Speech by Sushil Wadhwani*, Member of the Monetary Policy Committee
Delivered to the South East & East
Anglia Agency's Contacts in Norwich on 24 July 2001
(I am grateful to Joanne Cutler and Jennifer Greenslade for their help and advice. Kate Barker, Ed Nelson and Gus O'Donnell provided me with helpful comments on an earlier version. Of course, the views expressed in this speech are entirely personal, and do not reflect the views of either the MPC or the Bank of England. )
The Current Policy Conundrum
Ladies and Gentlemen
We meet today at a time of considerable uncertainty about the prospects for the world economy and the appropriate policy response. Several industrialists have asked the Monetary Policy Committee to lower interest rates in the near future, while some other commentators have suggested that the next move in interest rates should be upwards. I shall first review recent developments and discuss the "most likely" scenario. Then, I will turn to a discussion of the risks in the light of various imbalances.
1 RECENT DEVELOPMENTS AND THE "MOST LIKELY" SCENARIO
A significant fraction of the UK economy has weakened in recent months - for example, the CIPS Survey Indices for both private services and manufacturing are now weaker than at any time since early 1999(1), when sentiment was still weak following the Russian crisis in the Autumn of 1998. Of course, as is well recognised, not all sub-sectors of the economy are weak. The corresponding CIPS index for construction has risen in recent months and remains consistent with an above-average degree of growth. Similarly, the annual growth rate of retail sales has actually risen in recent months to remarkably robust levels even as the newsflow on the global economy has pointed to a significant weakening in many of our trading partners.
Overall, the published data and projections suggest that the growth rate of GDP has decelerated to a below-trend rate since the fourth quarter of 2000, with the last quarter (2001 Q2), perhaps, even weaker than the preceding two quarters.
As the surveys attest, sentiment in the corporate sector is quite depressed. Although some of the weakness of business investment in the first quarter may have been erratic, the combination of falling capacity utilisation, deteriorating profit expectations and an ongoing re-evaluation of the profitability of investments in the information and communications technology (ICT, hereafter) area is consistent with a subdued outlook. It is also likely that there is an ongoing inventory correction - in a special survey conducted for the June MPC, the Bank's Agents reported to us that a net balance of firms said that their stocks had been higher than expected in 2001 Q1, and that they planned to reduce them over the following six months. Corporate employment expectations have also turned down recently - for example, survey responses to employment questions in the CIPS(2) and REC(3) surveys have recently fallen below the neutral level for the first time since early 1999.
Turning to the global economy, forecasts for GDP growth in continental Europe appear to be in the process of being reduced further. It is perhaps worth recalling that the German economy was already weakening in the second half of last year even before it had to contend with the effects of growing global pessimism about the ICT sector and the US slowdown. The Japanese economy might already be in recession again, and that is even before we see the deflationary effects of the proposed structural reforms. The US economy has, of course, slowed significantly with growth in the last quarter being, perhaps, even weaker than the preceding quarter. Although some of the recent survey data has pointed to some improvement, sentiment is still at a relatively low level. Deteriorating profitability appears to have led to a significant reduction in capital spending. Importantly, it is historically rare for a recovery from an investment-led slowdown to be quick and sharp. Given the current expectations of a sharp rebound in profits growth, it is unlikely that the current level of equity prices in the US is compatible with the prospect of a longer-lasting period of sluggish growth. Of course, it remains possible that the prompt policy response that we have seen in the US leads to an atypically quick recovery from a starting-point of a significant investment overhang. We shall have to wait and see on that, but the ongoing labour market weakness is rather discouraging and the inventory correction in the hi-tech arena still appears to have some way to go. Meanwhile, it appears that the slowdown in the three largest economies is taking its toll elsewhere. For example, several emerging economies have weakened by rather more than was expected only recently. Overall, the rather indifferent outlook for the global economy does not bode well for a small, open economy like the UK.
Given the deteriorating international and corporate outlook, it is quite remarkable that we have seen a rise(4) in retail sales growth, consumer confidence and house price growth this year. It is plausible that a combination of falling unemployment, healthy real income and interest rate cuts has helped sustain consumer confidence so far, notwithstanding the fact that household wealth has fallen during 2001(5). If, though, the forward-looking employment surveys do accurately presage some decline in employment growth, we may well see some deceleration in consumption growth.
We have, in recent weeks, also seen a significant increase in the inflation expectations of financial market participants, though there has been no discernible effect on survey-based measures. Historically, the market-based measures of inflation expectations have been rather volatile, and their predictive record is uninspiring. Nevertheless, it would be a dereliction of duty if one did not enquire into the possibility that the markets' concerns are justified. It would appear that the markets were especially worried by the jump in measured RPIX inflation in May to 2.4%, a level that it remained at in June. However, a significant fraction of that increase in inflation is directly attributable to transient factors like food prices. Note that a combination of adverse weather conditions and the foot-and-mouth disease has led to a significant increase in food prices, with the current annual rate of seasonal food price inflation of 24% not having been exceeded since 1984. Relatedly, although RPIX inflation has risen by 0.6 pp since January, RPIX excluding food has risen by a more modest 0.2 pp. It is notable that, unlike the US and Europe, where current inflation is temporarily above target, recent energy and price shocks have only taken UK inflation to around target. It is worth noting that, if, this time next year, the level of food prices was unchanged, which is a conservative assumption since most knowledgeable observers expect prices to be lower, then, other things being equal, this factor alone will subtract around 0.7 pp from the RPIX inflation rate in June 2002.
More generally, it is difficult to see why inflationary concerns have intensified. Recently, oil and metals prices have fallen. Import prices have been weaker than expected, and survey-based measures of pricing intentions suggest a more benign outlook for producer prices in the US and Europe. In the UK, the CIPS Survey questions on average prices charged are consistent with price deflation in manufacturing, and barely changing prices in services. This diminution in pricing pressure should affect RPIX inflation over the next two years. By contrast, we have seen a notable increase in the survey-based measures of selling prices in the retail sector, in part because the degree of deflation a year ago was unsustainable for very long, and also because the recent buoyancy of retail sales growth has made it easier to raise prices. Hence, provided that retail sales growth does not accelerate, it is unlikely that the retail sector will contribute to a higher RPIX inflation outlook a year out.
Turning to the labour market, falling profitability and prospective falls in employment growth suggest declining wage growth over the next year or so. We have already seen some signs of bonuses making a smaller contribution and it is likely that some moderation in regular pay growth will follow. It is important that we do not project ahead recent out-turns for regular pay growth as it is a lagging indicator.
Overall, in terms of the "most likely" scenario for inflation over the next 12-18 months, a combination of falling capacity utilisation and employment growth, more benign pipeline pressures and a subdued external price outlook should deliver falling inflation. This is why I personally believe that in the absence of large, unexpected shocks (e.g. assuming that the exchange rate and the oil price are broadly constant), and making the conventional assumption that interest rates are unchanged, inflation is "most likely" to be "around 2%" a year from now, and likely to remain around that level for at least another year, i.e. we are "most likely" to undershoot the target over the most relevant policy horizon. However, given their erratic nature, I do not exclude the possibility that the recent food price shocks temporarily tip us over the 2.5% threshold over the next few months, but that, in any case, is less relevant to policy.
Of course, in formulating policy, one is not merely interested in the "most likely" forecast, but also the balance of risks around it.(6) Although we do not, of course, target "imbalances" in the economy, they may yet be relevant to policy because they might affect the balance of risks.
2 THE CURRENT IMBALANCES AND RISKS TO THE "MOST LIKELY" FORECAST
A THE EXCHANGE RATE
It is sometimes argued that because the consensus forecast is for sterling to decline significantly, this is something that we should build into our inflation forecast. Of course, relative to our current treatment of the exchange rate in the forecasting process, this would have the effect of producing a higher forecast of inflation, which, other things being equal, might be associated with higher interest rates. Often, the justification for the view that sterling might depreciate usually either involves valuation considerations (e.g. sterling is rather high in relation to the euro on purchasing power parity grounds), and/or the statement that the trade deficit might eventually become problematic.
I am rather uneasy about these arguments. First, consensus forecasts of the exchange rate seem, to me, to be a rather thin reed to base policy on. Indeed, a recent study(7) argued that, in recent years, if one were interested in forecasting five exchange rates, flipping a coin would generally have yielded superior results to using consensus forecasts! While, in my opinion, sterling is "overvalued", and likely to fall at some point, it is important to remind ourselves that this is a view that many have held for some time. The Bank of England's own record in terms of currency prediction has been similarly indifferent. For example, during a period when the sterling exchange rate moved up from 83 during the first quarter of 1996 to about 106 in November 1999,(8) at each stage of this appreciation amounting to around 27%, the Bank of England predicted a depreciation.
Note that some of the projected widening of the UK trade deficit is, in any case, occurring because we are forecast to grow faster than many of our trading partners in 2001. Hence, it is plausible that a large part of any prospective rise in the trade deficit will be a temporary, cyclical phenomenon, and one would expect the financial markets to recognise this factor when assessing the appropriate value of sterling's exchange rate.
Second, our exchange rate forecasts influence interest rates today and thereby potentially affect the exchange rate today. Hence, a belief that the exchange rate might depreciate significantly over the next two years does typically lead to higher interest rates today, which, other things being equal, leads to the sterling exchange rate being higher than it would otherwise be. This has the effect of potentially exacerbating the imbalances today, as a higher current value of the exchange rate leads to a higher trade deficit.
Relatedly, for many years, students of macroeconomics have, therefore, been taught(9) that a central bank which is interested in stable output and inflation should 'lean against the wind' of significant asset price movements if these disturbances do not reflect underlying economic fundamentals.
About a year ago, some academic work(10) that I was associated with showed that this basic insight was still valid in the context of a more general model of the economy currently used at the Bank of England, i.e. if one were interested in minimising the volatility of inflation around an inflation target, then, the correct policy response to a high, overvalued exchange rate is to keep interest rates a little lower than what might be necessary to achieve a 2½% inflation rate at a two-year time-horizon. This is primarily because the alternative of somewhat higher interest rates is likely to exacerbate and prolong the "overvaluation" of sterling, which would then increase the volatility of future inflation.
Another relevant consideration is whether one thinks that there is time to react to an exchange rate fall if and when it occurs. Recall that those who feel that we might need to increase interest rates pre-emptively because of the current strength of the exchange rate are implicitly recommending a strategy of continuing to undershoot the inflation target if sterling does not, in fact, fall. They believe that the economy would be better placed to absorb the inflationary shock associated with an exchange rate shock if we were already starting below target.
However, one problem with this argument is that the appropriate policy response to a fall in sterling will depend on why the exchange rate falls, therefore making it rather risky to be pre-emptive. In general, an exchange rate depreciation could lead to a change in relative import prices without necessarily increasing the overall inflation rate. If, for example, sterling falls because there is a questioning of the 'new economy' and global equity prices fall,(11) then, it is possible that, over the medium-term, the disinflationary effects associated with the further global economic weakness more than offsets the inflationary effects normally associated with a fall in sterling.
In any case, even if a hypothetical exchange rate fall were not associated with such a disinflationary shock, our remit explicitly permits us to allow one-off price-level shocks to take us away from target, as long as we do not accommodate any second-round effects. Therefore, there would be no reason for the MPC to overreact to a temporary overshoot of the target that might be caused by an exchange rate fall if and when it occurs - a more pre-emptive policy of keeping interest rates higher today may therefore be, both, unnecessary, and damaging in terms of its effects on output through a higher exchange rate today.
B IMBALANCES IN THE PRIVATE SECTOR
It is sometimes argued that a cut in interest rates now in order to prevent inflation undershooting the target on a two-year horizon could be problematic because a fall in the savings ratio could lead to a rise in indebtedness that would make the economy vulnerable to a "hard landing" at some future date. A potential trigger for such an outcome could be a rise in interest rates associated with a fall in the exchange rate. A variant of this argument has consumption growth acquiring such great momentum that it is then hard to stop. Recall that our remit is to hit the 2½% target at all times. Therefore, allowing inflation to modestly undershoot on a two-year horizon while hoping that a much larger deviation from target is avoided at some future date is an attempt to more effectively fulfil the remit.(12)
However, in the current conjuncture, it must be noted that household debt servicing ratios are not high by historical standards, and therefore not yet a great source of concern. Moreover, a conscious strategy of tilting policy to prevent consumption growth from accelerating, simultaneously has the effect of bearing down on corporate investment through, both, the direct interest rate channel and through the indirect exchange rate channel. This, in turn, has costs to the economy in terms of a lower level of future productivity and capacity, at least for a time. Hence, in a desire to avoid an "imbalance" in the form of indebtedness, one could be storing up another "imbalance" in the form of a level of capacity that might, at least in the short-run, be "too low".
At present, it is possible to envisage circumstances where the deterioration in company financial balances causes significant retrenchment by the corporate sector and/or further gloomy news from overseas actually leads UK consumers to re-evaluate future economic prospects. This could lead to a significant rise in the savings ratio and a pronounced slowdown in GDP growth. Of course, the existence of this risk does, other things being equal, actually favour a cut rather than a rise in interest rates now, illustrating that there is no clear-cut link between the "imbalances" in the household sector and interest rate policy.
Of course, I do not exclude the possibility that there would be circumstances in which it might be appropriate to respond to imbalances in the household sector. One could, for example, envisage a situation where it might be appropriate to increase interest rates to forestall a house price "bubble" from gaining momentum. I should re-emphasise that one would only be worrying about the bubble because of its implications for future inflation volatility, which is entirely consistent with our remit. In my judgement, taking the UK as a whole, we are not in that situation as yet, but one must remain vigilant. Even in that situation difficult judgements about whether a house price "bubble" is more important than an exchange rate "bubble" will have to be made.
C OTHER RISKS
I have focussed on the exchange rate and imbalances in the UK household sector today as they have attracted a great deal of attention recently. There are, of course, a variety of other risks associated with our inflation forecast, the most notable of which is that the global economic slowdown proves to be worse than that which we have assumed(13) and this continues to have an important effect on the policy judgements that I make.
CONCLUSIONS
I hope that I have communicated to you today some of the considerations that are currently affecting policy. These are turbulent times and we may have a bumpy ride ahead of us. I now look forward to hearing how you see things.
Footnotes
* and Visiting Professor at the City University Business School and the London School of Economics
3 i.e. The Recruitment and Employment Confederation survey.
7 Lehman Brothers' Global Weekly Economic Monitor, 15 June 2001.
8 We changed our forecast convention in November 1999.
