Introduction

Until recently, with some welcome exceptions such as Elster (1979, 1994, 2000), political philosophers had not discussed in depth the philosophical problems arising from independent central banks. However, interest among political philosophers in the independence of central banks has increased significantly since the beginning of the 2007–2009 financial crisis and the implementation of unconventional monetary policies such asQuantitative Easing (QE) (Fontan et al. 2016; Dietsch et al. 2018; Tucker 2018; van ’t Klooster 2019, 2020; Dietsch 2020).

The Global Financial Crisis led to two distinct concerns about the independence of central banks. The concern about legitimacy asks whether it undermines political legitimacy for democratic governments to delegate very important decisions to an independent body that is not subject to reelection and not easily removed by the legislature. The concern about economic justice asks whether the policies of independent central banks, which have a profound distributional impact on citizens, are in line with economic justice. This paper explores the legitimacy of independent central banks from the perspective of these two concerns. I will argue primarily that we should avoid construing central bank independence in binary terms, such that central banks either are, or are not, independent. This way of construing the possibilities confronts us with a dilemma. Either we must endorse the orthodox economic case for central bank independence-- namely, that it is necessary for economic efficiency-- or we must reject that argument for the sake of democratic legitimacy and economic justice (I assume economic justice differs from economic efficiency).

I will argue that we should instead construe central bank independence in scalar terms, so that independence admits of degree. This helps us to avoid the dilemma, by allowing us to develop an account of independence in which central banks can retain it to the extent necessary for economic efficiency, while meeting reasonable concerns regarding democratic legitimacy and economic justice. The mandate of the central bank has to include, besides its main goals of price and financial stability, other societal and distributional goals, such as protecting citizens from unemployment and other forms of absolute deprivation. However, once it has secured these goals, the government can choose to design a fully independent central bank only in charge of reaching the Pareto efficiency frontier, or a less independent bank that promotes other values co-extensive to those of the government even if they do not maximize the prospects of the least advantaged.

The article proceeds as follows. 'The Normative Standing of Central Banks' briefly explains the normative significance of central banks. 'The Orthodox Economic Case for Central Bank Independence' presents the orthodox economic case (OEC) for independence, traditionally endorsed by central bankers, which proposes a clear division of labor between the central bank, which it assumes seeks efficiency, and the fiscal authority, responsible for distributing benefits and burdens among citizens according to sound principles of distributive justice. 'The Case for Democratic Central Banking' examines the second institutional design for central banking-- what I call the case for democratic central banking (CDCB)-- which requires that a central bank’s goals are not restricted to price and financial stability, but are rather co-extensive with those of the government, including for example economic justice and greening the financial system. I conclude that we should take into account current concerns about the consequences of monetary policy, and, in particular, unconventional monetary policies such as QE in distributive and climate justice and the OEC’s inability to respond to them. In 'Distributive Justice, Political Legitimacy, and Central Banks', I examine the claims of political legitimacy and distributive or economic justice of these two cases for the institutional design of the central bank offered in earlier sections. Finally, I defend my scalar view of independence, which proposes reasonable trade-offs between efficiency, protected by independence, distributive justice, and a democratic justification of the legitimacy of independent central banks.

The Normative Standing of Central Banks

Independent central banks raise familiar worries about their legitimacy. Political legitimacy is understood here to involve a political authority possessing a justified right to rule. This right involves a permission to exercise coercion through law, and to issue commands and thereby create valid duties for the authority’s subjects (Raz 1985; Shapiro 2002). Some claim that given the distributive impact of monetary policy, the government, with a democratic mandate, should recover control over central banks, because independent experts lack the procedural or democratic legitimacy necessary to take decisions with such profound distributional consequences.Footnote 1

However, the problem with the common understanding of central banking is that it forces us to think that the political legitimacy of independent central banks can only be secured by giving more control of the central bank to the government, thus making the central bank more directly accountable to the electorate. In response, I will argue that we should instead construe central bank independence in scalar terms, so that independence admits of degree between two opposing views of central banking.

The first view is what I call the orthodox economic case (OEC) for independence. In contrast, the second view is based on government control of the central bank. I will call this the case for democratic central banking (CDCB). I will try to present the best arguments for each of these two cases, and the consequent institutional designs for central banking they propose to defend their respective claims to political legitimacy and economic justice. My own view tries to reconcile the classical time inconsistency justification of central bank independence and the OEC with our concerns about the procedural legitimacy of this institution and the justice of monetary policy. I will proceed by offering a plausible account of the duties of distributive justice and the justification of the central bank’s legitimacy, for each of the two views. Finally, as anticipated, I will argue that we should think about central bank independence in scalar terms, so that independence admits of degree, thus making this institution adaptable to different economic scenarios that might vary across time.

Before drawing on philosophy, it is worth noting some empirical facts relevant to what follows. Independent central banks standardly pursue certain general goals, such as controlling the money supply and securing price and financial stability. Traditionally, the main role of central banks, according to Goodhart (2010), is to be in charge of the money supply through open market operations, in order to adjust their balance sheet, fix the interest rate, and monitor the risks of strategic financial institutions. To understand how the open market operations of central banks work, we need to take a look at their most important feature, that is, their monopoly on the issuance of currency.

There is a hierarchy of money, and central bank money is the ultimate form of settlement between economic agents (Pistor 2013; Tcherneva 2017). It is true that the central bank is not the only institution that creates money: private banks create deposits out of nothing when they grant loans to their customers, but they do not lend out reserves and cannot increase the money supply as such; this results only from actions of the central bank, the public, or the government (Sheard 2013). Privately created money constitutes 97% of the money created in our economies (McLeay et al. 2014a, 2014b). However, the special power of central bank money to issue the currency and create reserves to compensate its own debts as the ultimate form of settlement between economic agents makes it the best institution of our democracies for achieving price and financial stability.We have changed 'Mcleary et al. 2014a, 2014b' to 'McLeay et al. 2014a, 2014b' to match the reference list. Please check.Ok, thanks

A simple explanation is that in order to achieve price stability, the central bank has control over the short-term interest rate that is charged to commercial banks. Since commercial banks hold accounts in the central bank, this official short-term interest rate affects their operational costs, and thus they adjust the interest rate that they charge to other market participants. These changing costs for economic agents influence their decisions about investment and consumption, which in turn change the level of inflationary pressure on the economy (Dietsch et al. 2018, p. 7).

However, in order to understand the hierarchy of the financial system and the central bank’s role in it, we need to be a bit more precise. The main channel for monetary policy implementation consisted before the Global Financial Crisis of open market operations. To get access to more liquidity, commercial banks can turn to each other in the interbank money market, that is, the market where commercial banks lend to each other to meet their short-term liquidity needs. To influence the effective interest rate in the interbank market, the central bank changes the amount of liquidity to which commercial banks have access through open market operations. Central banks swap with commercial banks an amount of liquidity for specific assets that act as collateral. To inject liquidity, the central bank acquires assets from commercial banks, creating central bank reserves, and these open market operations 'affect all the other interest rates by first affecting the availability of cheap credit on the interbank lending market'(Dietsch et al. 2018, p. 8). In this pyramidal system, the central bank is at the top, since it is the only institution that can settle its own debts by creating its own reserves, something that private banks are not allowed to do.

As explained above, the central bank’s open market operations are used to fix the interest rate and to manage the money supply. Suppose the bank wants to expand the amount of money in the economy; then it buys bonds and pays for them by creating money. In contrast, if it wants to contract the money supply, it sells bonds and removes from circulation the money that it receives from the exchange of bonds. As the central bank buys bonds, the demand for bonds goes up, increasing their price while the interest rate on bonds goes down. In contrast, when the central bank makes a contractionary open market operation, it decreases the bonds’ price and increases interest rates on them. The management of these open market operations aims at driving market rates into line with the separately set official rate by the central bank.

Historically, the central bank’s capacity to lend via open market operations was also the main focus for the stabilization of financial markets (Goodhart 2010, p. 9). These open market operations to fix the interest rate and secure price and financial stability imply profound political choices with severe distributional consequences made by unelected experts. However, central banks responded to the Global Financial Crisis with unconventional measures that went well beyond the previous open market operations mechanism. For instance, on October 9, 2008, the Federal Reserve Bank started paying interest on the reserves held by private banks. After Lehman Brothers’ fall, the Fed also began implementing Quantitative Easing programs, which involves the central bank supplying excess reserves, 'as an easing policy after the interest-rate ammunition has been exhausted'(Sheard 2013, p. 9). It is a policy that comes on top of near-zero interest rates and allows the central bank 'to change the composition of the aggregate portfolio held by the private sector; the central bank takes out of that portfolio the government debt and other securities it buys and replaces them with reserves and bank deposits'(Sheard 2013, p. 9). The aim of the central bank with QE is to increase credit creation over time, once the classical mechanisms of open market operations and the interest rate policy have been exhausted.

A helpful starting point for thinking about the normative standing of central banks is the Rawlsian notion of the basic structure (Rawls 1999, pp. 6–10). The financial system is part of the basic structure of society. It is important to note that this system contains private elements, such as commercial banks and investors, and public ones, such as the central bank and regulators (de Bruin et al. 2018). Central banks and regulators, being the public elements of the financial system, are part of the basic structure and connect the politics of finance with questions of legitimacy and justice (Fontan et al. 2016; Tucker 2018; Dietsch et al. 2018; van ’t Klooster 2019, 2020; Fontan and van ’t Klooster 2020; Dietsch 2020).

These claims to legitimacy and the effects of central banking in social justice have changed profoundly since the beginning of the Global Financial Crisis. The overnight rate policy changed the interaction between central banks and private financial institutions, even if it did not put more money into the economy. Furthermore, the 'portfolio rebalance effect' of QE is a form of increasing money supply which has new and unprecedented distributional effects. If one assumes that traditional monetary policy before the financial crisis had relatively little impact on other policy objectives (e.g. containing economic inequalities), as long as monetary policy achieved its goals of price stability and financial stability, central bank independence did not jeopardize the legitimacy of political institutions. However, if one also thinks that since the financial crisis the collateral damage of monetary policy actions to other policy objectives has substantially increased (Fontan et al. 2016), then one might argue that (certain) central banks now threaten legitimacy, precisely because the massive purchases of corporate bonds and financial derivatives performed by central banks under QE programs have such profound distributional effects, which are not compensable by the fiscal authority. Therefore, it might seem that they threaten democratic legitimacy and economic justice, as I will explain in what follows.

The Orthodox Economic Case for Central Bank Independence

After the Great Depression and the Second World War, the main role of the central bank focused on employment and promoting stable economic growth. The acceptance of the Phillips Curve led to the widespread assumption that central banks could choose between different combinations of unemployment and inflation (Phillips 1958). A country could achieve low unemployment if it was willing to tolerate higher inflation. Alternatively, it could achieve price stability if it were to tolerate higher unemployment. According to Goodhart (2010), this era of central banking was characterized by strong government control over the central bank.

In the 1970s, after the appearance of stagflation, the idea of a stable trade-off between employment and inflation was abandoned. Central banks became committed to macroeconomic stability, low inflation, and moderate interest rates. Friedman (1968) and Phelps (1968) argued that in the medium run, government attempts to obtain low unemployment by accepting higher inflation are prejudicial. The initial beneficial effects of inflation on employment disappear once economic actors adjust their expectations regarding inflation. To understand this influential argument, suppose the central bank is in government hands and the latter officially commits itself to price stability. However, when the next election looms, the government is unlikely to avoid using the money supply to promote its electoral interests. Because the ruling party aims to maintain power and the electorate can be influenced by manipulating interest rates, the government is less likely to maintain its previous commitment. This is known as the time inconsistency problem of optimal inflation policy (Kydland and Prescott 1977; Rogoff 1985).

The OEC argues that a government has an interest in reelection, and to gain votes it is prone to choose to expand the money supply to promote economic growth and reduce unemployment. In the short run, this increase in money supply leads to a lower interest rate, which leads to an increase in investment, and, in turn, employment. Nevertheless, in the medium run, an adjustment in price level expectations takes place. The lower unemployment rate also leads to an increase in prices. As a result, prices are higher than the wage setters expected. They then revise upward their expectations for the increase of price rates in their wage claims. Finally, businesses increase their prices at the same rate. The real interest rate, or the nominal interest rate minus inflation, reflects the increase in inflation. The real interest rate is the basis for calculating the return on investment, and thus investment decreases. The higher output after the government’s initial intervention in favor of employment comes back, in the medium run, to where it was before the increase in money supply, while inflation remains high.

Summarizing, justifications of central bank independence can be divided into three arguments. (i) Independence design neutralizes the impact of one irrelevant factor on the supply of money. Central bank bureaucrats not subject to electoral pressures and time inconsistency problems provide a more stable monetary policy, which leads to lower inflation in the long run, which is good for investment and employment in the long run. (ii) Independence is also a pre-commitment device that prevents voters from being manipulated by an unconstrained government, because an independent central bank eliminates the political failures of a government that is subject to electoral pressures. And (iii) it is designed to contain the growth of the money supply as such. Given the nature of markets, if independent central banks control the money supply, investors have more faith that the long-term low-inflation policy will protect returns on their investment.

For decades, during what has been called 'the Great Moderation Era'(Stock and Watson 2002), central bankers were seen as apolitical bodies. The independence of central banks was instrumentally justified when their only goal was to fix inflation with a single instrument, the short-term interest rate, through open market operations. Time inconsistency problems caused by electoral pressures make governments less able to promote long-term stable inflation (Kydland and Prescott 1977; Rogoff 1985). Thus, delegating monetary policy to unelected experts was seen as a self-binding device for overcoming such electoral pressures and promoting long-term price stability. It is better for the government to tie its hands if it wants to achieve the target of price stability. Central bank independence ensures this goal through the appointment of financial experts who are not subject to reelection, and who cannot easily be removed by the legislature. Insofar as independence involves the delegation of powers by the government, it is similar to establishing constitutional constraints. The government acts as in the case of Ulysses and the Sirens. It exercises its ability to bind itself in order to achieve a target in the long term (Elster 1979, 1994).

It is important to note, however, that setting up an inflation target is a decision with distributional consequences. If the central bank raises the interest rate to fight inflation, this will have an impact of less economic output and more unemployment. This decision is clearly a political choice with distributional consequences. Van ’t Klooster (2019) argues that unemployed people are among the worst-off members of society, so this contractionary monetary policy seems to contradict our standard view of distributive justice. That is, the view that claims that inequalities should serve to maximize the prospects of the least advantaged (Rawls 1999 [1971]). However, for several decades before the Global Financial Crisis of 2007–2009, it was normally thought that the fiscal authority had the tools needed to achieve distributive justice-- despite the distributive effects of central banks’ decisions-- and that they could compensate for such distributive effects of the central bank’s decisions.

The Great Moderation Era ended with the deepest global financial crisis since the 1930s, leading to central banks’ political scrutiny. After the 2007–2009 global financial meltdown, central banks recovered with tremendous energy their interest in broader financial stability, and started using various instruments besides managing the short-term interest rate. QE has to be understood as an instrument for achieving financial stability in the financial system and the government’s debt markets after the 2007–2009 Global Financial Crisis once the interest rate was near zero. New policies such as QE and their distributional impact have attracted interest among central bankers themselves, economists, and political theorists. Some think it is less acceptable that independent experts can choose any unconventional means to achieve price and financial stability when these policies have profound distributional consequences (e.g. Fontan et al. 2016; Tucker 2018; van ’t Klooster 2019, 2020).

For these purposes, it is helpful to distinguish between the ends and the means of the central bank (Dietsch 2016). The central bank’s mandate establishes the ends or goals that this institution needs to achieve, namely price and financial stability, and the central bank has a freedom of means to achieve these ends. Before the crisis, the central bank used just one instrument, the short-term interest rate, to achieve the ends established in its mandate by the legislature. Since the Global Financial Crisis, central banks have started to use multiple instruments and means to provide price and financial stability. This new role for the central bank implies a much broader set of political choices, almost unlimited, since each of these means has, for example, different distributive effects, and this is exactly what political theorists find troubling.

A growing number of political philosophers have been examining the social responsibility of finance and the distributional consequences of QE programs. The central worry within the recent literature on the political philosophy of finance is that monetary policy implies decisions made by independent unelected officials that have enormous distributional consequences. For instance, QE programs favor bondholders and stockholders who see how their assets increase in value (Montecino and Epstein 2015). Since these bondholders and stockholders are among society’s most advantaged members, some might claim that QE is unjust (Fontan et al. 2016).

These distributional consequences are problematic for two reasons. First, central banks make profound political choices, and some authors are concerned about whether these decisions serve distributive justice. Second, some question whether unelected officials can have such freedom of means to achieve price stability and take these deeply political decisions without undermining democratic or procedural legitimacy. New policies such as QE do not fit well with the traditional justification of central bank independence as a self-binding device for achieving stable long-term inflation with a single instrument, the short-term interest rate. Some authors claim that it is especially problematic that its current narrow mandate, focused on price stability, can include such profound political decisions, given central banks’ independence and weak democratic accountability. Fontan and van ’t Klooster (2020), for instance, claim that environmental monetary policies are deeply political issues, and challenge central banks’ legitimacy when they buy 'brown assets' through their QE programs. This opens the door to the second model of central banking, which I call-- following van ’t Klooster (2019), but also departing from him in my account of it-- the case for democratic central banking.

The Case for Democratic Central Banking

In this section and the next, I will use a Rawlsian framework to examine these two issues concerning distributive justice and the political legitimacy of central banks. The OEC claims that the best arrangement is for the central bank to promote an efficient allocation of economic resources in the long run by keeping inflation low. In A Theory of Justice, Rawls argues that his two principles of justice apply primarily to the basic structure of society: the major political and economic institutions that exert profound and unavoidable effects on citizens’ motives and expectations of primary goods (Rawls 1999 [1971]), p. 6, 1977, pp. 159–165). The central bank is certainly one of the key institutions of the basic structure, and thus some might argue, against the OEC, that the central bank should be governed by the same principles of justice as the government. I call this view the case for democratic central banking (CDCB).

Before examining the CDCB’s claims to political legitimacy and economic justice, I should cast some light on the issues of distributive justice and the OEC. According to the latter, there is a clear division of labor between the central bank, which seeks economic efficiency, and the fiscal authority, which implements distributive justice. In A Theory of Justice, the difference principle regulates the differences in lifetime expectations of primary goods and requires such inequalities to be arranged to maximize the lifetime expectations of primary goods enjoyed by the least advantaged members of society. However, Rawls also distinguishes between four branches of the government: the allocation branch serves to keep the price system workably competitive; the stabilization branch serves to maintain full employment and to ensure that free choice of occupation and the deployment of finance are supported by effective demand; the transfer branch guarantees a certain level of well-being and a social minimum; and finally the distribution branch serves to preserve approximate justice in distributive shares (Rawls 1999 [1971], pp. 243–244). The allocation and the stabilization branches of government, according to Rawls, aim to 'maintain the efficiency of the market economy', and, then, the transfer and the distribution branches are necessary 'because the efficient market outcome that could be secured by the allocation and stabilization branches working together gives no consideration to needs and therefore requires supplementation by a system of social transfers'(O’Neill and Williamson 2015, p. 65).

On this view, Rawls’s theory of social justice can be read as requiring that 'different institutions meet different claims'(Rawls 1999, p. 244). O’Neill and Williamson (2015, p. 66) state that 'Rawls’s commitment to this division of labor leads him to favor certain policies over rival ways of addressing the same problems'. One might suggest, something that Rawls did not do, that following this branch distinction, the central bank is committed to maintaining price stability and efficient market outcomes, while the transfer and the distribution branches are the parts of the basic structure responsible for securing a sufficient level of advantage for everyone, and for other demands of social justice.Footnote 2 Seen in this light, one might argue that there is a neat division of labor between the government and the central bank. The former is guided by principles of justice, while the latter should promote efficiency alone. According to the OEC, the best arrangement is for the central bank to reach the Pareto efficiency frontier, allowing the government to choose amongst the set of Pareto efficient allocation of resources. Once the central bank has done its part, the government is responsible for choosing the efficient distribution that maximizes the lifetime expectations of primary goods for the least advantaged. As an example that illustrates the OEC view of distributive justice, if low inflation promotes the long-term maximization of productivity by creating unwanted unemployment, then there are other government branches that should provide enough redistribution, unemployment benefits, and social security to compensate for the distributive effects of inflation on the unemployed.

Central bankers have traditionally been reluctant to assume any responsibilities regarding distributive or economic justice. The central bank, they usually claim, has no responsibility for distributive justice, and need not consider this issue in its policy-making. For instance, concerning climate justice, in a recent report the Bundesbank claimed that central banks should operate under a principle of market neutrality and cannot substitute for climate policy-makers (Weidman 2019).Footnote 3 'Neutrality' can have two different meanings here, however. First, it might mean that monetary policy should not interfere with the markets and should be neutral in its effects on the markets. This, however, is an implausible view given the distributive effects of inflation and QE (Montecino and Epstein 2015). It is more reasonable to think that 'neutrality' here refers to neutrality of justification (Raz 1986; Kymlicka 1989); therefore, central banks’ preferences for broad and liquid assets in their transactions are justified in order to achieve price and financial stability, regardless of the effects of monetary policy on distributive or climate justice. However, some central banks have other goals besides price and financial stability. For instance, the Federal Reserve Bank’s mandate includes the goal of maximizing employment, and the European Central Bank’s mandate aims to promote the EU’s overall economic goals. In these cases, these central banks depart from market neutrality, in the second sense mentioned above, in order to include these goals. As a result, they must make trade-offs between these different goals, at the cost of achieving a suboptimal inflation policy. One natural response to the concern about the legitimacy of QE is to include distributional goals in the central bank’s mandate, and accept that this institution must make these trade-offs between different values and thus depart from neutrality. The delegation of powers to an independent central bank may result from a law passed by the legislature, which can then impose limits on the central bank’s remit. Therefore, including distributional goals in the central bank’s mandate is a way of departing from the OEC and coming closer to the CDCB.

The Bundesbank claims that the central bank should not get involved in climate policies. But that is because they recognize the political status of these issues, and then conclude that central banks are not political institutions able to make such political choices. Meanwhile the CDCB claims that it is worth changing the mandate of the central bank, where this amounts to recognizing the political status of monetary policy. But in contrast to the Bundesbank, the CDCB concludes that the central bank should get involved in economic and climate justice-- albeit with its powers explicitly delegated and limited by the legislature and the government, reinforcing its procedural legitimacy.

One possible problem with this solution is that changing the mandate of central banks might not be politically feasible for nonideal reasons like political partisanship, or it might take too long given the climate emergency facing humanity here and now (Diestch et al. 2022). Therefore, we might want to constrain the central bank’s means for achieving price and financial stability in another way. In this case, the central bank and the government should coordinate to promote economic or climate justice. This again will reinforce the central bank’s procedural legitimacy and can be accommodated by the CDCB, albeit at the cost of its independence and optimal monetary policy, reducing this time its instrumental value.

Distributive Justice, Political Legitimacy, and Central Banks

Central Banks and Distributive Justice

To examine the OEC’s and CDCB’s distinct claims to distributive justice, I will continue using Rawls’s framework for the assessment of political institutions. It is useful for these purposes to distinguish between different interpretations of Rawls’s difference principle, which may help us to understand the OEC and the CDCB’s distinct claims to economic justice.

Williams (2011) distinguishes between the maximinizing and the non-maximinizing interpretation of the difference principle. The former allows inequalities that are not detrimental to the least advantaged and maximize their benefits (Rawls 2001, pp. 59–60). Thus it requires a broader range of inequality in the distribution of benefits than the non-maximinizing view of the difference principle, because it requires maximizing the promotion of benefits for the worse-off. As mentioned earlier, the OEC claims that there is a neat division of labor between the independent central bank and the government. The best arrangement is that the independent central bank tries to maximize efficiency and provide the government with a set of Pareto efficient allocation of resources. Meanwhile, the government is responsible for choosing the Pareto efficient allocation that maximizes lifetime expectations of primary goods for the least advantaged.The reference: (Rawls, 2001) is present in the text but not in reference section. Could you please check?Rawls, J. (2001) Justice as Fairness: A Restatement, (Cambridge: The Belknap Press of Harvard University.

This view of the duties of economic justice of the OEC, and the division of labor between central bankers and elected politicians, is opposed to the non-maximinizing interpretation of the difference principle that permits but does not require maximizing wealth for the worst-off, but merely prohibits increasing inequality in ways detrimental to the least advantaged (Rawls 1999[1971], p. 68; Williams 2011). Thus, the CDCB’s claim to distributive justice allows the government to constrain the political choices of the central bank, even if it does not maximinize the prospects of the least advantaged. The CDCB does not require the central bank to provide a set of Pareto efficient allocation of resources, and in this sense, monetary policy becomes suboptimal. Given that in the real world Pareto improvements are not really feasible, since political decisions in general, and monetary policy in particular, always create winners and losers, it adopts a non-maximinizing view of the difference principle because, falling short of the best arrangement, it permits but does not require maximizing wealth for the worst-off, and instead merely prohibits increasing inequality in ways detrimental to the least advantaged. According to the CDCB and the non-maximinizing view, the central bank does not need to reach the efficiency frontier because the government, departing from the benchmark of equality, can choose any point between equality and the Pareto efficient point that maximally improves the situation of the least advantaged. However, once everyone enjoys a social minimum, even if it is permissible, the government is not required to promote their lifetime expectations of primary goods any further, which means that it can accommodate a suboptimal monetary policy for the sake of a more egalitarian distribution of resources, or to preserve the climate.

Reflecting on the unintended distributional effects of monetary policy might lead us to the view opposing the one traditionally held by central bankers. Democrats might well argue that if the central bank is not to threaten political legitimacy, what we need is for the government to recover control over the institution and apply sound principles of justice in its policy-making, since only a democratically elected government is entitled to make such choices. Considering the adverse effects of central bank independence on economic justice, democrats might claim that we should abandon central bank independence and recover democratic control over the institution, and this too might lead us to an opposing view of the central bank’s duties of distributive justice.

On this democratic reading, central banks should pursue the same distributional goals as the government. This view appeals to the fact that the central bank is one of the main institutions of the basic structure of political institutions, and makes decisions that have profound distributional effects (Rawls 1999 [1971], pp. 6–10, 1977, pp. 159–165). The central bank’s goals should then be coextensive with those of the government. Contrary to the division of labor between independent central banks and the government favored by the OEC, the central bank should adopt the full range of distributional goals applicable to governmental decision-making.

However, do central banks need to be guided by the full range of distributive principles that apply to government decision-making? Those who endorse the OEC may claim that this coextensive view of the values of central banks relies on the ambitious assumption that all decision-makers with profound distributive impact should pursue the same distributional goals. This assumption is questionable, however, since we can identify examples where policy choice has a profound distributional impact that certain decision-makers may disregard. Suppose, for example, that a government agency decides to impose on milk producers certain conditions on rearing cows (e.g. more space for each cow in their farms), and that these regulations in turn will increase the price of milk for consumers. It does not seem reasonable to claim that the agency must attend to these distributive consequences, and thus has reasons against adopting these regulations, because of the negative distributional impact on milk consumers. It seems more likely that if this distributional effect is relevant then it should guide the decisions of some other government agency. Still, the unintended distributive effects of unconventional monetary policy are relevant if they are not compensable by the fiscal authority, and we have reasons to claim that elected governments are better situated than unelected central bankers to take decisions that have profound distributive consequences, or at least, that the central bank must serve the government’s goals by enhancing the government’s overall economic policy and how it serves social justice (van ’t Klooster 2020).

Central Banks and Political Legitimacy

With regard to political legitimacy, one might wonder whether the OEC is compatible with democratic values. Recall that the concern about the legitimacy of central banks asks whether it is legitimate for democratic governments to delegate very important decisions to an independent body that is not subject to reelection and not easily removed by the legislature. To proceed, we may recall the distinct ways in which the political legitimacy of democracy has been defended. One main guiding distinction is between proceduralism, which defends the authority of democracy solely on the basis of the fairness of its procedures in which everyone has an equal say. By contrast, instrumentalism defends democracy on the basis that complying with its directives produces the best consequences over time compared with any other feasible form of political authority.Footnote 4

The independence of central banks is relatively uncontroversial from the perspective of instrumentalism. Instrumentalism can easily accommodate the OEC and the claim that democratic regimes should include precommitment devices that help to diminish time inconsistency problems, thereby enhancing the quality of monetary policy over the long term, and that this is valuable given the nature of financial markets. The legitimacy of central bank independence and the OEC is much harder to explain, by contrast, if one adopts a pure proceduralist account of legitimacy. It appears very difficult to reconcile the delegation of very important decisions to an independent body not subject to reelection, and which cannot easily be removed by the legislature, with the idea that decision-making procedures ought to be sensitive to the wishes of the electorate.

On this reading of the legitimacy of central banks, because the decisions of central banks benefit some at the expense of others, such conflicts of interest should be resolved fairly. If the government is faced with these choices, and selects some particular just option, then it can at least say, in its defense, that it has a democratic mandate. Now imagine a central bank that pursues a particular option without consulting the government. It cannot say that its choice is uniquely favored by the elimination of inefficiency, and it also cannot say that it was elected. This lack of a justification for the specific choice made sounds troubling.

Yet I claim that this analysis of the implications of pure proceduralism is too quick. An uncompromising pure proceduralist faces the objection that it is deeply implausible to say that all highly technical issues should be under direct democratic control. On further reflection, a pure proceduralist might find the delegation of the government’s decision-making powers to an independent central bank unproblematic, at least provided two conditions are fulfilled. First, the decision to isolate the government from monetary policy has been taken by an elected democratic government, with popular support, which has decided to protect citizens from manipulation by political representatives who are pursuing their own interests of reelection. Secondly, if the bank’s brief is to avoid the time inconsistency problem, then it can do this without complete indifference to government values and the electorate’s preferences. A more plausible position for the pure proceduralist is to allow central bank independence provided that the right institutional design is chosen for the bank: a design that includes the promotion of some form of democratic participation in central bank decisions, such as public forums in which the decisions of the bank can be reviewed, perhaps by a special committee of parliament assisted by experts.

So far, I have argued that central bank independence can be reconciled with instrumentalist and procedural conceptions of political legitimacy, provided we avoid construing 'independence' in overly binary terms. The independence of the central bank normally consists of a freedom to set only the means for achieving certain ends (price and financial stability), and not the ends themselves, which are set by democratic authorities. To accommodate proceduralist concerns, the political choices of independent central banks should not be unlimited: the legislature and the government might not only impose the ends the bank has to achieve, but also constrain the means available to achieve such goals. Notice that the OEC for independence allows the restriction of means by the government so long as the means that are removed are not necessary for avoiding the time inconsistency problems caused by the political failures of a government subject to electoral pressures.

A special committee of the parliament should discuss these facts and alternative unconventional policies, and provide guidelines for the central bank’s future decisions. Furthermore, on this matter, Elster (1994) suggests that if disagreement between elected institutions and the central bank persists, then the legislature, by a majority of two-thirds, should be able to remove the president of the central bank. This mechanism is one way to make the members of the central bank at least weakly accountable, without undermining the virtues of independence that we deemed necessary to avoid certain political failures associated with a government that is unconstrained in its monetary policy.

Therefore, we should avoid construing central bank independence in binary terms, so that central banks either are, or are not, independent. As mentioned at the beginning, this way of construing the possibilities confronts us with a dilemma. Either we must endorse the instrumental case for central bank independence-- namely, that it is necessary for economic efficiency-- or we must reject that argument for the sake of procedural legitimacy and economic justice. We should instead construe central bank independence in scalar terms, so that independence admits of degree. This helps us to avoid the dilemma by allowing us to develop an account of independence in which central banks can retain their independence to the extent necessary for economic efficiency, while meeting reasonable concerns regarding legitimacy and economic and climate justice. However, as mentioned earlier, we might have to face trade-offs: the Federal Reserve Bank in the United States, for example, already accepts that it has to make trade-offs to maximize employment at the cost of optimal inflation. The role of central banks has changed over time, as have their relations with the government (Goodhart 2010, pp. 1–14; Tucker 2018, pp. 27–47). Now, we should construct a degree of independence that allows us to promote political legitimacy and distributive (and climate) justice, while still meeting the primary goals of price stability and financial stability.

Instead of assigning the government’s full range of distributional goals to the central bank, as suggested by the coextensive view, here the distinction between the means and the ends of central bank policy is instructive. The institutional design of independent central banks should allow the government to constrain the means which the central bank has at its disposal to achieve price and financial stability. The constraint on the delegation of powers to the central bank should have its source in a law passed by the legislature. This law might, for example, regulate, or even forbid, some kinds of massive purchases undertaken by the central bank, e.g. of bonds of companies that have a negative impact on global warming. The law could also create an ethics committee to oversee such purchases and propose new prohibitions and regulations to parliament. An institutional design of this kind-- i.e. one in which governments can constrain the means used by the central bank-- reduces the extent to which important decision-making is delegated from elected representatives to unelected central bank officials, and this should diminish the concern that a proceduralist would have with central bank independence.

Some might claim that I was too quick to say that instrumentalists can easily accept central bank independence and the OEC. Taking into account that monetary policy has unintended distributive effects, Dietsch et al. (2018) argue that the central bank should no longer have a narrow mandate exclusively focused on price and financial stability. They argue that the neutrality of monetary policy is no longer defensible. Central banks should instead take on distributive responsibilities when pursuing price and financial stability, and thus aim for policy coordination with other agencies of government, mainly the fiscal authority, in order to limit inequalities exacerbated by unconventional monetary policy, so as to promote the lifetime expectations of primary goods for the least advantaged.

Nevertheless, I have admitted that recent concerns about central banks and economic justice arise because unelected central bankers can choose between unlimited policies that have profound and distinct distributional impacts on citizens. Some instrumentalists might regard these effects as unjust, and because they are produced by an independent institution this exacerbates worries not merely about justice, but also about the institution’s legitimacy. That was the point of the CDCB in claiming that the central bank should take on distributive responsibilities and follow goals coextensive with those of the government.

Defenders of the OEC may point out another objection to the CDCB. Some might argue that if we allow the government to take control over monetary policy in order to apply sound principles of justice, or to choose more egalitarian unconventional monetary policy to stimulate the economy, then we might then lose the good effects of independent central banks and fall prey again to the time inconsistency problems that distort future investment. The alternative I defend is for the government to include societal and distributional goals in the mandate of the independent central bank, such as protecting citizens from unemployment and other forms of absolute deprivation, or promoting long and stable demand for green bonds.Footnote 5 The view is also scalar regarding the distributive effects of central banking. Once the central bank has secured sufficiency, the government can choose between dividing the labor between a fully independent central bank only in charge of reaching the Pareto efficiency frontier, and a government securing social justice, or instead, a less independent bank that promotes other values coextensive to those of the government even if they do not maximize the prospects of the least advantaged. I claim this view lies between the OEC and the CDCB, reconciling the good effects of central bank independence with our concerns about distributive justice and the political legitimacy of this institution.

Conclusion

My purpose in this article has been to reconcile Elster’s classical justification of independence with more recent concerns about the distributional consequences of monetary policy and the political legitimacy of central banks. I claimed that the OEC for central bank independence is compatible with legitimacy and economic justice if we adopt a purely instrumental view on the one hand, and a maximinizing view on the other. I also claimed that according to the CDCB, the institution’s claim to legitimacy is made on procedural grounds, and its claim to economic justice depends on non-maximinizing views of distributive justice. Different times require distinct institutional designs for the central bank, and after the Global Financial Crisis, and given the climate emergency too, we should be able to constrain the financial means available to the central bank for reasons of economic and climate justice when this institution pursues its main goals of price and financial stability. This in turn helps us to avoid compromising the legitimacy of this institution from a procedural point of view when the government delegates very important decisions to an independent body, which is not easily removable by the legislature, and is not subject to reelection.

If we think of independence as a feature of the institution that admits of degree, then such requirements can be met and there is no reason to think that the central bank undermines the government’s political activity; meanwhile, democracy continues to have authority and create duties of obedience for its subjects. Under these conditions, it is unlikely for independence to produce bad effects on the government’s activity, hence any reason to think that it jeopardizes legitimacy seems weak. The main conclusion is that independent central banks are not necessarily illegitimate; rather they can be legitimate if their institutional design accommodates current concerns about economic and climate justice.