I will argue, first, that central banks must have very clear and transparent policy mandates, which are consistent with the policy tools that they have at their disposal. Without such mandates, which must command political consensus, central banks cannot be held to account for their performance. This requires that central banks must focus exclusively on monetary policy and bank regulation, and not undertake other roles which might create policy conflicts. Secondly, I will argue that central banks need operational independence to pursue their politically determined policy goals. I will explain why this is so important and what it means in practise. Thirdly, I will argue that policies which have distributional implications are inherently political and, therefore, should not be the responsibility of a central bank; instead they should be the responsibility of the relevant government ministry. This principle requires a clear separation of monetary and fiscal policies.

What is the proper role of a central bank?

Central banks are institutions which have a monopoly on the issuance of fiat money; notes and coins which are legal tender within a given jurisdiction or set of jurisdictions. From this monopoly of the central bank is derived its two critical functions. First, the central bank can determine the quantity of money in circulation in the economy or, alternatively, set the price of this money, which is the interest rate. Determining the quantity or the price of money is the essence of monetary policy. No other type of institution can conduct monetary policy because no other institution has the monopoly on the issuance of legal tender. Secondly, the central bank's monopoly of the issuance of legal tender means that it is the only institution which is able to carry out the function of lender of last resort to illiquid banks, which is sometimes necessary to avert a banking crisis. From this function there arises the need to impose prudential regulation on banks, because in the absence of prudential regulation a bank which has access to lender of last resort facilities when it incurs financial distress does not have adequate incentives to manage itself in a sound manner. All countries impose prudential regulation on banks and in most, but not all, countries this regulation is carried out by the central bank.

Monetary policy is one of the three main tools of macroeconomic policy; the others are fiscal and exchange rate policy. Because monetary policy affects private sector expenditures, which constitutes the vast bulk of spending in a market economy (approximately 88 percent of total final expenditure in the Ugandan economy is by the private sector), it is particularly important for macroeconomic stability. Sound monetary policy is a prerequisite for a stable macroeconomy, with low inflation, although it is not always sufficient in the absence of sound fiscal policy.

It is also important to recognise what monetary policy can realistically achieve and what it cannot. Monetary policy can influence the level of private sector spending in the economy, and thereby the level of nominal aggregate demand; it is a tool of demand management. Controlling the level of aggregate demand in the economy is crucial for the control of inflation, but it is much less relevant for the long term growth of the economy, because the latter depends on supply side factors, such as capital investment, the growth and education of the labour force, and improvements in productivity. The supply side of the economy is beyond the orbit of monetary policy and, as a consequence, it is not realistic to expect that monetary policy can shape the long run growth rate of the economy. For this, different types of policies are required, such as those which affect incentives for private investment or the quality of the labour force. Fiscal policy is far more important for the supply side of the economy than is monetary policy.

What does this mean for the mandate of the central bank? The central bank should have clearly defined policy goals which it can realistically be expected to deliver, given the monetary policy tools it has at its disposal. That is why the BOU has, as its primary policy objective, the control of inflation. A well formulated and implemented monetary policy should be able to achieve a target for inflation, on average over a medium term horizon, although not in every single month over that horizon because prices are subject to short term shocks, such as those caused by food supply shocks. If a central bank is given a clear policy objective to control inflation over the medium term, it is then relatively straightforward to judge its performance, by measuring the outcome against the policy objective, and thus to determine whether or not the central bank is doing a good job. The BOU aims to hold annual core inflation to a maximum of 5 percent over the medium term. The public should judge the BOU on how well it performs relative to its target for inflation. Over the last 24 months, annual core inflation in Uganda has averaged 4.8 percent; hence the inflation outturns indicate that the BOU's monetary policy has been successful over this period.

In the 1960s and 1970s, central banks in many countries, including Uganda, were given multiple objectives, many of which were developmental in nature, in addition to their objectives for controlling inflation. Although the developmental objectives were not, per se, unwarranted, pursuing them was highly problematic for central banks for at least three reasons. First, as I have already explained, monetary policy has very little impact on the supply side of the economy and, therefore, cannot be used to achieve developmental objectives such as raising the rate of capital investment or long term economic growth. Secondly, there was a conflict between some of the developmental objectives and that of controlling inflation, with the result that the ability of monetary policy to control inflation was undermined. Thirdly, it is very difficult to evaluate the performance of a central bank which has multiple and conflicting policy objectives, and thus hold it accountable. In turn, if the central bank cannot be held to account for poor performance, it will not face strong incentives to deliver a good performance.

Similar principles should inform the role of the central bank as bank regulator. A bank regulator cannot prevent all bank failures, but provided that it has the requisite statutory authority, which is usually set out in the banking laws, it is reasonable to expect the bank regulator to supervise banks in a manner which both minimizes bank failures and ensures that, if a bank does fail, it is subject to regulatory intervention before its losses erode all of its deposits. This is the essence of the prudential regulation of banks; the objectives of which are the avoidance of banking crises and the protection of depositors. These objectives are both clear and, other than in the most exceptional of circumstances, achievable. What is problematic is to expect central banks to use their regulatory tools to achieve other objectives besides those of prudential regulation; for example attempting to direct the credit allocations of banks. As with monetary policy, this will create conflicts of interest which are likely to undermine prudential regulation and make it harder to evaluate the performance of the central bank and hold it to account.

The operational independence of the central bank

In common with many other central banks around the world, the BOU has operational independence. In Uganda this independence is guaranteed by the Constitution which explicitly states that: "in performing its functions the Bank of Uganda shall conform to this constitution but shall not be subject to the direction or control of any person or authority." What does operational independence actually mean in the context of the central bank and why is it important?

In essence operational independence pertains to the independence of the central bank to set its monetary policy instruments, such as the policy interest rate, free of any interference from other persons or authorities, such as ministries of finance. It does not mean that the central bank is free to set its own policy priorities or objectives. Whether or not the central bank should prioritise the control of inflation or some other objective is an inherently political question which must be determined by a political organ. Often this will be determined by the legislature, with the policy objectives set out in legislation, as is the case in Uganda where the Bank of Uganda Act specifies the functions of the BOU and places emphasis on "achieving and maintaining economic stability". Within the specific framework of politically determined policy objectives, a central bank with operational independence is free to set its policy instruments to best achieve those objectives.

Operational independence is regarded as being of fundamental importance in the institutional governance of central banks because of the notion of the "time inconsistency" of economic policy making. This refers to the incentives that policymakers with short time horizons have to take decisions which may generate short term gains but at the expense of long term costs, and the recognition that rational private sector agents will understand the incentives facing these policymakers and take actions which will circumscribe any short term gains without necessarily mitigating the long term losses.

Policymakers with short term horizons include politicians seeking election. A central bank which is not independent of political interference would face pressure to pursue expansionary monetary policies which generate short term gains in output at the expense of higher inflation in the future. Granting operational independence to the central bank is, therefore, an institutional mechanism to insulate the central bank from pressures to undertake actions which would have damaging long term consequences, and thus can deliver better monetary policy over the long term.

Moreover, monetary policy is likely to have more credibility with the private sector if the central bank is insulated from political pressures. In turn, the greater credibility of monetary policy means that the private sector is likely to take a more optimistic view of inflation prospects and, as expectations about inflation are to some degree self-fulfilling, this will help to bring about lower inflation.

The separation of fiscal and monetary policy

The implementation of monetary policy and bank regulation does not, in general, have significant distributional consequences. For example, an increase in the policy interest rate has macroeconomic affects but it does not usually favour one section of the population at the expense of another. If the implementation of monetary policy and bank regulation had significant distributional consequences, the operational independence of the central bank would be much more problematic, because distributional issues should be determined in the domain of politics, not by technocrats alone.

In contrast, fiscal policy is intrinsically distributional in its consequences. How taxes are raised and where government spends public resources does not have an equal impact on all citizens. Consequently it is imperative, in a democracy, that elected politicians are closely involved in determining the details of fiscal policy. For example, in Uganda all government expenditures must be approved by Parliament, as must all public borrowing and all changes to tax policy. This principle is enunciated in the Ugandan Constitution.

Given that fiscal policy has distributional consequences and requires Parliamentary approval if it is to command political legitimacy, it is imperative that the central bank is not expected to undertake policies which are quasi fiscal in nature, such as providing subsidized credit from its own resources to industries or firms which are regarded as priorities for development. If government desires to pursue such policies, they should be implemented through the government budget, after the necessary Parliamentary approval has been obtained. It is essential to maintain a strict separation between monetary and fiscal policies.

Sound macroeconomic policy also requires the avoidance of "fiscal dominance", which refers to government persistently borrowing from the central bank to finance its own budget deficits. Central bank financing of budget deficits creates money and thus is a source of inflationary pressure. Consequently, the ability of a central bank to achieve its targets for inflation will be undermined if it also has to finance government budget deficits. Sound economic governance requires that governments fully fund their borrowing requirements by issuing debt to the markets, and as is often the case for developing countries, by mobilising concessional external finance. This is a second reason for insisting on a clear separation between fiscal and monetary policy.

Conclusion

I have outlined what I believe are the essential elements of an institutional framework for central banking in the post 2015 era. These essential elements are: clear policy mandates for the central bank focused on monetary policy and bank regulation; a primary monetary policy objective of controlling inflation over the medium term, the operational independence of the central bank and a clear separation of monetary and fiscal policy. Over the last two decades, legislative, institutional and policy reforms have been implemented in Uganda which have put in place a framework which incorporates these essential elements, although we still have some way to go to make them fully effective. I am confident, therefore, that the Bank of Uganda is well placed to support Uganda's development agenda beyond 2015.

By Emmanuel Tumusiime-Mutebile
Governor of the Bank of Uganda.