Italy has a rapidly growing public debt, which exerts a pervasive influence on the working of its financial markets. Heavy administrative controls have been imposed on both domestic and international flows of funds. In recent years, fiscal adjustment has become more difficult because of a huge increase in real interest rates, which has followed in the wake of a process of financial deregulation.

This paper is concerned with the relation between government deficit and direct controls on financial markets. Section I presents an accounting framework to analyze the increase of Italian public debt during the last ten years; in particular, it estimates the impact of changes in interest rates and the amount of seigniorage collected by the Treasury.

Sections II and III examine the administrative controls that have been used to sustain the private sector’s demand for government bonds and monetary base, and thus to collect implicit taxes to help finance the deficits. These sections also describe the recent trend toward liberalization. Section IV analyzes the relation between the “tax from controls” and the well-known concept of “inflation tax.” Section V looks at the stock market and discusses the consequences of controlling international capital movements on the cost of raising equity capital by the corporate sector and the implications for the working of the crowding-out mechanism. Section VI contains some concluding comments on the possible effects of financial deregulation and its impact on the process of fiscal adjustment.

I. Accounting for Italian Public Debt Growth

During the last 12 years, the real value of the Italian public debt held outside the central bank increased at an average annual rate of about 10.5 percent; its ratio to gross domestic product (GDP) is now around 78 percent (93 percent including the part held by the central bank), having increased at a rate of around 3.5 percentage points a year.

This growth of public indebtedness is probably unsustainable.¹ To analyze its impact on financial markets, it is useful to divide it into three parts: first, the contribution of the public sector deficit net of interest expenses; second, the average real ex-post interest rate paid to private sector holders of the debt; and third, the negative contribution of the Treasury’s seigniorage obtained by borrowing from the central bank. Both the second and the third components have to be computed net of the Treasury’s interest payments on its balances with the central bank,² which in Italy approximately correspond to the payment by the central bank of a 5.5 percent interest rate to the banking system on its compulsory reserve deposits.

Table 1 shows this breakdown of the rate of growth of real public debt. Table 2 presents the equivalent breakdown of the increase in the debt-to-GDP ratio, where the real growth rate of the denominator is subtracted from the real interest rate component. (See Appendix for an explanation of the simple algebra used for the calculation.)

Table 1.

Italy: Breakdown of Growth Rate of Real Value of Public Debt Held Outside the Central Bank, 1976-87

(Average yearly rates in percent)

Note: See equation (1) in the Appendix. Calculations are based on Bank of Italy data and official forecasts. The decomposition for 1987 has been partially estimated by the authors; it is not exact, and small errors of approximation have been distributed proportionally for rounding-off purposes.

Table 1.

Italy: Breakdown of Growth Rate of Real Value of Public Debt Held Outside the Central Bank, 1976-87

(Average yearly rates in percent)

			Contribution of:
	Growth Rate		Deficit Net of		Real Interest		Treasury’s
Years	of Real Debt	=	Interest	+	Rate on Debt	-	Seigniorage
1976-81	8.8		17.0		−3.5		4.7
1982-87	12.2		10.8		4.0		2.6
1976-87	10.4		13.5		0.3		3.4
1984-87	10.7		8.3		5.3		2.9

Note: See equation (1) in the Appendix. Calculations are based on Bank of Italy data and official forecasts. The decomposition for 1987 has been partially estimated by the authors; it is not exact, and small errors of approximation have been distributed proportionally for rounding-off purposes.

View Table

Table 1.

Italy: Breakdown of Growth Rate of Real Value of Public Debt Held Outside the Central Bank, 1976-87

(Average yearly rates in percent)

			Contribution of:
	Growth Rate		Deficit Net of		Real Interest		Treasury’s
Years	of Real Debt	=	Interest	+	Rate on Debt	-	Seigniorage
1976-81	8.8		17.0		−3.5		4.7
1982-87	12.2		10.8		4.0		2.6
1976-87	10.4		13.5		0.3		3.4
1984-87	10.7		8.3		5.3		2.9

Note: See equation (1) in the Appendix. Calculations are based on Bank of Italy data and official forecasts. The decomposition for 1987 has been partially estimated by the authors; it is not exact, and small errors of approximation have been distributed proportionally for rounding-off purposes.

Table 2.

Italy: Breakdown of Change in Ratio of Nonmonetary Public Debt to GDP

Note: See equation (2) in the Appendix and Note to Table 1.

Table 2.

Italy: Breakdown of Change in Ratio of Nonmonetary Public Debt to GDP

			Contribution of
	Average Yearly		Real Interest Rate
	Change in GDP		on Debt Less Real		Debt/GDP		Deficit Net of		Treasury’s
Years	Ratio	=	Growth of GDP	×	Ratio	+	Interest	-	Seigniorage
	(in percent)						(in percent of GDP)
1976–81	1.6		−6.9		0.42		6.6		2.1
1982–87	5.4		1.7		0.62		6.1		1.7
1976–87	3.3		−2.4		0.51		6.4		1.9
1984–87	4.3		2.2		0.70		4.9		2.1

Note: See equation (2) in the Appendix and Note to Table 1.

View Table

Table 2.

Italy: Breakdown of Change in Ratio of Nonmonetary Public Debt to GDP

			Contribution of
	Average Yearly		Real Interest Rate
	Change in GDP		on Debt Less Real		Debt/GDP		Deficit Net of		Treasury’s
Years	Ratio	=	Growth of GDP	×	Ratio	+	Interest	-	Seigniorage
	(in percent)						(in percent of GDP)
1976–81	1.6		−6.9		0.42		6.6		2.1
1982–87	5.4		1.7		0.62		6.1		1.7
1976–87	3.3		−2.4		0.51		6.4		1.9
1984–87	4.3		2.2		0.70		4.9		2.1

Note: See equation (2) in the Appendix and Note to Table 1.

The analysis can go deeper if the Treasury seigniorage is also broken down. Seigniorage is obtained when the deficit is financed via the central bank’s issue of the monetary base. New injections of these issues will find their place in the private sector portfolio for three reasons. First, a certain amount of inflation will develop, cutting the real value of the pre-existing base. Second, an increase in the real demand for money base can take place as a consequence of a change in real income and/or in the velocity of circulation of the base. Third, the composition of the monetary base can change, with an increase in the portion issued to finance the Treasury. Thus, for a given real demand for total monetary base, the private sector will accept a larger amount of the Treasury base when the other parts of the base (issued or destroyed as a counterpart to changes in official reserves and/or in the central bank’s credit to the private sector) decrease or become proportionally smaller. Table 3 evaluates the contribution of these three causes of the “seigniorage tax” collected by the Treasury, subtracting the negative contribution of the Treasury’s interest payments to the central bank. The tax can be expressed as the product of a tax rate and a tax base,³ and the latter can be related to the stock of nonmonetary debt or to GDP. (See Appendix for a formal derivation of the analytical division.)

Table 3.

Breakdown of Italian Treasury’s Seigniorage, 1976-87

(Average yearly rates, in percent except where otherwise noted)

Note: See equation (1) in the Appendix and Note to Table 1. The average interest rate on the monetary base has been estimated by setting the rate on the overdraft account of the Treasury with the Bank of Italy at 1 percent, and the rate on Treasury notes and bonds held by the Bank at 5.5 percent.

^*Public debt held by the private sector.

Table 3.

Breakdown of Italian Treasury’s Seigniorage, 1976-87

(Average yearly rates, in percent except where otherwise noted)

	Seigniorage Tax Rate
		Change in Real Stock						Seigniorage Tax Base
		of Total Monetary Base			Change in	Avg Interest		(monetary base of
					Treasury Base/	Rate on		the Treasury)		Total Seigniorage
		Real GDP		Total	Monetary	Treasury	Total			(7)x(8)	(7)x(9)
	Inflation	Growth	Velocity	(2)–(3)	Base Ratio	Monetary Base	(1) + (4) + (5) - (6)	(percent of debt*)	(percent of GDP)	(percent of debt*))	(percent of GDP)
Years	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
1976–81	16.6	3.4	3.0	0.4	−1.8	4.1	11.1	0.42	0.19	4.7	2.1
1982–87	10.1	2.2	−0.7	2.9	1.4	3.2	11.2	0.23	0.15	2.6	1.7
1976–87	13.4	2.8	1.2	1.6	−0.2	3.6	11.2	0.30	0.17	3.4	1.9
1984–87	7.5	3.1	−1.8	4.9	4.2	3.3	13.3	0.22	0.16	2.9	2.1

Note: See equation (1) in the Appendix and Note to Table 1. The average interest rate on the monetary base has been estimated by setting the rate on the overdraft account of the Treasury with the Bank of Italy at 1 percent, and the rate on Treasury notes and bonds held by the Bank at 5.5 percent.

^*Public debt held by the private sector.

View Table

Table 3.

Breakdown of Italian Treasury’s Seigniorage, 1976-87

(Average yearly rates, in percent except where otherwise noted)

	Seigniorage Tax Rate
		Change in Real Stock						Seigniorage Tax Base
		of Total Monetary Base			Change in	Avg Interest		(monetary base of
					Treasury Base/	Rate on		the Treasury)		Total Seigniorage
		Real GDP		Total	Monetary	Treasury	Total			(7)x(8)	(7)x(9)
	Inflation	Growth	Velocity	(2)–(3)	Base Ratio	Monetary Base	(1) + (4) + (5) - (6)	(percent of debt*)	(percent of GDP)	(percent of debt*))	(percent of GDP)
Years	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
1976–81	16.6	3.4	3.0	0.4	−1.8	4.1	11.1	0.42	0.19	4.7	2.1
1982–87	10.1	2.2	−0.7	2.9	1.4	3.2	11.2	0.23	0.15	2.6	1.7
1976–87	13.4	2.8	1.2	1.6	−0.2	3.6	11.2	0.30	0.17	3.4	1.9
1984–87	7.5	3.1	−1.8	4.9	4.2	3.3	13.3	0.22	0.16	2.9	2.1

Note: See equation (1) in the Appendix and Note to Table 1. The average interest rate on the monetary base has been estimated by setting the rate on the overdraft account of the Treasury with the Bank of Italy at 1 percent, and the rate on Treasury notes and bonds held by the Bank at 5.5 percent.

^*Public debt held by the private sector.

Table 1 shows that in the post-1982 years the real debt growth rate was much higher than in the previous six-year period, in spite of the much smaller contribution of the accumulation of deficits net of interest expenses. This rise in the growth of debt is mainly explained by the huge increase in the real interest rate paid on the debt, although the more moderate resort to the seigniorage tax was also partly responsible. If one focuses on the last four years, the increase in the interest component is even more impressive.

The picture looks the same in Table 2, which shows a breakdown in the increase in the debt-to-income ratio. The third column of the Table makes it clear that besides the increase in the real rate of interest, which in the last few years has been significantly higher than GDP growth (as shown in the second column), the explosion of interest expenses was caused by the higher debt-to-income ratio accumulated over time.

When seigniorage is measured as a percent of GDP, as in Table 2, its reduction during the 1980s does not appear so strong; to be sure, in the four years 1984–87, the weight of Treasury seigniorage returned to its average level during the high-inflation period of 1976–81,⁴ and constituted a high percentage of total explicit tax revenues.⁵ This development makes the recent strength of debt growth more worrisome, because it can hardly be considered the result of a more prudent resort to monetary financing. It also makes clear that seigniorage does not coincide with the inflation tax.

We can now consider Table 3, which is a breakdown of seigniorage. Columns 8 and 9 show a definite decrease in the seigniorage tax base, even if, when expressed in proportion to GDP, the trend appears somewhat weak and has not continued in more recent years. The stock of Treasury indebtedness with the central bank remains high and constitutes a robust base from which seigniorage can be extracted in favor of the Treasury. Nevertheless, the decrease in that stock (in proportion to total debt and GDP) is the only cause of the lower weight of tax seigniorage in the post-1982 period, because, as shown in column 7, the tax rate has remained constant in the two periods of the past 12 years (1976–81 and 1982–87), but has clearly increased in the years 1984–87.

How could this happen with a sharply decreasing inflation rate? The answer can be found in columns 4 and 5 of Table 3. In the last four to five years, the real demand for total monetary base has been rising rapidly because of a pronounced decrease in its velocity of circulation (see column 3); at the same time, a higher proportion of its supply has been provided through the Treasury’s financing.⁶ Had the ratio of GDP to monetary base, and the proportion of the monetary base issued in favor of the Treasury, been stable since 1984, the same seigniorage could have been collected by the Treasury, other things being equal, if the yearly inflation rate had been only 6 percentage points (1.8 percent + 4.2 percent) higher; that is, if prices had increased at about the same rate as in 1983.

The increase in the base’s velocity is in part a consequence of the behavior of money velocity, an international phenomenon that is not easily explained but that is definitely related to the decrease in interest rates and inflation. However, as we argue in Section IV, the higher demand for monetary base has also been caused by a compulsory reserve regulation imposed on banks, which has functioned as a taxing device to collect seigniorage. The increase in the Treasury’s monetary base in relation to the total monetary base cannot be attributed entirely to a strategy directed toward collecting seigniorage; it is rather the indirect result of exchange rate policy and the destruction of the monetary base through the balance of payments.

It is fairly unlikely that both velocity and the Treasury share of high-powered money will be able to change in the same direction and at the same speed in the near future. This will complicate the problem of stabilizing the rate of growth of public debt in the coming years. If the real rate of interest, the rate of inflation, and the rate of growth of GDP stay at present levels, the reduction in the deficit net of interest will have to be very strong and fast, in order to make up for a probable decrease in the collection of seigniorage.

To better understand the past dynamics of the various components of Italian public debt growth and the current and future problem of fiscal adjustment, administrative controls on financial markets have to be taken into consideration. We can think of them as a form of implicit taxation, which can in part substitute and/or complement Treasury seigniorage, even if, unlike Treasury seigniorage, they do not have a universally acknowledged fiscal dimension. We shall try to show how the “tax-from-controls” concept can be especially helpful in a discussion of the behavior of the interest component of debt accumulation; it can also be used to help explain the working of the seigniorage-inflation tax mechanism and the reaction of the noninterest deficit to monetary stabilization.

II. Controls Sustaining Demand for Public Bonds

Administrative controls on domestic and international financial flows can have the indirect effect of artificially sustaining the private sector’s demand for public bonds for any given level of their interest yield by lowering the interest rate required to finance a given public deficit through the issue of bonds. The decrease in the interest cost of public sector borrowing that is obtained in this way can be considered an implicit tax collected by the Treasury.

The presence of a substantial tax of this kind is undoubtedly the explanation for the significant negative differential between the real interest cost of Italian public debt and the international real interest rate that obtained during the 1976–81 period.⁷ Conversely, the removal of an important set of financial controls is certainly one of the main reasons for the huge increase of the interest cost that took place, as mentioned above, after 1982. A reintroduction of controls might therefore be considered one of the instruments required to bring about fiscal adjustment by slowing down the rate of growth of public debt. However, the negative effects and distortions caused by this solution should be stressed. Controls diffuse the pressures that would otherwise force the authorities to make the political decision to tackle the first-best solution—that is, to reduce the basic deficit, net of interest, and to rationalize the structure of taxes and public expenditure so as to enhance the productivity of the public (and private) sector. In what follows, we shall comment on both the distortions and on the decrease of fiscal discipline that we think result from financial protectionism.

The characteristics and time profile of the main administrative controls that have been phased in and out of the Italian financial system during the 1970s and the 1980s are summarized in Table 4, which shows the evolution of the most important controls on both domestic financial markets and on international capital flows. Many of these controls were considered short-term measures, designed to enhance the authorities’ ability to influence aggregate demand and external accounts. In fact, the controls have played an important structural role in sustaining the demand for public bonds, even if this objective was achieved in an indirect way and without explicit declaration by the authorities.

Table 4.

Controls Sustaining Demand for Public Bonds: Main Measures, 1970-87

Table 4.

Controls Sustaining Demand for Public Bonds: Main Measures, 1970-87

Control		1970–79	1980–87
Domestic Controls
	Ceiling on bank loans	July 1973. First introduction (renewed on several occasions). March 1975. Suspension. October 1976. Re production, confirmed by subsequent measures.	March 1980. Strengthened by introduction of penalty on overdrafts. June 1983. Suspension. January 1986. Reproduction for six months. September 1987. Reintroduction for six months.
	Portfolio requirement (to buy bonds issued by special credit institutions)	June 1973. First introduction.	January 1987. Suspension.
Controls on International
Capital Flows
	Capital outflows	June 1972. Suspension of external convertibility of Italian banknotes. June 1973. Introduction of 50 percent zero-interest deposit on investment abroad.	November 1984. Reduction of compulsory deposit to 40 percent (30 percent for EEC securities). October 1985. Reduction of deposit to 25 percent. August 1986. Reduction of deposit to 15 percent. May 1987. Abolition of deposit.
Foreign trade financing		Measures directed at: —discouraging deferred settlements of exports and advanced settlements of imports; —imposing foreign currency financing requirements. Many policy measures; their stance has changed according to evolution of foreign accounts. Controls favored sizable expansion of foreign liabilities.	No radical changes in objectives. Recent increases in terms of settlements and attempts to reduce foreign currency financing requirements.
Regulation of external position of banks		June 1972. Banks authorized to assume position of net external indebtedness and forbidden to assume a creditor position. July 1974. Imposition of ceiling on net external indebtedness. July 1975. Ceiling lifted on net external indebtedness.	July 1984. Imposition of ceiling on net external indebtedness. December 1985. Ceiling lifted on net external indebtedness. March 1987. Introduction of reserve requirement on net external indebtedness. September 1987. Abolition of reserve requirement on net external indebtedness.

View Table

Table 4.

Controls Sustaining Demand for Public Bonds: Main Measures, 1970-87

Control		1970–79	1980–87
Domestic Controls
	Ceiling on bank loans	July 1973. First introduction (renewed on several occasions). March 1975. Suspension. October 1976. Re production, confirmed by subsequent measures.	March 1980. Strengthened by introduction of penalty on overdrafts. June 1983. Suspension. January 1986. Reproduction for six months. September 1987. Reintroduction for six months.
	Portfolio requirement (to buy bonds issued by special credit institutions)	June 1973. First introduction.	January 1987. Suspension.
Controls on International
Capital Flows
	Capital outflows	June 1972. Suspension of external convertibility of Italian banknotes. June 1973. Introduction of 50 percent zero-interest deposit on investment abroad.	November 1984. Reduction of compulsory deposit to 40 percent (30 percent for EEC securities). October 1985. Reduction of deposit to 25 percent. August 1986. Reduction of deposit to 15 percent. May 1987. Abolition of deposit.
Foreign trade financing		Measures directed at: —discouraging deferred settlements of exports and advanced settlements of imports; —imposing foreign currency financing requirements. Many policy measures; their stance has changed according to evolution of foreign accounts. Controls favored sizable expansion of foreign liabilities.	No radical changes in objectives. Recent increases in terms of settlements and attempts to reduce foreign currency financing requirements.
Regulation of external position of banks		June 1972. Banks authorized to assume position of net external indebtedness and forbidden to assume a creditor position. July 1974. Imposition of ceiling on net external indebtedness. July 1975. Ceiling lifted on net external indebtedness.	July 1984. Imposition of ceiling on net external indebtedness. December 1985. Ceiling lifted on net external indebtedness. March 1987. Introduction of reserve requirement on net external indebtedness. September 1987. Abolition of reserve requirement on net external indebtedness.

Among the domestic controls, the most significant were the administrative constraints placed on bank assets. These constraints did not directly force the banking system to hold government securities; rather, this objective was accomplished indirectly through a ceiling on bank loans that imposed severe limits on the growth of private sector credit. The ceiling was introduced for the first time as an emergency measure in July 1973. Although it was only supposed to be in effect for eight months, it was subsequently renewed on several occasions, and, with the exception of the period March 1975–October 1976, it remained in force until June 1983.

The ceiling increased the placement of public bonds with the banking system; by the end of the 1970s, their share in total public debt, which was 15 percent in 1973, had more than doubled (Table 5). The ceiling also reduced the availability of credit to the private sector and increased its cost, while at the same time it created many distortions in financial markets, among the most important of which was a reduction in competition in the market for bank loans.⁸

Table 5.

Italy: Public Debt, 1970–86

(In percent)

Table 5.

Italy: Public Debt, 1970–86

(In percent)

Source		1970	1971	1972	1973	1974	1975	1976	1977	1978	1979	1980	1981	1982	1983	1984	1985	1986
Borrowing from the Bank of Italy		25.6	24.0	23.3	28.0	31.7	36.7	40.0	30.7	27.3	22.7	23.2	23.5	21.8	17.5	16.6	17.6	16.5
Postal deposits		21.1	20.5	20.7	19.5	17.3	16.0	15.6	15.5	15.2	16.2	14.5	12.6	10.9	9.7	8.9	8.7	8.9
Borrowing from credit
	institutions	30.9	35.1	36.3	34.5	33.8	37.0	33.9	40.9	40.9	40.0	37.0	31.8	35.4	35.5	32.0	27.9	24.8
	Bills and securities	13.6	17.0	17.2	15.0	14.8	20.1	17.5	29.4	32.0	31.2	29.3	25.3	28.9	29.1	25.2	22.6	20.1
	Loans	17.3	18.1	19.1	1.95	18.9	16.9	16.4	11.5	8.9	8.8	7.7	6.5	6.5	6.4	6.8	5.3	4.7
Bills and securities placed with economy		17.6	16.2	16.3	14.9	14.3	7.9	8.0	10.7	14.5	19.1	23.1	29.2	28.7	34.1	39.2	42.7	47.2
Other debts		2.1	1.9	1.6	1.4	1.3	1.1	1.0	0.9	0.9	0.8	0.7	0.6	0.5	0.5	0.4	0.4	0.3
Foreign debt		2.2	2.3	1.8	1.7	1.6	1.3	1.5	1.3	1.3	1.2	1.5	2.3	2.7	2.7	2.9	2.7	2.2

View Table

Table 5.

Italy: Public Debt, 1970–86

(In percent)

Source		1970	1971	1972	1973	1974	1975	1976	1977	1978	1979	1980	1981	1982	1983	1984	1985	1986
Borrowing from the Bank of Italy		25.6	24.0	23.3	28.0	31.7	36.7	40.0	30.7	27.3	22.7	23.2	23.5	21.8	17.5	16.6	17.6	16.5
Postal deposits		21.1	20.5	20.7	19.5	17.3	16.0	15.6	15.5	15.2	16.2	14.5	12.6	10.9	9.7	8.9	8.7	8.9
Borrowing from credit
	institutions	30.9	35.1	36.3	34.5	33.8	37.0	33.9	40.9	40.9	40.0	37.0	31.8	35.4	35.5	32.0	27.9	24.8
	Bills and securities	13.6	17.0	17.2	15.0	14.8	20.1	17.5	29.4	32.0	31.2	29.3	25.3	28.9	29.1	25.2	22.6	20.1
	Loans	17.3	18.1	19.1	1.95	18.9	16.9	16.4	11.5	8.9	8.8	7.7	6.5	6.5	6.4	6.8	5.3	4.7
Bills and securities placed with economy		17.6	16.2	16.3	14.9	14.3	7.9	8.0	10.7	14.5	19.1	23.1	29.2	28.7	34.1	39.2	42.7	47.2
Other debts		2.1	1.9	1.6	1.4	1.3	1.1	1.0	0.9	0.9	0.8	0.7	0.6	0.5	0.5	0.4	0.4	0.3
Foreign debt		2.2	2.3	1.8	1.7	1.6	1.3	1.5	1.3	1.3	1.2	1.5	2.3	2.7	2.7	2.9	2.7	2.2

In July 1973 another administrative constraint on bank assets was imposed with the introduction of a portfolio requirement, by which banks were committed to buy bonds issued by special credit institutions. These institutions extended long-term loans, and at the beginning of the 1970s had been experiencing serious problems in their financing because of the high liquidity preference of Italian savers. The requirement, which obliged the banks to give to the credit institutions the funds they were not able to secure on the market, was initially stringent (40 percent of the increase in banks’ deposits in the first half of the 1970s), but was relaxed subsequently.

The portfolio requirement also had an effect on the market for public bonds for at least two reasons. The yields on bonds issued by the special institutions were kept below the market levels, thus dampening the competitive pressure they would otherwise have exerted on public bonds. It should also be remembered that during the 1970s, special institutions had extended many subsidized loans; the portfolio requirement, by allowing a reduction of the public sector expenditure on those subsidies, helped to reduce the public deficit.

Foreign exchange regulation has also had an important effect on the effort to sustain the demand for public bonds. Until the beginning of the 1970s, Italian foreign exchange policy was directed at achieving a progressive financial integration. But in 1972–73 the policy stance was reversed and was replaced by a policy characterized by increasing financial protectionism—the acquisition of financial assets abroad by the economy was severely restricted, and the raising of funds on international financial markets was stimulated.

The measures taken fall into the three broad categories shown in Table 4:⁹ (1) controls on capital outflows; (2) controls on terms and conditions of foreign trade financing; and (3) regulation of the external position of banks.

The most important among the measures designed to control capital outflows was the imposition in June 1973 of a compulsory zero-interest deposit on property and portfolio investment abroad. This measure affected the assets side of the process of financial integration, leading to a drastic reduction of the share of external financial assets held domestically, particularly by the household sector. By preventing the legal acquisition of foreign bonds, the compulsory deposit forced Italian savers to increase their demand for domestic assets, thus helping to sustain the demand for public bonds.

The primary aim of the measures designed to control the terms and conditions of foreign trade financing, on the one hand, was to prevent speculative capital movements connected with foreign trade. Their principal effect was to increase the demand for foreign currency loans from the banks. The regulation of the external position of banks, on the other hand, was intended to increase the supply of foreign loans by the banking system. On the whole, these measures encouraged the banks to borrow abroad, thereby enabling them to meet the domestic demand for foreign currency loans. In fact, the regulations prevented the banks from assuming a net creditor external position and permitted them to maintain an unlimited net debtor position, except for short periods. These two different categories of controls were complementary and were used intensively during foreign exchange crises to keep official reserves above their equilibrium levels. By stimulating the growth of foreign liabilities, they also achieved the objective of dampening the demand by enterprises for domestic financing, thereby creating conditions for a more favorable placement of public bonds in the domestic markets. Foreign exchange regulations also helped to keep Italian real rates of interest at levels lower than those prevailing in other countries (see Tables 6 and 7).¹⁰

Table 6.

Interest Rates: Italy and Other Industrial Countries

(Real short-term rates in percent)

Table 6.

Interest Rates: Italy and Other Industrial Countries

(Real short-term rates in percent)

Country	1981	1982	1983	1984	1985	1986
United States	6.62	7.03	4.20	5.99	4.42	3.75
Japan	4.48	4.76	4.22	4.34	5.04	5.83
Federal Republic of Germany	6.8	4.90	2.92	3.68	4.72	5.12
France	1.96	5.64	3.91	5.41	6.45	4.50
United Kingdom	3.18	7.39	5.06	4.54	7.36	7.04
Italy	3.87	3.85	56.14	6.96	6.45	7.69
Canada	7.26	6.72	4.38	7.54	5.39	5.18

View Table

Table 6.

Interest Rates: Italy and Other Industrial Countries

(Real short-term rates in percent)

Country	1981	1982	1983	1984	1985	1986
United States	6.62	7.03	4.20	5.99	4.42	3.75
Japan	4.48	4.76	4.22	4.34	5.04	5.83
Federal Republic of Germany	6.8	4.90	2.92	3.68	4.72	5.12
France	1.96	5.64	3.91	5.41	6.45	4.50
United Kingdom	3.18	7.39	5.06	4.54	7.36	7.04
Italy	3.87	3.85	56.14	6.96	6.45	7.69
Canada	7.26	6.72	4.38	7.54	5.39	5.18

Table 7.

Interest Rate Differentials: United States and Federal Republic of Germany

Note: Average of the differentials with respect to the United States and the Federal Republic of Germany calculated using monthly data on short-term (three months) interest rates.

¹Deflated with the consumption price index.

²Forward discount of the lira (monthly averages).

Table 7.

Interest Rate Differentials: United States and Federal Republic of Germany

	January 1977–	April 1979–	April 1983–
Differential	March 1979	March 1983	July 1987
Real differential1	−2.73	−3.84	+0.95
Nominal covered differential2	−1.89	−2.45	+0.80

Note: Average of the differentials with respect to the United States and the Federal Republic of Germany calculated using monthly data on short-term (three months) interest rates.

¹Deflated with the consumption price index.

²Forward discount of the lira (monthly averages).

View Table

Table 7.

Interest Rate Differentials: United States and Federal Republic of Germany

	January 1977–	April 1979–	April 1983–
Differential	March 1979	March 1983	July 1987
Real differential1	−2.73	−3.84	+0.95
Nominal covered differential2	−1.89	−2.45	+0.80

Note: Average of the differentials with respect to the United States and the Federal Republic of Germany calculated using monthly data on short-term (three months) interest rates.

¹Deflated with the consumption price index.

²Forward discount of the lira (monthly averages).

During the 1980s there has been a reduction in the recourse to administrative constraints. The deregulation began first in the domestic markets and has subsequently been extended to international capital flows.

The ceiling on bank loans was suspended in June 1983. Its abolition was prefaced by radical changes in the conduct of monetary policy, which, beginning with the end of the 1970s, was increasingly directed at raising the amount of public bonds placed on the market. As shown in Table 5, when the ceiling was removed in 1983, the share of credit institutions in total public debt was already declining from a peak of about 40 percent reached in the second half of the 1970s. Following the removal of the ceiling, that share continued to decrease, falling to 24.8 percent at end-1986. The portfolio requirement, which had become progressively less stringent since its introduction, was also abolished in January 1987.

It must be noted that during 1986 and 1987 there was an interruption and a reversal of the deregulation process. On two occasions (January 1986 and September 1987) the authorities were compelled to reintroduce the ceiling for a six-month period. In both cases, the ceiling was used as an emergency measure to counter a worsening of the balance of payments. But there is no doubt that the recourse to the ceiling was also necessitated by the high level of public debt that continues to prevent the authorities from confronting foreign exchange crises with interest rate increases only, since this expedient by itself would have too strong an effect on the interest expenditure of the public sector.

Starting in 1984, foreign exchange policy underwent a gradual deregulation, beginning with the progressive reduction and eventual removal in May 1987 of the compulsory deposit on portfolio investment abroad.¹¹ In September 1986, the Italian Parliament passed a foreign exchange reform law empowering the government to overhaul the present regulation. The bill laid the foundations for a complete recasting of the regulatory framework and overrode the principle underlying the former system, replacing it with the principle that “everything is allowed except what is explicitly forbidden.”

A gradual liberalization has also been undertaken of the terms and conditions of foreign trade financing. However, this process has proven to be quite difficult; in May 1987 some constraints on foreign exchange financing were relaxed, but the authorities were compelled to reintroduce them a few months later.

III. Controls Sustaining Treasury’s Seigniorage

Seigniorage is determined by the supply of the monetary base created by the Treasury’s borrowings from the central bank, but it is also determined by the demand for monetary base by the private sector—that is, the quantity of real balances that the private sector absorbs in its portfolio. Seigniorage is influenced by regulation in two ways: by controls relating to the financing of the Treasury in the monetary base, which act on the relative supply of the Treasury’s base; and by controls relating to banks’ reserve requirements, which influence the demand for monetary base by the banking system. Table 8 summarizes the evolution of these regulations during the 1970s and 1980s.

Table 8.

Controls Sustaining Treasury’s Seigniorage: Main Measures, 1970-87

Table 8.

Controls Sustaining Treasury’s Seigniorage: Main Measures, 1970-87

Control		1970–79	1980–87
Treasury Financing in Monetary Base
	Treasury account with Bank of Italy	Regulation (from 1948) authorized Treasury to draw on its current account up to 14 percent of its planned expenditure.	No changes in regulation.
	Placement of government bonds with Bank of Italy	February 1975. Bank of Italy committed to intervene at Treasury bill auctions, buying bills not placed on the market.	July 1981. Commitment to intervene in Treasury bill auctions suspended.
Reserve Requirements
	Required reserves coefficient	January 1975. Marginal coefficient made uniform at 15 percent for all categories of banks. February 1976. Marginal coefficient increased to 15.75 percent.	March 1981. Marginal coefficient increased to 20 percent. December 1982. Marginal coefficient increased to 25 percent.
	Remuneration on required reserves	October 1970. Rate paid on reserves, which previously moved according to the official discount rate, fixed at 5.5 percent.	December 1982. Rate raised to 9.5 percent for reserves against certificates of deposit only.

View Table

Table 8.

Controls Sustaining Treasury’s Seigniorage: Main Measures, 1970-87

Control		1970–79	1980–87
Treasury Financing in Monetary Base
	Treasury account with Bank of Italy	Regulation (from 1948) authorized Treasury to draw on its current account up to 14 percent of its planned expenditure.	No changes in regulation.
	Placement of government bonds with Bank of Italy	February 1975. Bank of Italy committed to intervene at Treasury bill auctions, buying bills not placed on the market.	July 1981. Commitment to intervene in Treasury bill auctions suspended.
Reserve Requirements
	Required reserves coefficient	January 1975. Marginal coefficient made uniform at 15 percent for all categories of banks. February 1976. Marginal coefficient increased to 15.75 percent.	March 1981. Marginal coefficient increased to 20 percent. December 1982. Marginal coefficient increased to 25 percent.
	Remuneration on required reserves	October 1970. Rate paid on reserves, which previously moved according to the official discount rate, fixed at 5.5 percent.	December 1982. Rate raised to 9.5 percent for reserves against certificates of deposit only.

In Italy, the financing of the Treasury in monetary base takes place in two ways: either through the Treasury’s use of its current account with the Bank of Italy, or by the Treasury’s placement of its securities with the Bank of Italy. The financing of the Treasury in the monetary account allows it to draw on this account up to 14 percent of its planned expenditure at a very low interest cost (1 percent). This regulation has been in force, more or less unchanged, since 1948; in recent years, however, it has been the subject of discussion on several occasions because of the severe limitations it imposes on the autonomy of the Bank of Italy in the creation of the monetary base.¹²

The creation of monetary base through the placement of government securities with the central bank has also been regulated for many years by rules that also greatly restrict the autonomy of the central bank. This was particularly true after February 1975, when the Bank of Italy was committed to intervene in Treasury bill auctions and to buy all the bills not absorbed by the market. The commitment implied substantial purchases of bills by the central bank that were not always easy to neutralize with subsequent open market sales; furthermore, the confidence in central bank support created by this rule induced the Treasury to offer lower rates on its bills than would otherwise have been required by the market. In July 1981 this commitment was suspended by the so-called divorce between the Bank of Italy and the Treasury, and regulations concerning the creation of monetary base since then have tended to increase the autonomy of monetary policy.

It should be noted, however, that the reduction in borrowing from the central bank by the Treasury started well before 1981. As Table 5 shows, such borrowing reached its peak in 1976 (40 percent of total public sector debt) and declined thereafter, falling to 16.5 percent in 1986. This decreased reliance on borrowing from the central bank was achieved by means of heavily increased sales of public sector securities on the market. As has already been noted (see Section I and Tables 2 and 3), notwithstanding the reduction in the share of debt financed with monetary base, the monetary base of the Treasury has declined only slightly in proportion to GDP, and its seigniorage relative to GDP has returned in recent years to its 1976–81 level.

The stock of monetary base in Italy is still much higher than in other countries. Table 9 compares the Italian monetary base as a percentage of GDP with that of other industrial countries at end-1986; from the table it can be seen that the value for Italy is more than twice the average of the other countries. The data shown in the table refer to total monetary base, to which the Treasury contributes a much higher proportion in Italy than in other countries. If the comparisons had been based on the Treasury’s contribution alone, the ratio to GDP would be even higher for Italy.

Table 9.

Monetary Base/GDP Ratio at End-1986: An International Comparison

¹End-1985.

Table 9.

Monetary Base/GDP Ratio at End-1986: An International Comparison

Country	Monetary Base (total)	Bank Reserves	Currency
Italy	15.5	10.1	5.4
United States	5.7	1.3	4.4
Japan1	8.4	1.8	6.6
United Kingdom	4.1	0.5	3.6
Federal Republic of Germany	9.3	3.7	5.6
France	6.5	1.9	4.6

¹End-1985.

View Table

Table 9.

Monetary Base/GDP Ratio at End-1986: An International Comparison

Country	Monetary Base (total)	Bank Reserves	Currency
Italy	15.5	10.1	5.4
United States	5.7	1.3	4.4
Japan1	8.4	1.8	6.6
United Kingdom	4.1	0.5	3.6
Federal Republic of Germany	9.3	3.7	5.6
France	6.5	1.9	4.6

¹End-1985.

Table 9 also shows that the ratio of currency to GDP in Italy is only slightly above the average of the other industrial countries. In contrast, the ratio of bank reserves to GDP is particularly high, relative to the other countries. This difference may be partly explained by the greater degree of banking intermediation in Italy than in the other countries, but this intermediation has been declining rapidly in recent years. Reserve requirements provide a more probable explanation for the larger ratio of bank reserves to GDP. These requirements can be viewed as regulations that have substantially increased the demand for monetary base by the Italian banking system.

As shown in Table 8, during the 1970s and 1980s the reserve coefficient was increased many times. The upward trend began with the reform of January 1975, which imposed a uniform (15 percent) marginal coefficient in place of the previous system whereby the reserve requirements differed appreciably according to the type of deposit and category of bank. The reserve coefficient was progressively increased, reaching, with the measures taken by the authorities in December 1982, its present value of 25 percent.¹³

The influence of the reserve requirements may be seen both on the tax rate and on the tax base (see Appendix for the simple algebra used to calculate seigniorage). The reserve coefficient requires that during each quarter a substantial increase be made in the stock of monetary base held by the banking system; in this respect, the coefficient is analogous to a tax rate and its effect is captured mainly by changes in velocity. The rise in the reserve coefficient in recent years has led to a decrease in velocity, which has been stronger than that shown in Table 3, if one considers that the decrease in velocity caused by the reserve regulation has been partly offset by an increase in velocity in the portion of monetary base held in the form of free bank reserves and currency.¹⁴ By requiring substantial additions to reserves in each period, the regulation also had the effect of preventing a reduction of the tax base over time, which would have been prompted by the decline in the share of bank deposits in GDP.

To assess the implicit fiscality of the reserve requirements, one must also consider the provisions concerning the remuneration paid by the central bank on required reserves. Until 1970, the rate of remuneration moved according to the official discount rate. Thereafter, except for the portion of reserves pertaining to certificates of deposit, which was made subject to a rate of 9.5 percent at end-1982, the remuneration has remained fixed at 5.5 percent. The higher rate on certificates of deposit has so far made little difference to the average return on reserves because of their limited share in total reserves.

The fixed rate of 5.5 percent paid on reserves substantially increased the burden of the reserve requirements on the banking system during a period in which interest rates moved to levels much higher than those prevailing at the beginning of the 1970s.¹⁵ The fixed rate also allowed the central bank to pay back to the Treasury the interest (as perceived by the central bank) on the securities it was holding in its portfolio.

IV. Tax from Controls and Inflation Tax

In the previous sections, we have argued that administrative controls on domestic financial flows and international capital movements can help finance the public deficit in two ways: by lowering the real cost of interest-bearing public debt independently of the prevailing rate of inflation; and by raising the real demand for monetary base, which allows the collection of seigniorage. We will now briefly discuss the relation between these effects and the well-known operation of the “inflation tax.”

One connection is that higher inflation can increase the taxing power of a given set of controls. In more precise terms, a given set of direct controls on nominal interest rates becomes more binding when inflation increases. With higher inflation, controls on nominal rates have more pervasive distortionary consequences on credit and financial markets, but since they usually constrain the remuneration of bank deposits, a larger amount of seigniorage is also indirectly collected by the public sector.¹⁶

This type of connection between higher inflation and the effect of controls was widely recognized as obtaining in the United States during the 1970s.¹⁷ And it has often been stated that strong inflation stimulates financial deregulation by magnifying the disturbing consequences of fixed nominal rates.¹⁸ In the Italian case, however, explicit controls on nominal rates have never played an important role; consequently, the crucial connection between inflation and controls is of a different kind.¹⁹

In fact, there is a second type of relation, which moves in the opposite direction: namely, the imposition of controls can increase the amount of the implicit tax collected through a given rate of inflation. This is obvious when controls, such as compulsory reserve requirements, increase the real demand for high-powered money, because that money is also the base for the inflation tax. The effect on the bond market is similar.

With no controls, only unexpected inflation can levy a tax on bondholders. The anticipated component of inflation, via the Fisher effect,²⁰ will be incorporated in the nominal yield and will leave real returns unchanged. We argue that administrative controls can interfere with the working of the Fisher effect, by decreasing the substitutability among the various assets in the portfolio of the private sector. In particular, if it is relatively more difficult to substitute away from government bonds, the Treasury will collect a tax from expected inflation and from an expected acceleration in inflation.²¹ If purchasing power parity holds, for instance, exchange controls can decrease the availability of assets hedged against inflation, and it will be easier for anticipated domestic price increases to transfer wealth from creditors to debtors. If internal controls favor the holding of government debt, the public sector will be a privileged collector of that transfer. It is worth noting that this taxing mechanism (which more or less works in all countries) can rely upon a large variety of controls and on all the regulations and imperfections that hinder portfolio adjustments and asset or liability substitution. Controls that are not used for stabilization purposes—for instance, limits to entry in the banking industry, or protection of banks from the competition of nonbanks in the collection of savings or obstacles to securitization—these can work to lower the real cost of debt as expected inflation increases, and vice versa.²²

From this point of view, the introduction of real-indexed bonds works in the opposite direction (provided they are not forced into the system at lower-than-equilibrium real rates).²³ Not only do these bonds increase the supply of assets that are protected from inflation surprises, but they also prompt substitution away from bonds when inflation is expected, thus decreasing the collection of implicit taxes via anticipated price increase.²⁴

During the last two to three years, the ex-post average real interest rate on short-term government bonds (see Table 1) has been nearly 10 percentage points above its average level at the end of the 1970s. Such a large increase cannot be explained entirely by a decrease in both unanticipated inflation and administrative controls; nor can it be the effect of other often cited causes. Some liberalization has taken place, and this has allowed an upward movement of real rates, which have also been pushed up by the prevailing international fiscal-monetary policy mix; inflation risk-premium and some expected real devaluation of the lira are other probable causes. However, the increase in the real cost of public debt still appears too high. Therefore, the rise in interest rates must also be a result of the substantial inverse correlation between expected inflation and ex-ante real rates of interest that is generated by regulations and imperfections in the financial system. Even at the present, reduced level of administrative controls, an anticipated increase in inflation would exert some downward pressure on real rates of interest.

Inflation and administrative financial controls can thus be said to reinforce each other in lowering the real cost of public debt and in collecting seigniorage for the Treasury. To some extent, they can be substitutable sources of implicit taxes for a highly indebted government. If the adjustment of the Italian basic deficit is insufficient and inflation is kept low, it is highly probable that financial deregulation will be interrupted and/or new controls will be imposed. (Are the developments of fall 1987 a start in this direction?) In contrast, further liberalization could necessitate higher inflation.²⁵

But the substitution between the two implicit taxes cannot be pushed too far. An overdose of controls, besides giving rise to tremendous distortions, becomes ineffective as a consequence of all sorts of regulation-avoiding innovations. An overdose of inflation decreases the real demand for money, thus eroding the base of its taxing power and moving to the “unpleasant side of the seigniorage Laffer curve.”²⁶

V. Implicit Taxes, Capital Controls, and the Equity Market: The Issue of Crowding-Out

As documented in previous sections, the Italian Government has relied on administrative controls to collect implicit taxes. Although the money, credit, and foreign exchange markets have borne the brunt of the distorting effects of the controls, all financial markets have been affected by them. In this section, we focus on the indirect effects of across-the-board capital controls on the stock market and, consequently, on the cost of equity.

Until May 1987, Italian portfolio investments abroad were severely restricted. A liberalization process is now taking place, but many restrictions still remain in effect. By limiting the menu of assets available for domestic savers with capital controls, the authorities have reduced the interest rate on government debt because they have artificially increased the demand for it. Controls thus imply a subsidy for the government. But have other borrowers benefited from the same subsidy? The answer may be negative, because the controls may have raised the cost of equity—thus crowding out investment projects—even though the “risk-free” rate—the interest rate paid by the government on its debt—has been maintained below its equilibrium level.

The cost of equity for a company depends on the risk premium that investors demand in order to hold its stock. Stocks are viewed as riskier than fixed-income government debt because—apart from default risk—their prices are more volatile than bond prices, so that their returns are more uncertain. Consequently, investors must expect to receive a higher return from the share components of their portfolio to be compensated for the higher risk.

Stock prices tend to move together in the various capital markets. This empirical regularity has led to the conjecture that there is a common source of share price fluctuations, which is the risky factor. Because risk-averse investors dislike risk, demand will be high for those stocks that have a low correlation with all other stocks—and therefore, low correlation with the risky factor—whereas demand will be low for those stocks highly correlated with the rest of the market. In equilibrium, the first group of stocks will clearly demand a lower risk premium than the second.

When investors in a country are prohibited from investing abroad, they cannot reduce the variability of their portfolios by adding foreign stocks. Their portfolios will then be composed entirely of domestic stocks—and will necessarily be riskier than an internationally diversified portfolio. They will then demand a higher risk premium for holding the domestic stocks than they would have in a regime with free capital movements.

By segmenting the domestic capital market, a government may reduce the cost of funding its own debt but, at the same time, may increase the riskiness of domestic equities for domestic investors, thereby raising the cost of funding capital expansions for the corporate sector—in other words, the crowding-out effect. An interesting empirical question is whether the long history of capital controls has effectively raised the cost of equity financing for the Italian corporate sector. To examine this issue, we adopt a few simple international capital asset-pricing models in order to assess whether a “super-risk” premium characterized the Italian market during the last eight years.

In principle, if foreign investors were free to invest in the domestic market—and to repatriate profits—there would be no crowding-out because they would immediately reap the super-risk premium that would emerge in that market.²⁷ A premium found in the Italian data would be a “political-risk” premium—that is, foreigners are reluctant to invest in Italy because of institutional uncertainties, market illiquidity, and lack of information.²⁸

We begin our empirical investigation with the establishment of the European Monetary System (EMS). This monetary union amounted to a switch in regime that added credibility to Italian economic policies and presumably reduced foreign investors’ risk-perception of the country. Because we are testing for market segmentation, we have excluded the pre-EMS years from the sample to avoid spurious results due to the lack of homogeneous periods. We have ended the sample period in June 1987, when portfolio investments abroad were liberalized.

Any implementation of a capital asset-pricing model requires a measure of the risky factor for equity prices. A well-known approach uses the return on the “market portfolio” as the single measure of risk. In a world of perfect capital mobility, the world equity market is the appropriate market portfolio—a weighted index of the various national stock markets. In our study we consider ten countries—the United States, the United Kingdom, France, Canada, Sweden, Switzerland, Germany, Japan, the Netherlands, and Italy—which together account for 98 percent of the capitalization of world stock markets.

For each market we present both the average monthly real rate of return and the real return in excess to the risk-free rate, which was approximated by the rate on three-month government debt in every country. We computed real returns by taking the monthly percentage changes of the various stock indexes deflated by domestic consumption price indexes (we did not take into account reinvestment of dividends). Different indexes were needed because consumption baskets differ across countries and because purchasing power parity does not hold. Figure 1 shows that the Italian stock market provided the highest monthly real rate of return among the industrial countries with a rate of 1.5 per-cent, and was second only to Japan in terms of real excess return. The picture is even more clear-cut if we consider the last three years when the Italian stock market developed very rapidly following the introduction of mutual funds.

View Full Size

Real Rates of Return on Stock Indexes, January 1979–June 1987

(In percent)

• Download Figure
• Download figure as PowerPoint slide

To compare returns among different countries, we have to adjust returns for their risk. As explained above, we initially used the world market portfolio as the source of risk and measured riskiness by regressing the real returns of the various stock indexes on the real return of the world market portfolio. The cross section of the average returns in the different stock markets was then regressed on the estimated betas to obtain what is known as the security market line. This line gives the empirical equilibrium trade-off between real return and systematic risk in the international capital market. If a country is above the line, an international investor could have increased the return on his portfolio, without increasing risk, by investing more in that country. In Figure 2, we show the security market line for the period 1984–88; the same picture would emerge if we looked at the full sample period, which is not shown here.

View Full Size

Portfolio Model: Risk-Return Relation, January 1984–June 1987

(In percent)

• Download Figure
• Download figure as PowerPoint slide

Italy and Japan appear to have generated returns that were way above those prevailing in the other markets, even when they are adjusted for systematic risk. For example, an investor would have greatly enhanced the return on his portfolio per unit of risk by moving out of the U. S., U. K., and Dutch markets and investing in the Italian market.

The world market portfolio is not the only way to measure the source of risk. It has been shown that this measure is correct only for the truly “international investor,” whose consumption preferences are not biased toward goods of one country and therefore does not want to hedge against the losses of purchasing power of the currency in which the price of these goods is expressed.29 An alternative way to measure risk is to note that risk-averse individuals accumulate wealth to smooth consumption, even if their incomes fluctuate widely. To those investors, assets that have high returns when consumption is low are more valuable than assets with rates of return highly correlated with consumption. Risk can then be measured by consumption betas—namely, the empirical correlation between real asset returns and real consumption changes. The relevant consumption concept for the measurement of risk is world real consumption, because in equilibrium—and with perfect capital mobility—this is the only risk that is systematic to all stocks in the world and consequently cannot be diversified away by trading with foreigners.³⁰

In practice, we measured world real consumption by creating a gross national product (GNP)-weighted index of consumption at constant prices for eight countries with quarterly data—the same cross section of countries used before, with the exclusion of France and Switzerland for which data were unavailable. We used the same sample period 1979–87. The consumption betas and the average total real returns were then used to calculate the security market line, which is shown in Figure 3. The results are similar to those found with the market portfolio model. The rate of return per unit of risk in the Italian equity market exceeded the international norm as measured by the security market line.

View Full Size

Consumption Beta Model: Risk-Return Relation, March 1979–June 1987

(In percent)

• Download Figure
• Download figure as PowerPoint slide

No firm conclusion can be made based on the empirical evidence presented in the paper because of the limited sample size and the inevitable shortcomings of the pricing models used. However, one can point to the high real return per unit of risk, relative to the rest of the world, which Italian corporations appear to have paid to investors. One possible explanation for this phenomenon—although not the only one—is that domestic investors, not being allowed to diversify internationally, demanded an extra-risk premium to hold the existing stock of Italian equities. If this is the case, the liberalization process now undertaken by the government should gradually reduce the cost of equity financing and, therefore, bring the expected rates of return in the Italian stock market more in line with international markets.

VI. Concluding Remarks

As shown at the beginning of the paper, the growth of the real stock of Italian interest-bearing public debt has accelerated in recent years, in spite of some reduction in the basic deficit and notwithstanding a sustained collection of seigniorage by the Treasury. We have argued that inflation and financial controls are two sources of implicit taxes that reinforce each other. When inflation was high, even if anticipated, it could not be entirely incorporated in nominal interest rates because of controls and imperfections in capital markets. On the contrary, the decrease in inflation has been among the main causes of the higher levels of real rates.

When implicit taxes decrease the cost of public debt, they do not necessarily also lower the cost of capital for private firms. To be sure, controls on the allocation of credit and capital flows can be organized to obtain the opposite result,³¹ by placing part of the burden of implicit taxation on private debtors (thus relieving the burden on savers) and obtaining a crowding-out of private spending in spite of the lower rates paid by the Treasury. We have studied a particular effect of this kind—that is, the positive impact of foreign exchange controls on the risk premium built into the cost of equity capital. The Italian stock exchange commands a higher premium than that of other countries’ capital markets, which could be lowered by liberalization, thus compensating for a higher risk-free rate and limiting the possible crowding-out effect of deregulation.

But the problem remains. Can the financial situation of the Italian sector stand the real interest rates that are determined on free financial markets? Or will the authorities be compelled to choose a mix of stronger controls and higher inflation? We are convinced that this choice can still be avoided, and that fiscal adjustment can take place through a sufficient reduction of the basic deficit. To support this conviction, we make three final points, which have often been raised in the research conducted and proposals made by the Centre for Monetary Research and Financial Economics of Università Bocconi.

First, the qualitative composition of public debt must be carefully engineered. An efficient diversification of the various debt instruments can help reduce the Treasury’s interest expenses. In this respect, we think that in a period when inflation is low but inflation expectations are uncertain, issuing a substantial amount of real-indexed bonds could decrease the risk premium on real interest rates.³²

Second, liberalization must be balanced between domestic and international flows. If it proceeds too rapidly or too slowly on one front, the steps on the other front can become more difficult and more costly and less effective. Even with financial deregulation, there is an order-of-liberalization problem, not unlike the one encountered in less-developed countries when the removal of real trade and financial protectionism needs to be measured out in appropriate doses.³³

Finally, we think that political decisions that determine the basic deficit, net of interest expenses, are not independent from financial discipline nor from the level of real interest rates, and cannot be considered exogenous when the issue of deregulation is being confronted. Stronger administrative controls can certainly decrease the real interest component of public debt growth and/or sustain seigniorage; but they will probably also remove some of the pressure on the authorities to adjust noninterest expenses and explicit taxes. There are signs that some “disciplinary effect” of financial liberalization is already influencing

Italian fiscal policy. From this point of view, the international financial community and institutions perform a valuable role when they push for liberalization of international capital movements, thus raising the political cost of financial protectionism, stimulating the country’s financial discipline, and indirectly promoting fiscal adjustment.³⁴

Comment

Comment

Jorge Braga de Macedo

Three authors for an economics paper may seem a large number, but Franco Bruni, Alessandro Penati, and Angelo Porta suggest that, in their case, it is a small number. Because they are bringing together research in the areas of debt management, banking, and international finance conducted at the Centre for Monetary and Financial Economics of Università Commerciale Luigi Bocconi (UCLB), this work represents the views of yet more authors on Italy’s financial integration and monetary and financial policy. Claiming that economic survival with a high public debt will involve a mix of higher implicit taxes, stronger administrative controls, and faster inflation, Bruni, Penati, and Porta argue for fiscal adjustment by pushing for the liberalization of international capital movements. Borrowing from the title of a paperback about another successful business school, I will refer to these proposals as “the gospel according to UCLB.”

In this comment, I will briefly characterize the accounting framework used by the authors, and then describe the various solutions to the Italian public debt problem and their links to the literature. I will then show that the framework is useful for analyzing other public debt problems and, drawing on Macedo and Sebastiao (1989), refer to the recent experience of Portugal. My conclusion is that the UCLB gospel travels well, to the extent that it requires a careful analysis of the specific conditions of each high public debt country.

The analysis of Bruni, Penati, and Porta begins with a standard decomposition of the debt-accumulation equation, whereby the increase in privately held public debt equals the primary deficit plus the interest account minus seigniorage. This framework is used to uncover the various implicit taxes implied by the structure of the public sector. Seeing controls as sources of revenue makes it easier to understand the difficulties of fiscal adjustment and the dangers of financial deregulation in Italy. The collaboration with the International Monetary Fund (IMF) is thus welcome; the recent Occasional Paper No. 55 (IMF (1987)) still touches all too briefly on the link between financial programming and fiscal policy.

To understand the link, it is necessary to identify the various ways in which the public sector deals with financial intermediaries, a major undertaking begun by Bruni, Monti, and Porta (1982). A shortcut is to focus on the borrowing of the Treasury from the central bank, analyzed in great detail by Salvemini and Salvemini (1987). Here, the UCLB gospel asserts the rights of the lender and turns the subordination of the central bank to the Treasury on its head. The refusal to buy more public debt, rather than being “an act of rebellion” (as the 1973 Report of the Bank of Italy calls this possibility), may then be in accordance with the practices of the international financial community and institutions.

The 1981 “divorce” between the Bank and the Treasury was an incomplete and easily reversible liberating experience. There are still hidden taxes and controls that delay fiscal adjustment, and the change in operating procedures has not been ratified by a legislative act of Parliament, as emphasized by Tabellini (1988). The opening of the Italian capital market that is required by the objective of a single European market in 1992, by imposing financial discipline, is the only credible measure to end the direct financing of the Treasury by the central bank. The current system is unsustainable, and the alternatives, according to the gospel, involve a closing of the Italian market, perhaps sharing McKinnon-type “financial repression” with other soft currencies in the European Economic Community.

Because of the emphasis on the role of the central bank, the UCLB gospel departs from the fully consolidated view of the public sector. In such a view, shown for the closed economy in Buiter (1985), the change in the monetary base would be treated like any other source of government revenue. Instead, Bruni, Penati, and Porta prefer to focus on the public debt held by the central bank, which they call the Treasury monetary base. Other central bank assets, especially claims on foreign monetary authorities, may change the profit and loss account, but they are considered as lowering the deficit rather than financing it. The same may be said, incidentally, of foreign-held public debt, which turns out to be low in Italy but is of course high in many Latin countries involved in financial programming exercises with the IMF.

This accounting framework shows that the Italian Treasury has adjusted to high real interest rates by maintaining a sizable source of revenue from the central bank, not so much in the form of currency but rather bank reserves and controls, which increase the demand for public debt on the part of the banking system. Concentrating on the public debt held by the central bank at below market interest rates, the authors record roughly 2 percent of gross domestic product (GDP) from this type of seigniorage from 1976 to 1987—about the same as the total seigniorage reported by Giavazzi (1989). More interesting perhaps, Treasury seigniorage remained constant, while real interest rates net of growth rose from -7 percent in 1976–81 to 2 percent in 1984–87, and the primary deficit fell from 7 percent to 5 percent. This decline was due to an increase in the tax rate from 11 percent to 13 percent, since the base actually fell from 19 percent to 16 percent.

Aside from the Treasury seigniorage and the inflation tax, administrative controls help to finance the deficit because they enhance the demand for public debt relative to private domestic and foreign instruments. This is evident in a positive real interest rate differential relative to the United States and the Federal Republic of Germany, which was close to 1 percent in the period 1983–87. Actually, the higher cost of equity more than compensates for the lower real cost of issuing public debt, thus inducing international investors to require a premium to hold stock in Italian firms. A higher degree of capital mobility, by breaking this financial protectionism, may thus require a lower public debt to be credible. The trade-off is, once again, between a closed system with more inflation and controls and an open system with less government spending and less “disguised fiscal policy.”

Three closing points made by the authors illustrate the links of the UCLB gospel to other strands of the literature. First, the structure of the public sector matters, and, in particular, more attention should be paid to the qualitative composition of public debt. This point is indisputable and has a broad significance. Once again, in the Italian context it has often been seen in terms of marriages and divorces between Banca and Tesoro, but the broad implication is for the careful study of the variety of roles the public sector takes with respect to banks, as suggested in Bruni, Monti, and Porta (1982).

This particular aspect of the gospel is also documented by Tabellini (1988) and other contributors to Giavazzi and Spaventa (1988). Alesina (1988) and Salvemini and Salvemini (1987), moreover, show how this approach touches upon recent advances in the strategic analysis of government behavior and raises the issue of the independence of the central bank. The specific implication of this point, brought out in this paper, is to have more real-indexed bonds. Nevertheless, without a regime change, more Baffi bonds may induce more spending, as Morcaldo warns in his comment (see p. 246). In effect, the independence of the central bank cannot be advocated in a vacuum, it must be seen in the framework of the monetary constitution, so dear to the members of the public-choice school (see Brennan and Buchanan (1981)).

A second tenet of the UCLB gospel—that domestic and external financial liberalization must be balanced—touches upon an entirely different strand of literature, which has a flavor of Latin America rather than of Lombardy. The insight, both theoretical and empirical, about the order in which real trade and financial protectionism must be removed is of relevance here. Sometimes shown by the so-called Arriazu box, the preferred order seems to be real first and financial second (see Edwards (1984)). If domestic also precedes external, we see that the removal of financial protectionism in Italy requires domestic financial liberalization—namely, the end of the implicit taxes and administrative controls designed to favor borrowing by the Treasury. Skepticism about the ability to bring about an irreversible monetary reform without external pressure is another feature of the gospel. In light of the establishment of the single European market in 1992, however, one wonders whether expected external financial liberalization could validate a change in the domestic financial regime.

The importance of uncovering implicit taxes and disguised fiscal policy is that, once again, it shows the political element behind the financial discipline. The idea that external financial liberalization may be a way of neutralizing the domestic lobbying efforts for deficit financing is once again reminiscent of some of the development literature focusing on rent-seeking. It is also found, of course, in the game-theoretic approach of Alesina (1988) and has been used by economic historians.

We could even rationalize the super-risk premium on Italian equity discussed in the last section of the paper as a kind of “lira problem,” familiar to students in Chicago of the forward market for Mexican pesos and relate it to the degree of capital mobility. The approach of Frankel (1988) or Frankel and MacArthur (1988) is to take the real-interest differentials as measures of capital mobility. The figures in the paper suggest that both the real and nominal interest rate differential between the lira and the dollar turned from negative to positive after the 1983 realignment. We can see this change in the covered differential, which is a measure of political risk, as a move from restrictions on capital outflows to restrictions on capital inflows. The decrease in the magnitude of the spread, however, suggests that there is some expectation of liberalization but that the lira problem can be seen as a risk of future controls on outflows. Indeed, the most recent data suggest that the domestic return may have become lower once again. This approach is of course consistent with the finance literature summarized by Adler and Dumas (1983), which Bruni, Penati, and Porta use in their analysis.

Rather than pursuing this route, I will show how the gospel can be useful in an explanation of the Portuguese experience. Figure 1 shows the Portuguese public debt as a proportion of GDP from 1970 to 1987, as well as its decomposition between debt held privately—that is to say, mostly by commercial banks—and debt held by the central bank. The latter component began to rise after the 1974 Revolution, in the wake of which the domestic banking system was nationalized without compensation. There was a drop in 1980, offsetting the capital gain due to the revaluation of the gold reserves. The accumulation of both components of public debt accelerated subsequently, but after the stabilization program agreed on with the IMF in 1983–84, the central bank component began to decline. The external surplus of 1986/87 was a factor in the decline of the Treasury monetary base, but there is also increasing awareness that the “high public debt” (70 percent of GDP as seen in Chart 1, plus all the guaranteed debt) may reverse the anti-inflationary program initiated in late 1985, which managed to cut the rate of increase of the output deflator from 20 percent in 1985 to 11 percent in 1987, thus bringing ex-post real interest rates up from -11 percent to -1 percent. Meanwhile, the primary government budget deficit fell from 4 percent to 2 percent of output over the same period.

View Full Size

Portugal: Public Debt, 1970–85

(In percent of GDP)

Source: Data from Macedo and Sebastiao (1989)

• Download Figure
• Download figure as PowerPoint slide

Reproducing the decomposition used by Bruni, Penati, and Porta for Portugal requires some adaptations, mostly because the official budget figures (even “adjusted” like the ones on the primary deficit just cited) cannot be used in the debt-accumulation equation, due to accounting deficiencies. An “implied deficit” obtained as a residual is thus presented after the real interest payments and seigniorage are accounted for. As explained in Macedo and Sebastiao (1989), it matters whether we use total or domestic debt, and whether the interest payments of the Treasury to the central bank are net of central bank payments to the commercial banks due to the interest subsidies and the interbank market. With total debt and the gross interest figures, the implied nominal rate on privately held debt rises from 9 percent to 10 percent between 1985 and 1987, whereas the one paid on the public debt held by the central bank falls from 19 percent to 15 percent, implying that the contribution of seigniorage as a proportion of output fell from 3 percent in 1985 to -17 percent in 1987. The implied deficit in turn, fell from 38 percent to 13 percent of output, the difference from the actual figures being due to external borrowing and adjustments.

These figures underestimate seigniorage however, since the interest revenue of the central bank is spent compensating the banking system for the low profitability of its operations due to the system of credit ceilings. Taking these into account, the implied effective interest falls from 2 percent to -3 percent, and the contribution of seigniorage as a proportion of output falls from 14 percent to zero, whereas the implied deficit falls from 50 percent to 20 percent of output.

Tables 1-3, using the same format as Bruni, Penati, and Porta, are based on total debt and the net interest payments. The period averages since 1976 have a rough correspondence with governments, including the odd year of 1980 (see details in Macedo and Sebastiao (1989)). Overall, the tables show a situation that appears more distorted than the situation in Italy before the 1981 divorce, with highly negative real interest rates, high primary deficits, and high seigniorage. Despite the fall in the seigniorage tax rate from 28 percent in 1981/84 to 8 percent in 1985/87, the base remains high and is seen to be rising from 22 percent to 25 percent. This has kept total seigniorage at 2 percent of output. Although this is the same share as in Italy, the difference between the two situations is what the UCLB gospel—and these tables—bring out.

Table 1.

Portugal: Breakdown of the Growth Rate of the Real Value of Public Debt Held Outside the Central Bank

(Average yearly rates in percent)

Source: Data from Macedo and Sebastiao (1989).

Table 1.

Portugal: Breakdown of the Growth Rate of the Real Value of Public Debt Held Outside the Central Bank

(Average yearly rates in percent)

		Contribution of:
	Growth Rate		Deficit Net of		Real Interest		Treasury’s
Years	of Real Debt	=	Interest	+	Rate on Debt	-	Seigniorage
1976–79	3.0		64.0		−11.0		50.0
1980	31.0		0.0		−7.0		−36.0
1981–84	12.0		41.0		−8.0		22.0
1985–87	19.0		31.0		−6.0		6.0

Source: Data from Macedo and Sebastiao (1989).

View Table

Table 1.

Portugal: Breakdown of the Growth Rate of the Real Value of Public Debt Held Outside the Central Bank

(Average yearly rates in percent)

		Contribution of:
	Growth Rate		Deficit Net of		Real Interest		Treasury’s
Years	of Real Debt	=	Interest	+	Rate on Debt	-	Seigniorage
1976–79	3.0		64.0		−11.0		50.0
1980	31.0		0.0		−7.0		−36.0
1981–84	12.0		41.0		−8.0		22.0
1985–87	19.0		31.0		−6.0		6.0

Source: Data from Macedo and Sebastiao (1989).

Table 2.

Portugal: Breakdown of Change in Ratio of Nonmonetary Public Debt to GDP

Source: Data from Macedo and Sebastiao (1989).

Table 2.

Portugal: Breakdown of Change in Ratio of Nonmonetary Public Debt to GDP

			Contribution of:
	Average Yearly		Real Interest Rate
	Change in GDP		on Debt Less Real		Debt/GDP		Deficit Net of		Treasury’s
Years	Ratio		Growth of GDP	×	Ratio	+	Interest	-	Seigniorage
	(in percent)						(in percent of GDP)
1976–79	0.0		−16.0		16.0		10.0		8.0
1980	5.0		−11.0		19.0		0.0		−7.0
1981–84	3.0		−8.0		29.0		12.0		6.0
1985–87	6.0		−10.0		43.0		13.0		2.0

Source: Data from Macedo and Sebastiao (1989).

View Table

Table 2.

Portugal: Breakdown of Change in Ratio of Nonmonetary Public Debt to GDP

			Contribution of:
	Average Yearly		Real Interest Rate
	Change in GDP		on Debt Less Real		Debt/GDP		Deficit Net of		Treasury’s
Years	Ratio		Growth of GDP	×	Ratio	+	Interest	-	Seigniorage
	(in percent)						(in percent of GDP)
1976–79	0.0		−16.0		16.0		10.0		8.0
1980	5.0		−11.0		19.0		0.0		−7.0
1981–84	3.0		−8.0		29.0		12.0		6.0
1985–87	6.0		−10.0		43.0		13.0		2.0

Source: Data from Macedo and Sebastiao (1989).

Table 3.

Breakdown of Portuguese Treasury’s Seigniorage, 1976-87

(Average yearly rates, in percent except where otherwise noted)

Note: Data from Macedo and Sebastiao (1989).

^*Public debt held by the private sector.

Table 3.

Breakdown of Portuguese Treasury’s Seigniorage, 1976-87

(Average yearly rates, in percent except where otherwise noted)

	Seigniorage Tax Rate
		Change in Real Stock			Change in	Avg. Interest		Seigniorage Tax Base
		of Total Monetary Base			Treasury Base/	Rate on		(monetary base of		Total Seignoirage
		Real GDP		Total	Monetary	Treasury	Total	the Treasury)		(7)x(8)	(7)x(9)
	Inflation	Growth	Velocity	(2)–(3)	Base Ratio	Monetary Base	(1) + (4) + (5)-(6)	(percent of debt*)	(percent of GDP)	(percent of debt*)	(percent of GDP)
Year	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
1976–79	19.0	5.0	12.0	−7.0	38.0	5.0	44.0	12.0	19.0	49.0	8.0
1980	18.0	4.0	2.0	2.0	−52.0	1.0	−33.0	10.8	20.0	36.0	−7.0
1981–84	20.0	0.0	3.0	−3.0	18.0	7.0	28.0	7.6	22.0	21.0	6.0
1985–87	16.0	4.4	9.0	−5.0	−4.0	0.0	8.0	5.8	25.0	6.0	2.0

Note: Data from Macedo and Sebastiao (1989).

^*Public debt held by the private sector.

View Table

Table 3.

Breakdown of Portuguese Treasury’s Seigniorage, 1976-87

(Average yearly rates, in percent except where otherwise noted)

	Seigniorage Tax Rate
		Change in Real Stock			Change in	Avg. Interest		Seigniorage Tax Base
		of Total Monetary Base			Treasury Base/	Rate on		(monetary base of		Total Seignoirage
		Real GDP		Total	Monetary	Treasury	Total	the Treasury)		(7)x(8)	(7)x(9)
	Inflation	Growth	Velocity	(2)–(3)	Base Ratio	Monetary Base	(1) + (4) + (5)-(6)	(percent of debt*)	(percent of GDP)	(percent of debt*)	(percent of GDP)
Year	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)
1976–79	19.0	5.0	12.0	−7.0	38.0	5.0	44.0	12.0	19.0	49.0	8.0
1980	18.0	4.0	2.0	2.0	−52.0	1.0	−33.0	10.8	20.0	36.0	−7.0
1981–84	20.0	0.0	3.0	−3.0	18.0	7.0	28.0	7.6	22.0	21.0	6.0
1985–87	16.0	4.4	9.0	−5.0	−4.0	0.0	8.0	5.8	25.0	6.0	2.0

Note: Data from Macedo and Sebastiao (1989).

^*Public debt held by the private sector.

Comment

Panel Discussion

John Vanderveken

Mr. Chairman, fellow participants, let me start by thanking the International Monetary Fund (IMF) and the Università Bocconi for inviting me to join this panel for a discussion of fiscal policy adjustment and economic agents. I am the General Secretary of the International Confederation of Free Trade Unions (ICFTU). For those of you unfamiliar with our organization, it is a worldwide confederation of trade unions. We have 143 affiliates in 97 countries, with a total membership of 87 million unionists. Our affiliates are national centers, such as the American Federation of Labor-Congress of Industrial Organizations (AFL-CIO) in the United States; the Confederation General del Trabajo (CGT) in Argentina; and the Confederazione Italiana Sindacati Lavoratori (CISL) and the Unione Italiana del Lavoro (UIL) in Italy. Our role on their behalf is to develop policies on a wide range of issues of common concern; the protection and promotion of trade union rights and economic and social questions are two of the most important areas of our work. In recent years the issue of economic adjustment in both developing and industrial countries has been one of our major concerns.

As I understand the concept of adjustment, it is the process of adapting the economy to produce competitively a different range of goods and services. That easy phrase describes a difficult and challenging task affecting all workers. Some people may find it novel that trade unions wish to have a say in structural adjustment. However, the process is of such importance to the creation of jobs and the improvement of living standards that unions cannot simply sit back and criticize the failures of others. Furthermore, one of the key elements of a successful adjustment or transformation is, in my view, the cooperation and active involvement of workers in the process. Commitment, participation, and cooperation do not come automatically at the push of a button. They must be built up through discussion, exchange of information, and agreement both in the workplace and at the national and international policymaking level. My contribution to this morning’s discussion will be to argue that for a variety of economic, social, and political reasons, a great deal more needs to be done to ensure the involvement of trade unions and, by inference, other major groups, such as employers, in the development and application of adjustment policies and, in particular, fiscal policies.

As trade unionists, we have as a priority the protection and enhancement of the conditions of work and life of our members. More jobs, better wages and working conditions, pensions, and holidays are what members expect from their unions. Of course, behind these daily preoccupations are wider and deeper issues—dignity, justice, freedom—which decades of struggle have proven can only be gained and preserved by working people through trade union organizations. Most politicians and bankers—but not all—would admit that trade unions have a legitimate right to talk about wages and conditions. However, they are less willing to admit that we must also express our views about the running of the national and, indeed, world economy.

For many years now the ICFTU, the Trade Union Advisory Committee to the Organization for Economic Cooperation and Development (OECD), and our national affiliates have presented policy statements to governments and the international agencies. I think we have gained some respect for the seriousness of our approach, but I have to say that our members and working people generally do not see much return for our efforts.

Most developing country workers are no better placed now than they were six years ago. Many, in fact, have lost their jobs or found that inflation has cut into their wage packets. Poverty and hunger in both city and countryside are rising in most nations. Indicators of health problems among children are rising. The much-heralded recovery has just not arrived for millions of ordinary hard-working people. In industrial countries many groups of workers have not benefited from the slow growth since 1983, and many have experienced periods of unemployment devastating to their lives and savings.

Whenever I read or hear the economists or bankers say the worst of the world economic crisis is past, I point out that for the majority of working people in the developing world, the crisis continues and, for many, worsens. In the industrial countries there is considerable concern about job security in the private sector and wage squeezes in the public sector.

Is there a light at the end of the tunnel? Will these sacrifices result in improved economic performance, more jobs, and rising living standards next year, or the year after? I fear that my reading of the economic forecasts of the major international organizations leads me to say no. Each new forecast merely seems to explain why the last was overoptimistic, but that things should improve soon. But I do not see it. The United States and Japan are not growing at a sufficient pace to generate a worldwide fall in unemployment. A European recovery seems far away, and the greater part of the developing world is locked into tough programs of adjustment.

It is our strong view that economic policies must respond to the needs of the working people if they are to stand a chance of success. Free trade unions are ready to join with governments and industry, at both the national and international levels, to help define and implement the immediate and long-term objectives and policies needed to sustain, spread, and strengthen the process of economic revival and to overcome the urgent problems of unemployment and poverty.

Our conviction is that increased coordination of policies could yield benefits for all and would certainly help to reduce the risks of a pronounced downturn. We are looking for relatively modest first steps, but steps that would send a message that the major democratic economic powers are giving some degree of direction to a confused and worried world.

We believe that one important and necessary way to reduce uncertainty and inconsistent expectations is improved coordination nationally and internationally among policymakers, among whom I include ourselves. The first area is exchange rates and interest rates, which, of necessity, will also lead to some degree of fiscal coordination.

A reform of tax systems to spread the base of revenue raising, to improve the efficiency of collection, and to ensure a progressive sharing of the burden of government expenditure is important in most countries. Decisions about public expenditure, especially on issues such as health, education, housing, social security, and other essential aspects of the social and economic infrastructure, can help to reduce unemployment and provide efficient services. And as trade unionists, we accept that efficiency in public expenditure is as important as its level and rate of growth.

Mr. Chairman, trade unions are prepared to play their part in the promotion and implementation of national and international development policies that meet the difficult challenges of restoring growth, correcting trade imbalances, increasing employment, and tackling poverty. Dialogue between governments, industry, and unions is essential to increase economic efficiency and social justice. Fiscal policy is probably the most important area to tackle because of its linkages to the international scene through financial markets. The trade union movement is eager to participate fully in these processes, but on the clear understanding that progress cannot be achieved by creating more insecurity for large sections of the population. If we don’t start a serious and much broader debate on the current realities, we may all be facing a much more unpleasant world economic situation fairly soon.

Mr. Chairman, the painfully slow process of re-examination of current economic problems that is now underway will, we are convinced, eventually lead governments back to the realization that full employment, low inflation, a stable foreign trade balance, and steady growth in living standards can only be achieved simultaneously on the basis of a wider policy consensus arrived at through tripartite consultations and constructive international cooperation.

In this regard I was much encouraged by the outcome of the recent International Labor Organization’s (ILO) high-level meeting on Employment and Structural Adjustment, which set out a constructive program for that organization’s role in international development and cooperation—a role, incidentally, that I believe that the IMF and the World Bank will welcome.

I was also encouraged by some high-level meetings the ICFTU had in Washington in July last year with the IMF and the World Bank. The top officials recognized the value of considering social issues with economic policies. We believe that adjustment through growth makes it possible for socially necessary policies to be worked out that focus on employment and poverty alleviation and are economically efficient.

Effective tripartism can create widespread support for agreed policies of national development that address the urgent needs of working people and establish a sound economic structure for the future. This is where the ILO comes in. It can help to broaden the base for action by involving employers’ and workers’ organizations in the formulation of policy and its implementation.

In conclusion, let me make a couple of political points. Although many speeches by trade unionists are critical of the IMF, we are prepared to admit that if the Fund didn’t exist, we would have to invent it. And furthermore, it would be much harder in 1988 than it was in 1944 to establish a body with the necessary authority to oversee international monetary matters. However, the IMF needs support from its member governments, and if those governments do not in turn have support from public opinion for international cooperation through the IMF and other institutions, they will not be interested in associating themselves with internationalist solutions.

In addition, I think we are all concerned to see the deepening and strengthening of democracy. We must not get ourselves into a position where somehow obligations under international adjustment programs are in conflict with democratic institutions, and I count free trade unions as among the most fundamental. The Fund and its members must be responsive to the calls for wider participation in policymaking; this is not just a matter of attitude, it is a question of establishing mechanisms for discussion that can survive periodic disagreements. It is also not simply a political matter, because as I stressed earlier, in our view, adjustment is likely to work best on the basis of a broad national and international consensus derived from wide participation in the formulation of policies.

Ricardo Arriazu

I would like first to refer to some conceptual aspects of inflation. Economists do not all give the same explanation of inflation. If you are a monetarist you will say that inflation is the result of a disequilibrium in the monetary market; if you are a structuralist you will say that inflation is a reflection of a fight among sectors for the biggest increase in their respective shares of income. I am much more simple than that. I think that inflation is a fraud. The word in Spanish is estafa. I was looking for the word in Italian and was told that it is truffa. Actually, what it means is that the government has borrowed from the private sector and has promised to return real resources, paying back less than what it promised. If this were the only problem, what would happen is exactly the same as what happens in the private sector. If somebody does not fulfill his obligations, then people will stop lending to him, and he will be forced to go to the money market sharks—to the usureros, as we say in Spanish—and pay higher interest rates.

But the problem is that the government is making an estafa with an instrument that is the unit of account of the system. This has much wider implications than you might expect. Suppose, for example, that the government was borrowing a certain amount of wheat measured in tons and that it decided to change the unit of measurement of the system; say, it decided the kilo equaled only 900 grams, or only 700 grams. Of course, it would pay less. But if the unit of account “tons of wheat” is used by the rest of the system as the unit of account of economic flows (wages, prices, etc.), when the government cheats on its own debt, it produces tremendous transfers of real resources among other sectors in the economy. We economists tend to forget that money is used as a unit of account. Therefore, when a country is using inflation as a means to obtain financing and, as a consequence, uses “cheating” as a source of funds, the private sector abandons the unit of account of the government and also abandons the use of national currency as a way of making deals. For example, in Argentina most of the big transactions are not denominated in national currency.

How do you get to high inflation? It is very simple. I can assure you that with persistence, patience, and the right policy, you could all arrive at high inflation. You start with a fiscal deficit and then you persist with it. At the beginning, you obtain large inflationary tax revenues because the tax base is large, so that the tax rate needed (inflation) is low. Then when people start to fly out of the currency and the inflationary tax base starts to be reduced, you need a higher inflationary tax to finance the same deficit.

To give you an idea of what this means, consider the case of Argentina. In 1945 total national-denominated financial assets of the private sector were equivalent to 67 percent of gross domestic product (GDP). Now, they represent only 10 percent of GDP. The non-interest-bearing monetary base used to be equivalent to 25 percent of GDP. Now it is equivalent to only 2.5 percent of GDP. Private sector sight deposits used to be 20 percent of GDP; now they are 1 percent of GDP! The internal private debt used to be very high. Now it is only 7 percent of GDP and is mostly in the form of mandatory minimum reserve requirements, which means that capital markets have disappeared. A large part of this is used to finance the public deficit.

Most of these numbers are fictitious. The fact is that when you have inflation, most of your assets and savings are not held in the form of national currency instruments but rather in the form of foreign currency instruments. More than two thirds of our financial wealth is definitely held abroad; we have almost $30 billion abroad, and less than $10 billion inside the country.

However, the main problem with inflation is not financial but moral. The government starts with a deficit and is very happy with imposing an inflationary tax, because it is a simple way of getting real resources; what the government forgets is that the ones who pay the inflationary tax are normally the least informed segment of the population—the workers. When people start to learn, they begin to fly out of the national currency and to buy dollars, therefore eroding the inflationary tax base, so the government says, “We have a capital outflow problem,” and imposes controls on capital movements. Those people who are taking the money out are supposed to be unpatriotic. But actually, the only thing they are doing is protecting themselves against cheating by the government. We all know that our grandparents have lost all their savings because of inflation. But the government likes this form of taxation, and the public is forced to make a decision: either to buy dollars to protect themselves, which is illegal; or to stay and be cheated by the government. The public finally reacts, saying, “If I buy dollars it is not immoral, even if it is illegal; what the government is doing is immoral.” And for the first time, we start breaking up the relationship between morality and legality, which is the base of any organized society. Once I sunder the relationship between morality and legality, then nothing is an impediment to my going further.

The same thing happens with taxation. What the government normally does is to increase unproductive public expenditures—in doing this, it is increasing taxes only partially to pay for them, which reduces productivity and increases the deficit. Since the tax system is normally inefficient in the face of inflation, people can ask themselves how much tax they want to pay. But if I pay taxes for the government to spend the money in a really inefficient way, then I can ask myself: is it moral or immoral not to pay unproductive taxes? In a context such as the one described here, the public tends to decide how much taxes they wish to pay, and not paying taxes is no longer considered immoral.

Of course, in a context of excessive expenditures, countries will tend to, and most tend to, apply protectionist policies. Protective policies will never solve a balance of payments problem. It is always a question of one sector of the economy against another sector of the economy. But protective policies are widely applied to solve balance of payments difficulties. For example, in Argentina it is forbidden to buy, among other things, computers. We have factories that import all of their parts and sell their computers at prices three times higher than the international price. The question is, should I be noncompetitive, or should I get the computer in any way I can? And, of course, smuggling becomes a way of getting the computer. Again, the public is forced to break the law, because protectionism makes it impossible to become competitive.

Once people start to judge the lightness of certain legislation, where do they finish? What is the definition of morality? Each of us has his own definition of morality, and once I have severed the relationship between morality and legality, I can continue unchecked. In this context, for example, stealing can be moral. After these awful examples, I hope it should be quite clear why inflation is not only an economic problem but a political and a moral problem as well.

Let us go now to the final question, which is how you manage monetary policy in an inflationary environment. The answer is that you do not manage monetary policy; it is monetary policy that manages you. We do not have a capital market, since there is no way of introducing bonds into the system because nobody wants to hold a national-denominated debt instrument. To cover the fiscal deficit you have to print money. But if the non-interest-bearing monetary base is equivalent to just 2.5 percentage points of GDP, a deficit of similar magnitude will produce a 100 percent increase in the monetary base. If we had the Italian fiscal deficit in Argentina—which at this moment we do not have—the inflation rate would be in the range of thousands of percent. Since inflation itself tends to increase the budget deficit, once you are in this situation there is no alternative but to reduce the fiscal deficit practically to zero because you cannot make monetary policy.

How do we change it? Consider the accounting identity, which says that the current account of the balance of payments is equal to the sum of the public sector budget plus the private sector budget. Let us take the case of Argentina last year. We had a consolidated public sector budget deficit of approximately $6 billion. Assume that we could get external financing for only $2 billion as voluntary lending, and that, internally, nobody was willing to lend even 1 percent to the government in real terms. This leaves us three alternatives. Either we get more resources from abroad—voluntarily or mandatorily. Mandatorily means that if we have to pay interest of $4 billion, we do not pay it; so we get $4 billion worth of financing. The other possibility is to get $4 billion internally. But if the total deposit in the system is $10 billion, and we do not have any way of getting more financing internally, the only remaining recourse that we have is forced financing. And forced financing means increasing the minimal reserve requirements. But nonremunerated reserve requirements are an estafa. Perhaps it is possible to get $1 billion in that way—and we are left again with the problem of how to finance the rest of the fiscal deficit. Since I am using dollars as the unit of account, the way to reduce the fiscal deficit in terms of dollars is by inflation. And the way that we produce inflation is by devaluation. Devaluation is in substance a mechanism to obtain inflationary tax. You need to estimate how much you have to devalue in order to get an inflationary tax to reduce your budget in terms of dollars to something approaching equilibrium. Now, the calculus is not simple because the budget itself is related to inflation, so you have to know how much the deficit will increase when you are increasing inflation.

As I mentioned yesterday, there is some point at which the net inflationary tax becomes negative, when it does not produce any real resource transfer to the government. Then you are forced either to get the resources from abroad or to get them from inside the country, because the inflationary tax produces nothing. When this happens and you cannot get resources from abroad or within the country, you are close to hyperinflation. And in Argentina at this moment, what we have is a situation with no internal financing, no external financing, and no way of getting net resource transfers from the inflationary tax.

Therefore, we are very close to hyperinflation.

Luigi Spaventa

Since several references have been made in this seminar to the situation of Italian public finance, let me begin by briefly recalling some aspects of the debate on the control of public expenditure that has taken place in Italy and elsewhere—namely, the new legislation and stricter monetary policy.

Let me start with a few data. The development of the primary deficit goes back to the end of the 1960s and continued in the 1970s; therefore, it is by no means a recent phenomenon. The decisions made by both the Parliament and, no less frequently, by the Government, were made without any regard to their future consequences. The major examples are the legislation on social transfers and pensions, which in Italy are managed in a way that is both inequitable and expensive. When this legislation was passed, no prior thought had been given to demographic developments or the need to cover future expenditures with future revenues. This is particularly true in the case of the civil service—specifically, the measures relating to early retirement, salary increases, and creation of new employment.

It should be clear that in Italy there has never been any Keynesian “fine-tuning.” Expenditure decisions have not been inspired by any Keynesian philosophy. Some form of fine-tuning was attempted on occasion, but it was always ineffective. The response to the short-lived recessions of the 1960s was the decision to increase investment expenditures; but these sums were never used for investment and were gradually appropriated for current expenditures. So there was nothing Keynesian about the policy.

This practice led to debt accumulation; if we look at the profile of the accumulation of nonmonetary debt, we can distinguish two periods. In the early 1970s total debt accumulated, but nonmonetary debt did not because of the high degree of monetary financing by the Treasury. Nonmonetary debt began to accumulate in 1977, continuing until 1979, and then again after 1981. As was pointed out in the paper by Bruni, Penati, and Porta, we passed from a situation in which the main factor in debt accumulation was the primary deficit to one in which the main factor was the self-generating aspect. The change was due to the fall in the growth rate and the increases in the interest rate and the stock of accumulated debt. The primary public sector borrowing requirement (PSBR) was around 6 percent until 1984–85. It has been halved since, although debt accumulation continues at a rate of 4 percentage points a year.

In the wake of these developments, there has been a growing awareness of the need to control expenditure. Now, what are the possible and conceivable ways to control expenditure? Here, the Italian debate has something in common with the American debate. First, there has been lot of talk about constitutional provisions or law enforcement along the lines of the the Gramm-Rudman-Hollings Act. We have, of course, an article in our Constitution that limits expenditure—Article 91, paragraph 4—which, in principle, should ensure the financing of additional expenditure by means of higher revenues.

My problem with new laws restraining legislative power or constitutional provisions is that governments and parliaments are normally very good at circumventing these restrictions. They are very good at underestimating future expenditures and overestimating revenues that should offset future expenditures. They are very good at window dressing. As has recently been pointed out, the U.S. Senate has learned the Italian lesson very well; consider, for example, the extent of window-dressing evident in recent provisions for cutting the budget. The Italian experience does not instill much confidence in these kinds of remedies, even though I think things could be improved, and there actually has been some improvement. I do not want to go into the details here, but we do have a law that could work more effectively and could restrain the legislative power a bit more.

Another possibility that has been debated is an increase of controls. By this I mean external auditing of expenditure bills. The dream of the good Italian parliamentarians who worry about the growth of expenditure is the creation of an entity equivalent to the U. S. Congressional Budget Office—that is, some kind of independent body that audits bills, in the sense of determining that there is no overestimation or underestimation of revenues or expenditures. This would help, of course. But I do not think it would be enough, because the worst part of the story in Parliament does not always happen with the bills, but with the amendments, which are approved on the spot. It is extremely difficult to have an auditing of amendments that are proposed at the last minute, either by the Government, or by the members of the Parliament.

Another topic that has caused serious debate is the possibility for a change in parliamentary procedures. Here I think something could be done. One of the major causes of the uncontrolled increasing expenditure in Italy has been the fact that expenditure bills can often be approved at the committee stage, without going to the floor of the House. At the committee stage, it is extremely easy to find compromises among various parties and to have a sort of club-sandwich bill, in which if I like the ham there is the ham, and if somebody likes the omelet, there is the omelet as well. I think that if the Parliament were to agree that expenditure bills should go directly to the floor of the House, these compromises would become more difficult to effect, and some element of restraint could be introduced. Another method would be to set a limit on the possibility of amending expenditure bills.

The third hotly debated issue at the moment is whether a secret vote should be allowed on expenditure laws and expenditure bills. In the Italian Parliament almost everything is voted on by secret ballot, so that the majority can easily become a minority, especially on matters concerning the budget.

All of these measures, or some combination of them, may be useful. But I have never been able to understand why people worry so much about inventing ways to constrain their own behavior, instead of behaving virtuously; I mean, if you want to be virtuous, all right, be virtuous, instead of going through a whole process of legislation to invent a constraint on your own behavior. The real point, I think, is the political will to do something; you can invent a number of ways to proceed, but there will always be some lack of effectiveness unless there is a political will to act.

The Italian case is complicated by two other difficulties, which are perhaps more substantial. The first one is the political system. As you well know, we have the most stable political system in the whole Western world. It does not matter if governments change every six months, or every two months. At the same time, this is a nonpolarized system, in the sense that all parties fish in the same large pool.

What we really have in the Parliament is a coalition. We do not have a great coalition government in the sense that it includes all parties. The parties in the government have always been the same: there is a kind of implicit understanding that there is at least one party that will never go into the government in the foreseeable future and another party that will always stay in the government in the foreseeable future. However, when I speak of parliamentary coalitions, I mean there can be a convergence of interests in legislating, which very often results in larger public expenditure.

The other difficulty arises from the employment situation, regional inequalities, and insufficient growth.

What I want to point out is that often we remain at the surface of the problem; we simply consider external ways of controlling expenditures, while the deeper troubles and deeper causes are not sufficiently considered. This is a very important issue and there is much good literature on the subject. It is often maintained that nonpermissive, tougher monetary policy would compel the Treasury and the legislature to face their problems; whereas if they are always helped out of trouble, they will never face the stark reality, and they will never try to stick to the straight and narrow path. This view has much dignity and good theoretical support.¹ There are however some counterindications, concerning economic analysis or, if you wish, political theory.

One is the Sargent and Wallace story: the danger that if fiscal policy does not react in the expected way, you may end up in trouble. And you may end up in trouble in two cases: 1) if there is a limit to the amount of bonds agents are ready to hold, as in the Sargent-Wallace story; and 2) if there is a maximum accepted level of taxation, which is the Keynes story in his “Tract on Monetary Reform,” where he talks about France.

The second possible objection is that the effectiveness of this prescription depends on the arguments of the loss function of the political agents. I really think that this point should be investigated a bit more carefully. In the theoretical literature, there is a central bank that only cares about inflation, and a government that cares little about inflation and much about employment. That is a very simplified story. I would surmise that since the 1970s, control of inflation has become dramatically important in the loss function of governments. The most valuable feather in a government’s cap today is the claim: “I have brought down inflation.” So if this is the case, I think governments could be perfectly happy with a tough central bank that shielded them from inflationary dangers even though the cost of this protection might be debt accumulation. They can go to the electorate and say: “When we came to power, inflation was 12 percent; now it is 5 percent.” Debt accumulation has not been very important in the loss function of our politicians. It is a problem for the future. Probably they will not be in power when something happens. At the same time, everybody is happy with disposable income because of the interest paid on the public bonds. Those who maintain that tougher monetary policy and capital liberalization are good ways to provide an indirect incentive for controlling public expenditure should perhaps formalize their argument and try to model this issue with a different loss function from the one normally used in the theoretical literature, where this aspect of the problem is usually neglected.

However, the prescription I have discussed may be effective if tight monetary policy, capital liberalization, and other similar measures eventually precipitate a financial crisis, which may be the only way to force adjustment and may, thus, be a blessing in disguise.

Having begun with the Italian case, I shall end with it. As I said, in my view there has been some improvement. We have had a decline on the order of 2-3 percentage points in our primary deficit. And if we look at the future, we can see that not much effort is really needed to stabilize debt growth. There is a large literature on tax smoothing that maintains that the level of debt is irrelevant, provided you stabilize it, wherever it is. This argument has some merits, although it neglects distributional aspects. In our case, to stabilize the debt ratio in a few years, given a difference between the interest rate and the growth rate of 2 percentage points, we need to achieve a primary surplus of some 1 percent of GDP, with a turnaround of 3–4 percentage points—not a terribly demanding task. However, the fact is that our expenditure structure is very rigid. Our primary expenditure/GDP ratio has remained steady at 44 percent, no matter what has been done: if some items have been reduced, others have grown—because of demographic factors in the case of pensions, as well as for a number of other reasons.

Therefore, I am not terribly optimistic about the possibility of reducing expenditure, even though our chairman here, Professor Baffi, is a member of the commission that is supposed to propose means to this end. I would surmise that of this turnaround of 3–4 percentage points, not more than 1 percentage point can be obtained through expenditure reduction; the rest will have to come from an increase in the fiscal burden. Here I think our governments are behaving very rashly, because they keep promising that the fiscal burden will remain unchanged; with our debt situation, this is impossible. Their political business should be to increase the fiscal burden not only by increasing tax rates, but also by enlarging the tax base. This is not only a technical problem but a political one. At this stage the effort is still possible and feasible. But let another three or four years elapse and we might find ourselves in serious trouble.

Gert Haller

I would like to return to the question of international coordination. I was expecting Professor Spaventa to give me some powerful arguments that would force me to defend the German position. Now he has left me rather alone, and I will have to try to start from a different point. I took part in this conference, and I listened very carefully to what was said by other participants. One point that I found puzzling, not only in this conference but also in other discussions, is the idea some people have of the Federal Republic of Germany as a strange country, a country where people do not want any growth, where people sitting at home watching television do not need anything else. I do not know anybody in Germany who does not want more goods, more services, and let’s say, simply a better life.

This brings me to another point. If one discusses problems of growth in economic contexts and in conferences like this, one can gain the impression that the whole discussion focuses only on demand-management problems, in the sense that if Germany were to step up public spending or pursue an easier monetary policy and, by this means, increase total demand, everything would be fine. This is certainly a misleading point of view. Even if one approaches the question of faster growth from the supply-side view, it would still be only half of the truth, because supply-side economics implies that the government is able to remove obstacles. I think that is a half-truth, because growth is more than just having the management that helps us to reach aims. Growth is something that is in the mind of the people. Growth means being prepared to make a greater effort, to take risks and to be mobile, for example. And if, let’s say, German steel-workers decide to reduce their working time from 40 hours a week to 35 hours and this is done through a democratic process, it has to be accepted, by Germany as well as by other countries, even if it leads to lower growth.

I’d like now to move closer to the field of economics. I tried to explain yesterday how far the economic adjustment process has come in Germany in the past few years. I tried to give figures on that and on the development of the current balance in real terms. The current balance has gone from a surplus of DM 82.5 billion in 1985 to about DM 36.5 billion in 1988. In relation to gross national product (GNP), this means a reduction in the current balance from 5.2 percent to 2.2 percent at a rate of 1 percentage point a year. This is quite a bit and should be taken into account if one talks about an adjustment process, especially focusing on Germany. I also said that a further reduction of the foreign surplus in the years through 1990–91 by 1 percent or 1.5 percent would mean that the internal demand had to rise by about 3.5 percent a year. I’m pretty sure that this cannot be achieved. It would imply growth beyond our present capacity and could easily lead to the danger of rekindling inflation. Moreover, if one asks Germany to increase internal demand in order to help the process of international adjustment, one has to refer to the instruments that could make this possible. In principle, on a macroeconomic level there are two instruments: monetary policy and fiscal policy.

As regards monetary policy—and this is an issue I face not as an economist but as an administrator in the government—far-ranging discussions have been, and still are, taking place in the board of the Bundesbank on whether monetary policy is already too expansionary in Germany. I personally have the impression that the present monetary situation in Germany could be described as a kind of liquidity trap, in the classical sense. Money demand in Germany is highly elastic; all the additional money is easily absorbed by individuals and enterprises. And in the present situation, it seems that long-term interest rates cannot be brought down appreciably below 6 percent in Germany—a limit that long-term interest rates have fallen short of only in a few periods. During the 1950s and the 1960s, when long-term interest rates were somewhere around 6 percent, people bought long-term government bonds. At the end of the 1970s, the long-term interest rates rose enormously, and people consequently lost part of their wealth.

Another consideration that points in this direction is that the interest curve is very steep in Germany. Money market rates are very low and the long-term rate does not go down. This is also normally an indicator for the expansionary stance of monetary policy. Of course, one can discuss the whole question in the framework of real interest rates, and the real interest rate in Germany, compared to the very low rate of inflation, is relatively high. This indicator would probably give another impression. In my opinion, it is a question of whether or not one talks about a liquidity trap or about which additional criteria have to be found to evaluate monetary policy in Germany.

The second important point is fiscal policy. It is probably well known that the budget deficit in Germany will rise sharply to about DM 40 billion this year, which is DM 10 billion more than originally planned. Of course, it is due to the fact that the Bundesbank profit will be zero this year, but it is also due to other factors. What is more important, and was stressed in Vito Tanzi’s paper, is the question of the future burden that will fall on the public, since the German population is aging rapidly. This means a big burden for our social security system and for public finances.

The only question about expanding fiscal policy that could be discussed in Germany is whether it would or could be useful to bring the tax reform, which is planned for 1990, forward to 1989. You probably know that a part of this tax reform has already been brought forward to 1988 and has been added to the tax relief that had been decided for 1988. The amount has been raised from about DM 9 billion to about DM 14 billion. However, to answer the question, one has to know the German tax system, which is different from that of many other countries.

The Federal Republic of Germany is a federal state with strong individual states, each of which has extensive duties as far as education and public investment are concerned. Nearly half of public expenditure is effected by the states, whereas the tax system is nearly completely centralized. This means that all decisions on taxes are made by the central government, with far-reaching repercussions for the states. In the German tax system, the central government gives some amount of total tax revenues (income tax as well as value-added tax) to the states. But nearly every tax measure has to be agreed to by the states in the second house of parliament. This means that the Minister of Finance has to convince his colleagues from the states, which presents a lot of problems for the Federal Minister of Finance.

This point should be kept in mind when tax policy in Germany is being discussed.

Lord Roll of Ipsden

As you have suggested, Mr. Chairman, I have been on both sides of the barricades. At the moment I am on the same side as the so-called economic agents. So I will begin by speaking as an economic agent, particularly as one active in financial markets and in the financial services industry, but with an occasional glance across the barricade where I was some years ago. If you will turn a moment from the broad macroeconomic issues you have been considering for the past two days, I would like to emphasize that what concerns those in finance most as far as fiscal policy is concerned—and what, more than anything else, influences their day-to-day behavior—is the microeconomic aspect, the minutiae of fiscal policy.

Generally speaking, if you talk to someone in investment or commercial banking or securities and try to direct his attention to the raging debate on whether deficits do or do not matter, and whether deficits are responsible for crowding out or for high interest rates, you may hold his attention for 10 or 20 minutes, but he will soon move on to other concerns. For instance, he may ask you how the tax treatment of provisions for certain debts differs in different countries, or even in the same country—a subject that has become extremely important lately because of the enormous burden of sovereign debt and the involvement of commercial banks in it.

Another issue that directly concerns investment bankers is the question of capital ratios, about which there is increasing debate—again arising, at least in part, out of the growing concern over sovereign debt—and the attempt to coordinate differing national capital ratio systems and different types of banking.

Then there is the question of the tax treatment of exchange rate fluctuations. For example, in the United Kingdom, we treat losses on outward exchange rate transactions considerably differently from losses on inward transactions. And the repatriation of revenues derived from exchange rate transactions is often treated differently from the repatriation of profits from the same transactions. Another basic concern involves capital gains and corporation tax—two measures that were introduced by the Labour Government between 1963 and 1966. You may have read in the press that there has been great rejoicing in commodity trader circles in Paris, because the capital gains tax treatment of commodity trade was recently brought into line with that of other capital gains; hitherto, the treatment of commodity trade was much more severe and had caused a tremendous fall in such trade in Paris, compared to Tokyo, New York, and London.

All these examples are intended to make you aware that the economic agents—at any rate, those that I am familiar with—are much concerned with the minutiae of fiscal policy in its execution, whatever the broader concerns may be. However, having said that, I do not mean to ignore the other aspects of fiscal policy. It is quite clear that fiscal policy in its broadest application is perhaps the most potent instrument for affecting aggregate economic activity and growth and is thus of great concern to the operators in financial markets. And here I will turn to the still unresolved debate about the extent to which budget deficits do or do not matter and the perception of these problems in the minds of the economic agents, particularly in the financial services industries.

This question was at the center of an interesting debate in the columns of the Financial Times soon after “Black Monday”—the stock exchange crash on October 19, 1987—particularly in regard to the United States. One of the protagonists in that debate argued that it was really quite absurd to ascribe such importance to the U.S. budget deficit, when, as he put it, in relation to national income, it was similar in scale to the deficits of France and the United Kingdom, one half of that of Canada, and one quarter of that of Italy. I do not want to go into the intricacies of this argument (for example, whether these comparisons of ratios to national income really matter), but I can assure you that financial agents in the markets do not spend their days trying to discover whether the U.S. budget deficit has risen in relation to national income compared with that of Italy. What they are concerned about is the idea that has been drummed into them by some economic analysts—that the U.S. budget deficit does matter in one way or another; this explains why the haggling between the U.S. Executive and the Congress is followed with such interest in the financial markets. Thus, on days when there appears to be an optimistic view of the outcome, the market breathes a sign of relief; whereas, the next day if the view is pessimistic, the market will go down.

This seems to me to be something that economic analysis has not been, and perhaps never will be, able to cope with, but it does not exempt economic analysts from some responsibility. You remember Keynes’s famous dictum:

“The practical men who ignore theory are very often the victims of some defunct scribbler.” Well, often they are also the victims of some very much alive scribbler, and therefore I think you should remember that what concerns them is the perception operators have of what analysts are trying to say.

Another participant in the debate in the Financial Times agreed that the attention paid to the budget deficit had been greatly exaggerated, but ended his article by saying, “But, of course, the perception of the markets is not unimportant.” Indeed so! Let me just say that the concerns that are very properly expressed here (and I still remember enough about economics to share some of them) are not necessarily those that have the most influence on the behavior of markets.

There is a large area between the minutiae of tax treatment, on the one side, and the broader aspects, on the other; and that is the way economic analysis is reproduced in the minds of bankers, securities traders, brokers, investment bankers, and so on. This large area is much more difficult to analyze. For one thing, it contains the results of the mix of government policies at any one time, particularly fiscal and monetary policies. For instance, when I served in the first Wilson government—as an official, of course, not as a politician—we conducted a series of fiscal policy experiments, both microeconomic (the introduction of the capital gains tax and the corporation tax) and macroeconomic (an increase in personal income taxation, both on average and at the top end). I have not yet seen a totally satisfactory analysis of the macroeconomic policies of those years. I think that such an analysis would encounter some difficulty for two reasons. First, during those years, the mix of fiscal and monetary policy was very obscure, because at that time a major struggle was going on over which entity was the guardian of the monetary conscience, the Bank of England, on the one hand, or the government (that is, the Treasury), on the other. The composition of the mix that resulted from that continuing struggle between monetary and fiscal policy fluctuated throughout the period.

Second—and you referred briefly to this, Mr. Chairman—was that other important policy instruments were used that analytical economists regard as dubious, but which, unfortunately, politicians do from time to time resort to. These included incomes policy and what was called in those days “regional and industrial policy,” which was in a curious way a Labour version of supply-side policy. It was really an attempt to influence the supply side of the economy, but with a much larger degree of government intervention and participation—ranging from providing subsidies to directing the location of industries—than would be tolerated or even dreamed of by our present government and many other governments. As a result, you did not get—and this, notoriously, is nearly always the case in our discipline—a controlled experiment. It was quite impossible to isolate the consequences of fiscal policy—at least in the medium and short term—from all the other things that were happening at the same time.

I offer you an example. One fiscal policy measure was introduced with the definite intention, backed by intensive economic analysis, of influencing growth and the engines of growth in the economy. This was the selective employment tax, which was a re-emphasis of the Verdoorn law—that manufacturing industry was really the engine of economic growth. Now, how do you change the balance between manufacturing industry and services? You put a selective employment tax on services. And that is what happened. As I said at the time, it shows how dangerous economics can really be when applied to policy, because what happened—and this was all that happened—was that the next day your haircut cost more.

One final word about financial operators as economic agents. The important factor in this relationship is the way in which fiscal and monetary policy interact. The fluctuating relationship between the fiscal and monetary policy instruments makes itself felt in the stock and foreign exchange markets and all markets connected with investment banking through interest rates and exchange rates. I leave it to the distinguished economists here to trace the path by which fiscal and monetary policy influence these two important rates. But this is really what the operator in the market is primarily concerned with.

A study by the Sanford Bernstein Institute in New York, which came out toward the end of 1987, correlates fluctuations in interest rates with fluctuations in the revenues and net profits of various investment banks. The study showed that there was some correlation, as you would expect, because a large part of investment banking’s profits, or at any rate revenues, is dependent on the margin between different kinds of interest rates. But the correlation is sometimes difficult to identify; first of all, because there is always a time lag, and you can never be quite sure what kind of time lag to build into the correlation. Second, banks’ revenues, particularly those of investment banks, come from many different sources, some of which are not directly, and perhaps not at all, dependent on interest rates—for example, advisory functions, mergers, and acquisitions.

These activities of banks that are not directly related to interest rate differentials, particularly mergers and acquisitions, deserve considerable further study. The growth in these activities is important not only for the investment banks themselves, but because of the way this growth has been brought about through changes in regulations, the liberalization of markets, the tremendous amount of liquidity that is being generated in international financial markets, and the ability now, thanks to globalization and securitization, to raise funds in a variety of ways that were never thought of five, let alone ten, years ago. This availability of finance and the different methods of financing mergers and acquisitions are extremely important new features in financial markets, and beyond that, in the flow of investment.

In your earlier sessions, you appear to have spent a good deal of time debating the pros and cons of Japanese acquisition of U.S. industry, which recalls to my mind a similar debate about the acquisition of European industry by the Americans in the earlier part of the postwar period. This new development, which I think is going to change the location of industry, the location of investment, the direction and volume of capital flows, and all the rest of it, has been brought about by some of the broader fiscal and monetary policy developments that you have been discussing for the last two days. But I do not think that the path these changes will follow is clear yet. I would suggest to you that another conference might well concern itself with this question: what are the ways in which the adoption by governments and central banks of these new concepts of fiscal and monetary policy—changes in regulation, deregulation, reregulation, new institutions, and eventually, liquidity and methods of financing—will affect such activities as mergers and acquisitions and the flow of capital and investment across international boundaries.

Concluding Remarks

Biographical Sketches of Participants

List of Participants and Observers

References