometric model.

Public debt in Italy is now so large that its efficient management has an important impact on public finances and economic welfare. In 'The Management of Public Debt and Financial Markets', Marco Pagano (University of Naples and CEPR) explored the lessons economic theory offered for debt management. By issuing public debt the government can reallocate income across generations. Pagano therefore used an overlapping-generations model in order to capture these intergenerational effects. His analysis suggested that the 'representative household' would benefit if deficits behaved anticyclically with respect to income and absorbed abnormal bulges in public spending. Taxes, on the other hand, should move procyclically, and should be set so that on average they will pay for permanent government expenditure and real interest on debt. More strikingly, the optimal path of debt in these models was independent of the initial level of public debt.

Pagano then turned to the the optimal composition of debt issued by the government. In the presence of uncertainty and incomplete markets, the government can issue appropriately designed debt instruments which provide the private sector with new opportunities for risk-sharing, both within and between generations. Italian capital markets did not allocate risks effectively, Pagano argued, hence the Italian authorities had considerable scope for useful innovations in the menu of public debt. His analysis of portfolio choice within an overlapping generations model suggested that, given the strong negative correlation between the real yield on equities and inflation and the decreasing degree of indexation of labour income and pensions in Italy, the government should issue debt indexed to the inflation rate. Such debt would sell at a premium, as would any public debt instrument whose return was negatively correlated with returns on the market portfolio. Issues of indexed debt would not only improve risk allocation but would also help the Treasury minimize the burden of debt servicing.

Public sector deficits began to soar dramatically in Italy during the 1970s. At the same time, the rapid increase in inflation inflicted large capital losses on bond holders and made it increasingly difficult to finance the deficits via new debt issues. These difficulties were overcome after 1975: short-term Treasury Bills and, later, medium-term Treasury Credit Certificates (whose returns were linked to those of Treasury Bills) grew to occupy a large portion of private portfolios. Household financial assets have therefore undergone a substantial shift from currency and deposits to government securities, particularly after 1980. Andrea Bollino (Banca d'Italia) and Nicola Rossi (IMF), in their paper 'Italian Households' Demand for Monetary Assets and Government Debt', examined whether these shifts in the composition of household portfolios could be explained by econometric model of consumer choice. The household is assumed to maximize a utility function which includes non- durable consumer goods, labour services and the flow of services from consumer durables and monetary assets. Demand functions for financial assets can then be derived and their parameters estimated subject to a number of testable restrictions corresponding to the underlying household utility function.
Bollino and Rossi focused on the demand for services of monetary assets. They estimated the parameters of the 'technology' through which households combine the available financial instruments in order to produce a good which can be called 'liquidity services'. Their estimates implied a pattern of substitution among financial assets which allowed the effects of changes in interest rates on asset holdings to be examined. The pattern revealed that it was appropriate to treat both short-term and floating-rate government securities as a single good. The estimates also confirmed that households treat bank deposits and government securities as close substitutes in generating 'liquidity services', but that the degree of 'moneyness' of short-term and floating-rate government debt is not large enough to distort the meaning of traditional monetary aggregates which exclude such debt. Bollino and Rossi concluded that the composition of household portfolios is highly responsive to relative price signals. This has important policy implications in the light of the increasing degree of financial market integration, both within and outside the country.

The papers and proceedings of this conference, edited by Francesco Giavazzi and Luigi Spaventa, will be published under the title 'High Public Debt: The Italian Experience' by Cambridge University Press in April 1988.