Last week I blogged on the seven elements in Brown's economic reforms. I suggested that five of them (the fiscal rules, the stripping of prudential regulation from the Bank of England, the Tripartite system, taper relief for capital gains, and surplus ACT) had either collapsed or were now seen as negative. I argued that the 1997 inflation target (the one thing I endorsed) was collapsing because of the government's failure to operate it properly, and promised to argue against the final piece: independence of the Bank of England. Here goes.
Operational independence of the Bank of England
Most people were caught by surprise when the incoming Labour government made the Bank of England "operationally independent" in May 1997. They should not have been, really, since having an established independent central bank was a requirement of euro Membership, and since Labour and the Conservatives were committed to keeping their options open on euro Membership they would both have wanted to make the Bank of England independent at this point.
There are various forms of central bank independence. New Labour's 1997 scheme was what is called "operational independence". That meant that the Chancellor set the goals of monetary policy - the Bank of England Act specified "price stability" with the government setting out what that means at least once every twelve months; at present the Chancellor has set a target for CPI inflation of 2% - whilst the Bank of England's Monetary Policy Committee (MPC) decides how best to go about meeting the target (i.e. the MPC sets interest rates). An alternative would be "goal independence" (like the Federal Reserve or the European Central Bank) where the independent central bank sets the goal of its policies as well as deciding how to get there.
Because of the connection with euro membership, many Conservative MPs opposed Bank of England independence when it was introduced. John Redwood offered a different position: that it is impossible to have an independent central bank in a truly democratic society. A few others (mainly Europhiles) were in favour from the start. Francis Maude signalled a retreat from this opposition in 1999, and once Michael Portillo became Shadow Chancellor, he quickly announced that Conservatives now supported Bank of England independence. Since Portillo's time, the Conservative position has been that Bank of England independence needs to be strengthened, that under the Conservatives the Bank of England would be more independent, not less.
Arguments for and against central bank independence
Most economists imbibe a set of arguments for central bank independence almost with their mother's milk. Most of them agree with Friedman that operational independence is best. However, a significant number of monetary economists - those most expert in the matter - do not agree with central bank independence.
First I'll sketch four arguments commonly offered, and why I consider them wrong in the case of Britain. Then, to supplement my attacks on the arguments in favour of central bank independence, I'll offer four arguments the other way. My conclusion will be that independent central banks are good for some countries and may even have been good during the Labour government, but should not be thought good under a Conservative administration in Britain.
Four arguments in favour of central bank independence
1) Politicians are not experts in monetary policymaking
First, and very naively, many people favour central bank independence on the grounds that politicians don't know much economics, whilst central bankers do. This is patently a terrible argument, since politicians don't know anything about running health services or schools or an army or a police force either, but they do know how to take advice. The less said about this argument, the better.
2) Central bankers have the true interests of the country at heart, whilst politicians just want votes
Second, a much better argument often offered is that politicians just want to get elected, whilst central bankers will be interested in running things well. This is straightforwardly the ancient argument for benign dictatorship versus democracy. It's a very good argument, but since the eighteenth century we've decided that we had ways to make use of democracy, whilst on the other side it has proved unfortunately difficult to choose benign dictators or to keep them benign once they became dictators.
3) Politicians lack credibility because they won't want to keep their monetary promises
Economists of a "blind-you-with-science" bent like to offer something called the "time consistency argument". This goes as follows. One argues that, in the short-term, there is a trade-off between unemployment (or output) and unexpected inflation. That is, if inflation is higher than economic agents expect, then the price rises they see will appear to involve real increases in value. When production becomes more valuable firms will produce more. This means that output will be higher and unemployment lower. This is a standard result of economic theory, and is not an area of dispute between the major economic schools.
Now, since governments are more popular when unemployment is lower and output is higher, that means that governments have an incentive to create unexpected inflation, if they can. But economic agents are not stupid. They are rational. Since government has an incentive to pursue unexpected inflation, economic agents will take account of this. This means that even the most prudent government will face an inflationary bias. That is, even if the government does not pursue secret inflationary policies, inflationary expectations will reflect the risk that it might. This inflationary bias will be reflected in a need for higher interest rates, and in higher inflation (since inflationary expectations tend to be self-fulfilling).
Since inflation and interest rates are raised if government can create inflation secretly the obvious solution is to take that power away from government. The most powerful influence on inflation in the short-term is monetary policy. Hence by taking monetary policy away from governments and giving it to an independent central bank (e.g. the Bank of England) one would expect to have lower inflationary expectations and to require lower interest rates to implement monetary policy.
The key problem with using this as an argument for central bank independence is that it applies just as much to independent central bankers as to politicians. Independent central bankers will always find that, in the short-term, it is better to let inflation go a little bit higher than to be tough and keep inflation down. Now the idea is that being independent central bankers they will take a longer-term perspective and so be prepared to be tougher in the short term, but there is really nothing intrinsic about being an independent central banker per se that equips them for that, beyond the benign dictatorship argument above. Sometimes it is suggested that there should be a contract between the government and the central bank in which the central bankers are paid not to care about unemployment, so that they will be more willing to let unemployment rise in the short term in order to keep inflation down. But if that is the scheme then the central bank is clearly not independent, since it is subject to instruction by the government, and the scheme relies on the government not backing out once matters get tough. Indeed, it becomes clear that the "independent central bank" is intended as what economists call a "commitment device" - a way to make a promise. But an alternative way to make a promise is to make one and stick to it, like Mrs Thatcher - to take a long-term view. And the Thatcher example illustrates that it is possible to elect politicians that keep promises.
Of course, in some countries at some times then the commitment possible through using an independent central bank may be more credible than the commitment through politician promises. But there is nothing intrinsically obvious about that - it is dependent very much on the specifics of the case. And it seems clear to me that Britain is a country in which politicians can make and stick to promises.
Connected to this, one might have considerable concerns as to whether, under pressure, an independent central bank is well placed to be tough. For example, in late 2007 it became fairly clear that in 2008 the UK's inflation target would be breached materially and for an extended period. It would in principal have been possible to raise interest rates markedly so as to avoid inflation rising so high above target. But the cost would have been several hundred thousands of rise in unemployment. All standard opinion has been that that was not worth doing. Has the fact that we had an operationally independent (not truly independent, but let's ignore that for the moment) - has our independent central bank meant that monetary policy was tightened aggressively so as to meet the inflation target, ignoring the consequences for unemployment and output? No. I suggest that one reason for that is that a decision that has the consequence of putting hundreds of thousands of people out of their jobs should almost always be a political decision - no-one else is properly placed to make such a choice.
4) Countries with independent central banks have lower inflation
Moving on, the arguments above suggest that it is an empirical question, a question of the practical outworkings of the policy, whether independent central banks are better commitment devices than politicial decision-making. And many readers here will know that a number of famous studies gave the result that countries with independent central banks tend to have materially lower inflation than countries with political monetary policy-setting. One's first response to this might be to think it gives us the answer: it works. Of course, as I mentioned above it might be the best policy in some countries but not others, but let us put that aside for a moment. Do these data really tell us that (in most countries, at least) independent central banks deliver us lower inflation? The first reason one might doubt this is to do with the direction of causality: is it really that independent central banks lead to lower inflation? or is it that lower inflation makes countries more likely to bring in independent central banks? or might there, perhaps, be some features of countries that lead to them having both low inflation and independent central banks?
Of these three options, the one thing we can reasonably for certain is that if it is the first then it is not because of any increased credibility, i.e. the time consistency argument (3) above does not work in practice. We know this because there have been many studies that show that, in practice, when reducing inflation, there is no less lost output or less increase in unemployment if one introduces an independent central bank. Credibility doesn't come through institutional fixes. You have to earn it by showing a willingness to be tough, even if you are an independent central bank. And, also (crucially) anyone can earn credibility - politicians become credible just quickly as independent central bankers.
So, I have argued above that none of the four key arguments for central bank independence is really convincing. The only one with anything to it is the idea that, at some times and places, an independent central bank might be a good commitment device. I'll give you an example of such a time and place: 1997 in the UK. In 1997 there was a fear that the Labour Party would, if elected, be soft on inflation, and its promises not to be soft were not fully believed by financial markets, with the consequence that there was an inflationary premium in long-term interest rates in the run-up to the 1997 election. Once the Bank of England's operational independence was announced, that premium fell away. Now in my view this was a smart and useful move by Gordon Brown. Because his promises would have lacked credibility, he used the commitment device of an operationally independent central bank. Good for him. But that does not mean that the Conservatives should think that their own promises would not be believed or that we must endorse the concept of Bank of England independence, let alone promise to strengthen it as we have done.
Most economists would offer arguments (3) and (4) as their basis for wanting central bank independence, though as self-believing technocrats their truer motivations are probably (1) and (2) - they trust other economists much more than they trust politicians.
Now I'll move on to offer some reasons I am positively against independent central banks, as opposed to merely not in favour of them.
A) An independent central bank prevents the co-ordination of economic policy
First, and most famously, the old argument was that monetary and fiscal policy would work best if they were working together. If fiscal and monetary policy are not co-ordinated, if they are not working together as part of some overall strategy, then economic policy as a whole will not be as effective. But, by definition, if monetary policy is run by an entirely separate and independent institution it cannot be co-ordinated with fiscal policy, or indeed with any other part of government economic policy, such as labour market policy or industrial policy. Fiscal and other economic policies have an effect on inflation, so not even inflation will have a co-ordinated policy.
B) An independent central bank removes the incentive for government to avoid inflationary policies
A second, closely related argument is that if the central bank is put in charge of keeping down inflation the government will consider that controlling inflation is someone else’s job. This means that the government can pursue expansionary fiscal policies, or other inflationary policies, without fearing that it will be blamed for the consequent inflation. The clearest example of this is the period of German unification after the fall of the Berlin Wall. German Unification and a large budget deficit were pursued by the government, while the Bundesbank was expected to keep down inflation. This has also been a factor in Gordon Brown's huge rise in public expenditure in recent years.
The consequence of this argument is that, given that many policies which cause inflation are popular apart from their inflationary consequences, since the consequent inflation can be blamed on someone else an inflationary bias is introduced into the system precisely by having an independent central bank. That is, having an independent central bank leaves the government an incentive to pursue inflationary policies.
C) An independent central bank involves a loss of democratic control
This is just the flip side of the benign dictatorship argument. Having an independent central bank removes the ability of government to trade off short-term output rises for long-term inflation. But this is a genuine political choice. I do not favour having higher output today in exchange for inflation tomorrow and the next day, but that is my political philosophy. Other people might prefer to be able to make this trade. Removing democratic control over monetary policy may prevent people from making short-sighted errors. But democracy is all about letting people make their mistakes for themselves.
Perhaps we could appoint a permanent panel of experts to run our economy for us without any of those foolish politicians messing things up. But the failings of democracy compared with rule by experts are well-known, and have been discussed for thousands of years. The fact is that in mature political societies we have decided that it is best if we can elect and reject our rulers, particularly those who have control over our money. Control of monetary policy carried out by central bankers that turn out not to be good at it, or who decide to pursue goals thought undesirable by everyone else, could be extremely damaging. We need to be able to exercise democratic constraint.
It is no good responding to this objection by observing that, for example, the British Chancellor could remove the current decision-makers at the Bank of England and replace them with someone else. That is true, but it merely means that the Bank of England is not totally independent of political control. Yet the argument presented in favour of independent central banks implied that the less political control there was, the better!
D) Central Banks are intrinsically conservative
These days appropriate monetary policy basically means appropriate interest rate policy. Appropriate interest rate policy depends on a number of things. Let us consider just one : unemployment. There are always some people changing jobs or retraining. Some unemployment is entirely natural. The amount of unemployment which is considered natural is called the "natural rate of unemployment" (NRU).
If current unemployment is above the NRU then that is a source of deflationary pressure. The extra unemployed people will be looking for jobs and thus bidding down wages, hence production costs, and hence prices. Roughly speaking, the higher deflationary pressures are, the lower interest rates will be.
In contrast, if unemployment is below the NRU that is a source of inflationary pressure. There are not enough people switching between jobs, and there is not enough retraining going on, so workers are scarce. Since they are scarce their wages will be bid up, meaning higher production costs and hence higher prices. Roughly speaking, the higher inflationary pressures are, the higher interest rates will be.
Suppose we could reduce the NRU. That would mean that for any given level of unemployment, we should have interest rates lower than they would be if the NRU had not changed. Next suppose the government pursues policies intended to reduce the NRU. The government claims these policies will reduce the NRU, but the political opposition claims that they will not, and opposes them. Clearly these policies are a source of political controversy.
Now suppose I am an independent central banker deciding interest rate policy. Clearly, since I am non-partisan, it would be inappropriate for me to assume that the government’s policies have reduced the NRU until there is compelling evidence to this effect. (If the government changes then it might become even more difficult. It would seem impossible for me to assume that the last government’s policies were successful when the electorate have rejected them and those who opposed the policies have been elected!)
This means that during a period in which the government might believe, rightly, that its policies have been successful in reducing the NRU, and hence that appropriate interest rates should be lower, an independent central bank will keep interest rates higher until the lower NRU has been proven. Since central banks inevitably lag the political debate in this way, an independent central bank’s interest rate policy will inevitably not reflect the possible effectiveness of controversial and unproven government policies.
This conservativeness of independent central banks means that they may often pursue inappropriate policies (or at least policies which run counter to political will of the people, expressed through their elected government), and is another reason why it is a bad idea to have an independent central bank. Politicians might legitimately disagree with the central bankers for reasons other than wanting to increase inflation.
The important group of monetary economists opposing central bank independence mainly rest on the rejection of arguments (3) and (4) set out above, and counter-arguments (A) and (C).
The collapse of the operational independence regime
Further to the arguments above, I want quickly to mention one other matter I have raised repeatedly over the past eighteen months. The UK's operational independence regime has largely collapsed in this period. The original idea was that the Chancellor set the goals of monetary policy and the Bank of England carried them out - by analogy with a general tasked with invading a certain country but then left to do it as he thought best. But over the past eighteen months, the Bank of England's Monetary Policy Committee has decided that it did not need to abide by the inflation target set by the Chancellor, and when it has gone outside its target it basically said that that didn't much matter and the Chancellor didn't do anything about it. The MPC has now said, in its latest inflation report, that it would be undesirable for it to try to get inflation to target in 2009. That is all very well, but under operational independence it wasn't supposed to be up to the MPC to decide what was or was not a desirable goal for its policy. Operational independence has largely collapsed. The Bank of England is now, essentially, goal-independent: it sets its own goals as well as decides how to get them. It is the general that, having invaded one country, decides that he might as well invade the neighbouring country whilst he's about it, since that would be in everyone's interest.
Conclusion
I have argued against the idea that, as a general truth, independent central banks are a good thing. I do not believe that the Conservative Party should endorse this view. Instead, we should regard operational independence of the Bank of England as having been (at least for a while) a useful device that New Labour used to protect its policymakers from its backbenchers.* That does not mean that we should do things the same way. If you want monetary policy credibility you have to earn it, and in the UK politicians can earn credibility just like anyone else.
* I used to tell my students concerning operational independence for the Bank of England that, like Mr Burns in the Simpsons episode in which he was talking about art, when it comes to monetary policy I know what I hate, and I don't hate this.
