Banca d'Italia's independence in the political and economic spheres

Given the topic chosen for this event celebrating the centenary of Giovanni Spadolini, it is natural for our minds to go back to 1981 - the year when the first Spadolini government took office, and when Banca d'Italia and the Treasury went through a consensual 'divorce', initiating the process that culminated in the independence of monetary policy in Italy. This 'divorce' will be at the core of my remarks today, but before we get there, I believe it is essential to discuss the concept of central bank independence and how it has evolved over the past half-century.

Central bank independence

Although the question of independence for the money-issuing institution had arisen before,¹ the idea as we understand it today emerged in the 1970s, both in monetary theory and in central banking practice. The concept is linked to the purely fiduciary nature of money and to the need for institutional safeguards that are credible enough to preserve trust in its stability and value.

The temptation for a sovereign to exploit the privilege of issuing money to finance public spending by increasing the monetary circulation - thus generating inflation - is perhaps as old as currency circulation itself. Nor is the history of well- or ill-argued objections to this temptation, raised by philosophers, moralists, jurists and eventually economists, any shorter. Yet in ancient regimes, as long as gold and silver coin circulation prevailed, the problem arose in entirely different terms: i.e. in terms of physical debasement² and similar practices; nor were there any modern central banks whose independence needed protection. Even under the gold standard, when central banks began to play an increasing role in managing currency circulation, the peg to the precious metal - however conventional, indirect, often problematic, and with frequent intervals of fiat money - remained a benchmark in monetary theory and practice,³ acting as a brake on monetizing deficits. Protecting the independence of issuing institutions served to maintain and ensure the proper functioning of a monetary regime conceptually founded on an external anchor.⁴

The gold standard collapsed in the first half of the twentieth century under the blows of two world wars and the Great Depression, not to mention its growing inadequacy in an increasingly complex and dynamic economy. For about a quarter-century after World War II, it was replaced by a standard centred on the dollar, as the US remained the only country still committed to gold parity. In 1971, with the declaration of dollar inconvertibility, the Bretton Woods system fell, and with it the last link - however indirect - between circulating money and an underlying intrinsic value.

The breakdown occurred amid prolonged macroeconomic imbalances in the country issuing the peg currency - the United States: trade deficits, public deficit and creeping inflation. To some extent, it was due to a technical weakness inherent in the system itself (known as the Triffin dilemma), which was likely to surface sooner or later anyway. Yet to a greater extent it reflected the fact that events had undermined one of the pillars of the prevailing macroeconomic policy paradigm, in America or elsewhere, namely the misplaced confidence that it was possible to leverage expansionary monetary policy systematically to foster output and employment growth, by stimulating aggregate demand, without jeopardizing monetary stability.⁵ Only two years later, the first oil shock triggered the most violent inflationary surge across advanced countries since the war.A second shock followed a few years later.

The need to understand how to preserve monetary stability once any link to a real underlying asset had been severed - with money being based solely on the sovereign's 'fiat' - was no longer just a fascinating theoretical question: it became a pressing practical issue. It was in this context that the concept of central bank independence was developed as we understand it today.⁶

Theoretically, a series of contributions, including the seminal works of Finn Kydland and Edward Prescott (1977) and of Robert J. Barro and David B. Gordon (1983),⁷ formalised the problem of 'time inconsistency'. In a nutshell: monetary expansion can create a temporary stimulus for the economy, provided that economic agents (producers, consumers) do not expect it to lead to higher inflation. However, the stimulus will be short-lived because, if economic fundamentals do not change, the real long-run equilibrium will not change; ultimately, the only effect of monetary expansion will be to accelerate inflation. As economic agents grasp this mechanism, the effect of monetary expansion on economic momentum will be weaker and weaker, while its effect on inflation, through economic agents' perceptions and expectations, will be ever faster. These dynamics, combined with the exogenous shocks of those years, lie at the root of 'stagflation' in the 1970s, a term coined then to describe a situation that the Keynesian paradigm was unable to explain.

What, then, was to be done? Since the weapon of monetary stimulus was (almost) blunted, the conclusion was: better adopt a prudent monetary policy aimed specifically at anchoring inflation. This is where the problem of time inconsistency comes in. Once inflation expectations are anchored, monetary policy regains the ability to stimulate economic activity temporarily by 'surprising' its agents. Anchoring expectations is optimal ex ante, but creates an incentive for policy reversal ex post. The temptation is stronger if monetary policy is under the control of a government facing imminent elections. This game cannot go on indefinitely without generating an inflationary bias. The only way to break the vicious circle is to remove monetary policy from direct government control and entrust it to an independent and 'conservative' authority,⁸ vested with a clear mandate centred on price stability, limiting its discretion⁹ and strengthening reputational incentives.

The idea, of course, is not that central bankers are 'better' than politicians, nor even, strictly speaking, more forward-looking in a general sense, but that they should have a mandate, a set of rules, a system of incentives, and a framework of safeguards designed to strengthen their credibility and mitigate the risk of time inconsistency with respect to a specific, technically defined target. Hence the architecture that took shape in the following decades: price stability as the primary goal of monetary policy;¹⁰ independence, especially in terms of monetary policy tools; long terms of office; and protection against arbitrary removal. All of this, along with public transparency in central bank action, which is essential in an open, democratic society to prevent a central bank from becoming self-referential.¹¹

Later on came the practice of inflation targeting: a quantitative target, a central role for forecasts and high transparency - a triad that proved effective for macroeconomic stabilisation. Later still, in Europe, the Maastricht Treaty enshrined the principle of monetary policy independence, with the separation between spending power and money creation, the prohibition on monetary financing, price stability as the ECB's primary goal, and the institutional independence of the Eurosystem.

The empirical literature supports the new framework, showing that, in advanced countries, central bank independence is associated with lower inflation and with growth being no less in the long run.

However, let us go back to the period that concerns us: the end of the 1970s. Alongside the evolution of the theory, there was a marked shift in monetary practice, particularly in the United States. Under Paul Volcker, appointed in 1979, the Federal Reserve took a strongly tightening stance, despite being much criticised. At the cost of a temporary but severe recession, the inflation rate was brought down from nearly 15 to 3 per cent over a few years. The Bundesbank followed a similar path.

And what about Italy? In 1979, price growth in our country was in line with the US rate - the following year, it exceeded 20 per cent. Italian inflation was indeed quite sticky. Its persistence was fuelled by a high public deficit, an industrial relations system dominated by wage indexation and based on centralised bargaining without an 'incomes policy', and finally - what interests us most here - the fact that monetary policy was constrained by Banca d'Italia's commitment to acting as the residual purchaser of government securities that failed to find buyers on the market.¹² Even with the 'sterilisation' measures that Banca d'Italia implemented, as far as possible, to neutralise the creation of a monetary base through the Treasury, this situation effectively amounted to fiscal dominance,¹³ a condition that limits central bank independence and the use of monetary policy for anti-inflationary purposes.

In October 1979, Carlo Azeglio Ciampi became governor of Banca d'Italia, after Paolo Baffi was unjustly indicted and chose to resign. A year later, Beniamino Andreatta became Treasury Minister, first in the Forlani government and then in the first Spadolini government.

The divorce

'Dear Governor, I have long thought that many problems in the managing of monetary policy are made more serious because Banca d'Italia's conduct vis-à-vis the Treasury's financing needs is not sufficiently independent'.¹⁴ This is how a letter of 12 February 1981 starts, in which Beniamino Andreatta, in agreement with Carlo Azeglio Ciampi, set in motion what has gone down in history as the 'divorce' between the Treasury and Banca d'Italia, a set of changes giving full recognition to the Bank's independence.¹⁵

The letter ended with Andreatta's request to 'be informed about what Banca d'Italia thought about these proposals, in a context of a close and fruitful collaborative relationship'. Ciampi sent his reply on 6 March, expressing his agreement with the Minister:¹⁶ 'For monetary policy to remain unfettered by the size and the performance of the state deficit over time, it must be possible for Banca d'Italia to regulate its funding of the Treasury in order to achieve its monetary control objectives'.

Ciampi also expressed the same conviction in public, a few days after Andreatta's letter (18 February):¹⁷ 'In Italy, the Treasury's broad access to Banca d'Italia's financing further increases the risk that the control of monetary aggregates will be less effective in the short term. Imbalances between public expenditure and revenue affect the creation of a monetary base because Banca d'Italia either underwrites government securities initially or makes residual purchases of the part not placed publicly. It would be advisable to take steps in Italy as well to remove the elements of automatism in how the central bank finances the Treasury by purchasing securities'.

Maria Teresa Salvemini has written that 'the decision to proceed with the "divorce" was made during a kind of governmental interregnum, between the Forlani government, which fell in May, and the Spadolini one, installed at the end of June. It came into full effect in July.¹⁸

Spadolini claimed ownership of this in terms of operations and governance, referring to the divorce as the 'measures set out in July 1981 by the government led by me, with the contribution of the then Treasury Minister, Andreatta'. In his speech for Banca d'Italia's centenary in 1993,¹⁹ he recalled the various stages of this consensual separation, which also included limiting the overdraft that the Treasury could automatically access on its current account at the Bank to up to 14 per cent of its budgeted expenditure, and the formal power to change the discount rate.²⁰

The Spadolini government was thus the first to have to deal with an act (not 'seditious',²¹ but consensual) that shook the foundations of a system in which 'accommodating the Treasury's financial needs thwarted any aspiration to an independent monetary policy'.²² This was a significant cultural shift and a technical revolution. There was no 'easy learning for the Treasury':²³ it had to modify the auction mechanisms;²⁴ learn the principles of public debt management and how to assess the funding requirements; and decide on the amount of securities to be offered together with their maturity composition.

The twin revolution was needed, i.e. changing budgetary procedures and practices to achieve greater fiscal discipline, but it did not come about.²⁵ Ten years on from the divorce, the state sector's annual needs were still between 10 and 11 per cent of GDP. The primary deficit remained high throughout the 1980s (3.2 per cent of GDP on average), contributing significantly to the increase in the debt-to-GDP ratio, which exceeded 60 per cent in 1982 and 120 per cent 1994.²⁶

Some problems emerged immediately, at the time of the Spadolini government, which found itself leading the way in an entirely new context. Although he took an active part in setting up the 1982 Budget Law, as the fiscal package was being finalised, 'the content was affected first by the conflicts that emerged within the government and among the heads of the economic ministries, and then by the changes, mainly geared towards less stringency, made during the parliamentary debate'.²⁷ Ciampi himself, in his Concluding Remarks in May 1982, said that awareness of the problems was not matched by effective action.²⁸ His warning was addressed to all political parties, but Spadolini took it personally and recalled it for years afterwards, though with no hard feelings.²⁹

The new setup of the Treasury-Banca d'Italia relationship was put to the test in autumn 1982. In October, macroeconomic conditions, including a hint of an uptick in inflation, suggested raising the interest rate at which government securities were offered on the market. Yet the Treasury did not want to change the base rate, with the result that the monthly auction 'was a failure'.³⁰ At the political level, this event exacerbated the debate on debt management, resulting in the struggle between the Treasury and Finance ministers that led to the crisis of the second Spadolini government.³¹

The failure to place government securities led to the depletion of the credit margin available on the Treasury's current account. On 10 November, Ciampi wrote to Andreatta, the Treasury Minister, to point out to him 'for the purposes of the consequent and necessary interventions, the growing, prolonged overshooting of the maximum debt limit permitted by law'.³² The imbalance was allowed to last for up to twenty days, after which by law, Banca d'Italia was obliged to suspend payments to the Treasury account.

The Spadolini government fell in the same month. No countermeasures were taken and the Treasury account deteriorated further. A request for an extraordinary advance from Banca d'Italia was necessary.

On 23 December 1982, in a letter to the new Treasury minister, Giovanni Goria, Ciampi set out a detailed historical reconstruction of the rules that governed the Bank's regime of ordinary and extraordinary advances to the Treasury. It implied that the extraordinary advance could not be arranged by decree law and that a decision by Parliament was necessary.³³

This resulted in the law that authorized Banca d'Italia to grant the Treasury an extraordinary advance of 8,000 billion lire for a one-year period, at a rate of 1 per cent.³⁴ The law was passed by the Senate on 24 January 1983, the last day possible (the 20th consecutive treasury account imbalance) to prevent the suspension of state payments. This was on a Monday. In an oral history document, Ciampi says that the Senate was supposed to meet to vote on Tuesday, 25 January.³⁵ He himself convinced Tommaso Morlino, the then President of the Senate, to bring the session forward to 24 January because the next day would be too late: 'Sorry, I will suspend payments on Tuesday morning: the law is clear'.

These initial events did not stop the move towards central bank autonomy, spearheaded by Giovanni Spadolini's government. Despite the difficulties, Spadolini held a steady course, convinced that only an independent central bank could contribute effectively to the pursuit of stability and development by fighting inflation. In making his policy statements to Parliament in July 1981,³⁶ Spadolini had called inflation 'the most serious threat to the future of our society as an industrialised and advanced society'.

The inflation rate, which had reached 21 per cent in 1980, fell below 10 per cent in 1986, then stood at around 5-6 per cent between the 1980s and 1990s, and continued to decline in the 1990s. On the eve of the adoption of the euro in 1999, Italian inflation had fallen to below 2 per cent.

Spadolini and Banca d'Italia

Given the occasion, this conversation cannot end without a nod to the special relationship that Giovanni Spadolini had with Banca d'Italia, particularly with Paolo Baffi and Carlo Azeglio Ciampi.

Spadolini asked the former, who was now an ex-governor, to be an external consultant for his government on economic and monetary matters. Almost two years later, he proposed that Baffi run for election to the European Parliament.³⁷

Spadolini mainly interacted with Ciampi in the form of a dialogue between head of government and central bank governor.³⁸ In June 1981, ahead of the government's inauguration, Ciampi had a meeting with Spadolini in which he set out his recipe for regaining monetary stability, which had been presented two weeks earlier in his Concluding Remarks.³⁹

Ciampi wrote that 'returning to a stable currency requires a real change in the monetary constitution, involving the function of the central bank, the procedures for public spending decisions and those for the distribution of income'. The governor's approach was based on three pillars: central bank autonomy, new rules and constraints on public spending and a collective bargaining code that relied less and less on indexation.

Spadolini made these ideas his own. When he presented the government's policy statements to the Senate,⁴⁰ he underlined the 'vital need for the fight against inflation, or even for a real anti-inflation pact', he called for 'frequent dialogue between the social partners',⁴¹ and said that curbing public spending and labour costs was a pivotal part of the fight against inflation. He did not talk about Banca d'Italia's autonomy, but Ciampi wrote that '[he didn't] think it was an omission. Spadolini was a staunch supporter of central bank autonomy, but the best way to safeguard and increase it was to act rather than talk about it'.⁴²

In 1993, Spadolini took part in Banca d'Italia's centenary celebrations, with a masterful speech tracing the history of the Bank. The title was 'Banca d'Italia: autobiografia della Nazione',⁴³ which encapsulated the key theme of the speech, i.e. that 'the parabola of Banca d'Italia reflects the life of the unitary state'.⁴⁴

Spadolini traced this parabola, starting with the issuing banks operating in the aftermath of Italy's unification. He dwelt on the 'tragedy' of Banca Romana,⁴⁵ a disconcerting tangle of politics and the economic world in the late 19th-century Italy that was 'desolate and childish'. Yet he wrote that there was a reaction, and it was proof that 'the strength of a parliamentary democratic regime does not consist in the fact that it has no dishonest people in it: this is impossible [...]. Put more simply, in a democracy, dishonest people are found out and punished'.

He spoke about 'national destiny' when recalling that 'the great institutions' of modern Italy have only established themselves 'under the strain caused by dramatic and tormented sequences of events'. The establishment of Banca d'Italia was no exception: 'It came very late' compared with how things should have gone.

He emphasised how the Bank was able to return to its previous position in Italy in the post-war period, partly thanks to its professional skills, which had remained intact '[...] and to the dedicated spirit of service that had managed to survive in the Bank's Via Nazionale headquarters even under the rules of the Fascist dictatorship'. He defined Banca d'Italia as 'the cement and the symbol of our unity'.

Spadolini wrote that the Republic's Constitution does not expressly mention the central bank, but its role and responsibility are tacit in the 'economic constitution underlying the formal constitution' that attributes special powers to Banca d'Italia and makes it 'a pillar without which the government of the economy would be mutilated, incomplete, and in many ways ineffective'.

Conclusions

As Spadolini knew perfectly well, and demonstrated in practice in how his government acted, the protection by which many jurisdictions safeguard central bank autonomy is not, or at the very least, is not only a question of leaving the necessary room to manoeuvre to people with specific technical expertise. This requirement applies in many fields; there is nothing special about money here. What is special about money, especially since the value of a currency has been based solely on the trust of those who use it in transactions, is that autonomy is a necessary means for achieving a precise purpose: ensuring that trust in the currency is maintained, thus enabling it to play its central role in a trade-based economy.

Autonomy has to be established with robust rules and confirmed by consistent practices. The temptation to disregard it, by making the monetary function serve various needs, is always there, but it involves both short-term and long-term risks, of which we need to be aware.

On the other hand, it puts a great responsibility on central bankers, who are called upon to exercise it prudently and judiciously, and to be aware of their fallibility, ready to react to changing situations and to the evolution of delicate monetary and financial mechanisms, and to be transparent when reporting on their actions.

Spadolini would have called it civic virtue. He wrote that 'there is no republic without virtue'.⁴⁶ We would perhaps call it, though less emphatically, competence and spirit of service; maybe a touch of pride; but definitely/above all?, full accountability.