Select Committee on Monetary Policy Committee of the Bank of England Report


CHAPTER 4: THE TECHNIQUES AND TRANSMISSION MECHANISM OF MONETARY POLICY

4.1  Any analysis of monetary policy in general and the work of the MPC in particular must depend on gaining an understanding of how the policy instrument and the target are connected, whether the monetary authority can control its instrument sufficiently to achieve its objectives, and if the broader costs of doing so are acceptable. In the past it is clear that these criteria have not in general been met. The monetary authorities in the United Kingdom were unable to sustain control over their chosen intermediate target. The new method of monetary control by the MPC is based on using a short-term rate of interest, the repo rate[16], to achieve a final target, the inflation rate norm which lies within a symmetrical band, instead of an intermediate target. (We are aware that different techniques of monetary control are used in other countries. They are worth examining, but we have not pursued that topic in this Report.)

4.2  The Monetary Policy Committee has published an analysis of the transmission mechanism of monetary policy in its special report of April 1999[17]. It will be seen that our attempt at clarification, particularly of the role of the exchange rate and an estimate of the relative importance of the two main channels, the domestic and the exchange rate channels raises a number of important questions. The reason why we must look into these is partly as a response to the concern of many witnesses that the exchange rate is a more important channel of monetary policy than the Monetary Policy Committee has admitted. It is also an attempt to clear up the conflicting interpretations in the Minutes of the Monetary Policy Committee of May and June 1999 about how the interest parity condition should be used to help determine exchange rates.

4.3  We take it as given that some attention must be paid to the detailed connection between the instrument and the target. To rely on a black box approach, and not look at the mechanism has severe disadvantages. If something untoward happens, i.e. the instrument ceases to work or works in a perverse way, it will be impossible to know how to respond. In addition, if we do not know precisely what the instrument does, and how it works in varying circumstances, we shall not be able to assess accurately the contribution that monetary policy actually makes to the control of inflation.

4.4  We also take as given that there are several forces at work helping to determine the rate of inflation especially in the short to medium term. There is an obvious need to disentangle these forces, not merely to measure the net contribution of policy, but also to identify conditions in which an interest rate response is not what is required. To reiterate what has already been said and confirmed by witnesses, the present form of monetary policy is best adapted to demand shocks, and works via aggregate demand in the economy. They are not the only shock to which our economy is subject.

The background to current monetary policy

4.5  The broad aim of monetary policy is to provide a nominal anchor. This can be given a general interpretation such as price stability, as in the Section 11 the Bank of England Act 1998. It can also be given a more specific interpretation, such as that in the further details provided in Treasury's remit paper to the Bank of England, of achieving a particular inflation target.

4.6  Whatever the nominal anchor, in the long run the nominal interest rate will equal the inflation rate plus the real interest rate together with a risk premium in the case of long rates. The real interest rate is determined in the very long run by the forces of productivity.

4.7  The emphasis here is on the long run. What happens in the short run is immensely more complicated because the real interest rate and the risk premium are then highly variable. In addition, these long-run considerations are well set out in economic theory, and are supported by empirical evidence, but by their nature are not easy to prove. They do, however, provide a suitable logical structure within which to proceed further.

4.8  In the long run, the nominal exchange rate will depend on the difference between the domestic and the relevant foreign inflation rates. A higher domestic inflation rate will cause the exchange rate to depreciate by the amount of the difference. A nominal exchange rate target will therefore result in a domestic inflation rate equal to the foreign inflation rate. The domestic money supply will accommodate to this, and the domestic interest rate will equal the foreign interest rate.

4.9  Economies that are relatively closed, such as the United States, and now the European Union, have tended to prefer money growth targeting to an exchange rate target with the exchange rate being allowed to find its own level. (The ECB does, of course, use a short-term interest rate as its intermediate instrument.) An exchange rate target has been much more popular for small open economies in order to stabilize the terms of trade with their main foreign market.

4.10  The United Kingdom has at different times targeted the money supply, the exchange rate, and inflation. More than anything else, it is the failures of the first two forms of targeting that led to the adoption of inflation targeting. It has proved difficult to achieve the chosen money growth target due to an unstable money velocity brought about by structural changes to money demand, and a high degree of substitutability between different money aggregates. A monetary aggregate such as M4 is thus only loosely connected to inflation in the short run. The Governor's assessment of this episode is that "we missed the monetary target quite frequently, in fact we very rarely hit it" (Q 61).

4.11  The exchange rate has been proved even more difficult to keep on target. It is very volatile and subject to many unpredictable influences. High international capital mobility, not present under Bretton Woods, has meant that exchange reserves are unable to smooth out these fluctuations, and that interest rates have had to bear the burden. In the process the exchange rate volatility is transmitted to interest rates. Large short-term movements in interest rates have therefore been required from time to time in countries that have tried to peg exchange rates. In the United Kingdom, in September 1992, the cost to the economy of the interest rate increases that were needed to keep the pound in the ERM was judged to be too high, and sterling was withdrawn. It is against this background that inflation targeting has been adopted as the monetary policy.

Inflation targeting

4.12  Although, in principle, this can be implemented using monetary aggregates as the instrument, the discussion here is confined to the use of a short-term official interest rate. The Bank currently uses the repo rate, the rate at which the Bank is willing temporarily to repurchase eligible securities to provide liquidity to the monetary system. In his evidence to the Committee, the Governor has stated on more than one occasion that the Monetary Policy Committee seeks to control inflation by using interest rates to manage aggregate demand (Q 52). The report on the transmission mechanism by the Monetary Policy Committee identifies two main channels through which interest rates affect inflation: domestic demand and net external demand. The report does not mention supply-side effects as a cause of inflation. The judgement of the report is that temporarily raising the official rate by one percentage point for one year might be expected to lower output by something of the order of 0.2 per cent to 0.35 per cent after one year, and to reduce inflation by around 0.2 to 0.4 percentage points a year or so after that.

4.13  The argument lying behind this view is that when demand exceeds supply, prices are raised. This suppresses demand and induces additional supply. The implication is that by managing demand, inflation can be controlled. This ignores the fact that inflation, albeit rarely, may be a supply-side phenomenon and can occur even when demand is in balance with supply due to imported inflation. The experience of the oil price shock of 1973 suggests demand management on its own may not fully be the correct response to supply shocks.

The transmission mechanism

(a)  The domestic channel

4.14  Interest rates are an asset price. A change in the short-term official rate will be transmitted to all other asset prices: bank deposit and credit rates, bond yields, equity yields and the exchange rate etc. An increase in short rates will shift the whole term structure causing a rise in long rates. The increase in bond yields is brought about by a fall in bond prices causing a capital loss to bond holders, and an increase in the cost of issuing corporate debt. Equity prices may be expected to fall because the rise in the rate at which expected future dividends are discounted will reduce their present value. This raises the cost of equity finance.

4.15  Higher interest rates reduce the demand for goods and services by households and firms by:

  1. inducing households to defer consumption and save more (the inter-temporal substitution effect);
  2. raising the cost of new and old credit. This deters new borrowing and increases the interest servicing costs of outstanding liabilities such as mortgage payments and other loans (the cost of credit effect) thereby cutting the income available for new purchases;
  3. reducing the net wealth of asset holders due to the fall in bond and equity prices.

Virtually the only offsetting item is the benefit to savings in deposit accounts which will increase in value. This could have a positive effect on demand.

4.16  Several factors affect the size and timing of these effects, and hence their relative importance. The size will be affected by whether the interest rate change is thought to be temporary or permanent. Financial decisions are forward-looking, and what is expected to happen in the future may dominate what is happening at present. This is why the Fed seeks to create a climate of expectations about future interest rates and speaks about a bias one way or the other. What is clear is that uncertainty about future rates is damaging to economic decision making. An interest rate change will have much more impact if it is thought long-lasting than if it is temporary. If an interest rate change or differential is expected to be temporary, then the inter-temporal substitution and wealth effects of interest rates on expenditures will probably be small. Consumers are more likely to smooth expenditures in the short to medium term by temporary borrowing by drawing on financial assets. The main impact on consumption of a temporary change in interest rates is likely to arise from high floating rate liabilities which are difficult to unwind in the short term such as mortgage debt, and corporate borrowing.

4.17  In assessing the timing of the effects of an interest-rate change, a distinction can be made between first-round and second or later round effects. The demand effects mentioned above arise in the first round. In the second round, lower demand will lead to cuts in output and hence employment. This will cause a decline in income and lead to a further fall in demand.

4.18  We have set out the way this channel works in terms of a rise in interest rates. In a simple world, a cut in rates would have exactly the same effect in reverse. But the world is not simple and depending on circumstances, for example recently in Japan, interest rate cuts can be slower to stimulate the economy and, even when thought to be long-lasting, work less powerfully. (Again, to revert to the black box analogy, it is not the case that the instrument has a precise and well-determined impact on the target. Even the experts are sometimes surprised by the outcomes!)

(b) The exchange rate channel

4.19  We must raise a very important point at the outset. A good deal of economic theory, and much of the evidence we have received, emphasises the role of the exchange rate in monetary policy. At the same time we have also received evidence on how difficult it is to predict and explain exchange rate movements, not least the recent behaviour of sterling. It is very hard, therefore, to avoid the contradiction of simultaneously saying "this change in policy affects the exchange rate", and "we do not know what determines the exchange rate". Much of the theory on which we rely is predicted on the assumption that we do know how exchange rates are determined.

4.20  Having said that we note that while the Bank's report on the transmission mechanism does not emphasise its role, arguably, the exchange rate is the most important channel for monetary policy in the short run. It has a direct effect on inflation through its supply-side impact on import prices, as well as a demand effect arising from exports. The main disadvantage of the exchange rate channel is that it impacts disproportionately on the foreign trade sector. In his evidence to this Committee, the Governor has discounted the use of the exchange rate channel, especially recently, on the grounds that sterling's strength is due primarily to exogenous world factors, and not interest rate policy. Nevertheless, as its Minutes make clear, the Monetary Policy Committee looks very closely at the effects of the exchange rate on inflation, and of the consequences of interest rate changes on the exchange rate.

4.21  The evidence we have been given, not least by the Bank, distinguishes domestically generated inflation from what is generated by foreign sources. An increase in world inflation due to a supply shock would directly affect domestic inflation via import prices. It might also result in an appreciation of the exchange rate. In this special case, of course, excess demand is not the cause of the increase in inflation. An increase in interest rates would not, therefore, be the correct policy response as it would lead to a further appreciation of the exchange rate, and hence contraction of the economy.

4.22  An increase in world inflation may also be due to a world demand shock. Here the correct response would be to increase interest rates as the aim is to create an appreciation to partly offset the increase in export demand. It would also reduce total domestic demand enabling a transfer of resources to meet the extra foreign demand.

4.23  This illustrates some of the complexities in choosing the appropriate policy response to a foreign shock. It shows the important role of the exchange rate, and it highlights the danger of treating inflation solely, rather than mainly, as a demand-side phenomenon. It remains to be seen how the MPC would act in these circumstances.

4.24  In its wording, the Act makes no distinction between domestic and foreign generated inflation. Policy is to be directed at inflation (called price stability). To gear the interest rate solely, or largely, to so-called domestic price forces is a step which needs much more scrutiny than we have been able to give it so far, and must be yet another topic for the future.

THE EXCHANGE RATE AND INTEREST RATES

4.25  We are now obliged to enter another rather technical part of the problem. Our excuse for doing this is the need to clarify the role of the exchange rate. Although in the long run the exchange rate may be determined by the current account of the balance of payments, the removal of capital controls has meant that in the short run it is determined almost entirely by the capital account and the search for high returns. If domestic are higher than foreign interest rates then there will be a capital inflow which will cause an exchange rate appreciation. A more detailed description of the process is given in the appendix on the uncovered interest parity condition[18].

4.26  In practice, the connection between the exchange rate and interest rates is considerably more complicated, and less predictable than this broad description suggests. This is because capital flows are affected more by expected future interest differentials than by the current differential. Anything likely to influence future interest rates may, therefore, affect the current exchange rate. At times they may even prove a more powerful influence than a current interest rate change, and may cause the exchange rate to move in what appears to be a perverse direction. However strong these other factors, this does not mean that an interest rate change will have no effect on the exchange rate. This could be large, but not large enough to offset fully other factors.

4.27  The presence of exogenous factors is one of the main reasons why it is so difficult to predict the behaviour of the exchange rate. In his evidence to this Committee the Governor commented at length on the various foreign factors that might have affected the exchange rate apart from interest rates. For example, he said that "for reasons which were difficult to understand in terms of relative monetary conditions, we saw a sudden change in the exchange rate in the autumn of 1996, not against the dollar—we kept pretty good pace with the dollar—but against the core European currencies." (Q 91). He also commented on sterling's more recent strength against the euro in terms of "external influences" (Q 93).


16   The repo rate is defined later in this Chapter (at paragraphs 4.12 and 4.39). Back

17   The Transmission Mechanism of Monetary Policy, Bank of England, 1999. Back

18   Appendix 4. Back


 
previous page contents next page

House of Lords home page Parliament home page House of Commons home page search page enquiries

© Parliamentary copyright 1999