mpc.theyserveforyou.com

mpc:

MINUTES OF THE MONETARY POLICY COMMITTEE MEETING HELDON 8-9 DECEMBER 1999

1 Before turning to the immediate policy decision, the Committee discussed demand and output;money and asset prices; prices and costs; the world economy; the labour market; and otherconsiderations relevant to the decision.

Demand and output

2 Final domestic demand in Q3 had turned out as expected at the time of the November InflationReport. However, consumption had been a little stronger than expected, while investment was alittle weaker. On the basis of the pattern of revisions over recent years, it was quite possible that theQ3 investment figures could be revised up. However, it was also possible that the weaker figurespartly reflected a `millennium pause' in IT-related investment spending. The slightlystronger-than-expected consumption numbers reflected the ONS' attempts to capture the changingseasonal pattern of car registrations. The figures for consumption growth in the first half of the yearcould therefore be revised down, reflecting the latest estimates of the seasonal factors, when theNational Accounts for Q3 were published. Abstracting from these seasonal effects, the underlyingpicture for consumption growth in Q3 appeared broadly in line with expectations at the time of theNovember Inflation Report. Looking beyond Q3, the latest evidence suggested that car registrationswere turning out quite weak ­ possibly reflecting expectations of reductions in car prices in the nearfuture, in which case car sales, and hence consumption, might subsequently pick up at some point. 3 The Committee discussed other recent indicators of household consumption. Retail sales growthhad remained strong at the beginning of the fourth quarter, though the three-month on three-monthgrowth rate for October was slightly lower than it had been a few months ago. The Bank's regionalAgents' contacts were reporting some strengthening of retail activity, but the picture seemedsomewhat weaker than the official data. There were several possible explanations. First, salesvalues were rising less quickly than volumes, since the retail sales deflator was falling, and thismight be affecting contacts' responses. Second, retail floor space was probably still rising ­ and thiswas reflected in the difference between total sales and like-for-like sales in the British RetailConsortium survey. Third, it was possible that the Agents' retail contacts were weighted moretowards those sectors that were losing market share, such as department stores, and less towardsdiscount stores.

mpc: 2 4 Consumer confidence, as measured by the GfK survey, had fallen in November but remainedabove its long-run average. The MORI survey had risen, but this series was usually more volatilethan the GfK measure. While the recent fall in the GfK survey might reflect monthly volatility, therises in interest rates since the beginning of September might have had some dampening effect onconfidence. The disaggregated responses to the individual questions in the survey showed thatrespondents thought that the general level of unemployment would rise. This seemed at odds withthe recent evidence that employment had continued to grow. It was possible that, althoughemployment remained high, the pace of job creation and job destruction was higher now than in thepast and that this had heightened individuals' feelings about their job insecurity in an existing job. Ifso, that uncertainty could in turn have a dampening effect on consumption growth. But lookingbeyond the latest movements in consumer confidence and retail sales, the fundamental determinantsof consumption such as housing and financial wealth and real incomes seemed to be strengthening. 5 Turning to net trade, both exports and imports had turned out a little stronger than expected at thetime of the November Inflation Report. The UK share of world export markets probably increased inQ3, but was much lower than it had been prior to the appreciation of sterling from August 1996. Theexport orders index in the latest CIPS survey had remained above 50 for the seventh successivemonth, while the CBI monthly trends survey balance for export orders was negative, but less so thana year ago. There were, however, some signs that export optimism had deteriorated in recentmonths, for example in the DHL survey. 6 There had also been slightly stronger-than-expected import growth in Q3, so that the contributionof net trade to GDP growth in Q3 was only slightly more positive than had been expected. Stronggrowth in imports could be linked to re-exporting, and it seemed stronger than was consistent withrecently observed domestic demand growth. 7 Looking forward, the central projection for activity in the published fan chart assumed that nettrade would make negative contributions to GDP growth in the forecast period. The net tradecontributions in the central projection were somewhat more negative than projected by the average ofother forecasters, and that largely reflected a steeper assumed sterling depreciation by otherforecasters. The sterling effective exchange rate had strengthened since the November Report andother things being equal that would tend to depress net exports relative to the projection made then. 8 The latest data for non-EU trade in October showed a marked fall in exports but September hadbeen strong. That evidence seemed broadly consistent with the projections so far. The Agents'special survey tried to shed light on the recent export puzzle. The picture that emerged was of veryconsiderable price and market pressures. This was particularly marked for exports to the EU, which

mpc: 3 was to be expected given the movements in sterling against the euro compared with the dollar.Overall, the export picture in the survey seemed weaker than that suggested by recent official data. 9 On output, the latest estimate of GDP growth in Q3 was still in line with the 0.9% first estimateavailable at the time of the November Report. The latest manufacturing data showed a slight upwardrevision to Q3, although growth in October had been lower than the average monthly rate in Q3. TheCBI Industrial Trends survey for November recorded a positive balance for output expectations,though less positive than in October. The latest construction survey indicators also seemed quitebuoyant, more so than the official data on new orders. The monthly CIPS services survey had showna marked increase in the activity index. This index looked particularly strong when compared withthe average since the survey began in 1996, which itself reflected a period of strong service sectoroutput growth. The latest CBI/Deloitte & Touche survey however pointed to some recent weakeningof consumer services volumes. There were some suggestions that the latest rise in the CIPS surveymight be millennium-related rather than a more sustained acceleration in service sector activity.There was however little evidence from the Bank's regional Agents of strong millennium-relatedeffects on demand or output, apart from the possible pause in IT investment noted above and someisolated examples of increasing inventories. The latest National Institute of Economic and SocialResearch projection of GDP growth in the three months to November was 0.8%, down slightly onthe outturn for October. The staff's latest survey-based estimate of GDP growth in Q4 was broadlyat the same rate as in the third quarter, but higher than in November's central projection. 10 Overall there was not much evidence, on balance, that demand and output were turning outdifferently from what had been expected in November. If anything, the risks were of slightlystronger outturns in Q4 than thought a month ago.

Money, credit and asset prices

11 Broad money growth, excluding Other Financial Corporations, remained fairly stable. The moststriking monetary indicator over the past month had been the marked increase in narrow money inNovember. Notes and coin were growing at the fastest annual rate since 1980, with a particularlyrapid rise in November. The seasonal adjustment updated the seasonal factors each month using athree-year window. This meant that one third of the rapid increase in November had already beenattributed as a seasonal effect ­ the headline increase was even larger than the seasonally adjustedfigures suggested. The staff had tried to make allowance for the possible effects of the millenniumand the Government's Winter Allowance, but it seemed likely to explain only part of the totalincrease in November, though there was great uncertainty over the magnitude of these effects. Anincrease in money balances ahead of the year-end might occur for several reasons. For example, itmight be precautionary on account of concerns about computer-related Y2K problems, or it might be

mpc: 4 to spend on millennium-related celebrations. Even in the latter case, the effects might only betemporary if spending was transferred from other months. It might have no effect on aggregatespending if it reflected a shift from other types of spending to more cash-based items. In any event,it seemed unclear why households should withdraw substantial balances in November, so far fromthe year-end. There was no anecdotal evidence to support a large effect. Although there was a linkbetween retail sales and narrow money, it had not been particularly close from month to monthrecently. There was, however, a risk that higher money balances would eventually feed through toconsumption, even if the balances were not spent immediately. There was a range of views on howmuch weight to attach to the risk of higher spending. For some, the risk was mitigated by the factthat the large monthly increase came just before the year end, when interpretation ofmonth-on-month changes was usually difficult and was made more so by any potentialmillennium-related effects. 12 Asset prices had risen in the UK and overseas. The FTSE 100 had risen by 5% over the pastmonth and the FTSE All-Share index was now more than 6% up, and over 9% above the fifteen dayaverage used as the starting assumption in the November Inflation Report projections. There hadbeen rises in Japan, continental Europe and the United States too. The FT Small-Capitalisation indexhad outperformed the FTSE 100 index over the past month. Some of the rise in share price indiceswas accounted for by high-tech IT stocks, but this did not account for the entire rise. Nor was it clearthat the rise in IT stocks should be excluded in any event, because it was necessary to look at theaggregate. However, fluctuations in these stocks might be contributing to greater volatility. The risein indices both here and abroad seemed broadly consistent with a general rise in market confidenceabout the strength of world activity. 13 For the UK the rise in equity wealth, coupled with the strength of borrowing for consumption,suggested a robust outlook for consumption growth. House prices had again been volatile, with theNationwide index recording a strong rise on the month, while the Halifax index had fallen. Annualpercentage rates of increase in house prices, and hence gross housing wealth, were in double digits,and many short-run growth rates were at least as fast as the annual rates. Activity indicators in thehousing market had been mixed over the month, but on balance showed activity broadly stable overthe past few months and stronger than at the beginning of the year. The staff's provisional estimateof the increase in mortgage equity withdrawal in the third quarter, though well below the levels ofthe late 1980s as a share of personal income, was quite marked, and appeared to be closely correlatedwith recent growth rates in household consumption spending.

mpc: 5

Prices and costs

14 The Bank's commodity price index had fallen in October, reflecting both the fall in oil prices anda fall in metals and food prices. However oil prices had risen again in November. Producer inputprices continued to reflect higher oil prices. At the retail level, oil prices had been reflected in higherinflation in continental Europe. But despite higher petrol prices in the UK too ­ contributing around0.5 percentage points to annual inflation in the year to October ­ inflation as measured by theharmonised consumer price index in the UK was now below the European average. This was not toosurprising given the past strengthening of sterling. 15 Retail price inflation on the targeted RPIX measure had risen to 2.2% in October. This was inline with expectations at the time of the Committee's November meeting. The latest CBIDistributive Trades survey balance for retail prices had turned negative in November, and was wellbelow its historical average, which might point to further weakness in retail prices. However, thehistorical average reflected periods of much higher inflation, so it was difficult to interpret the recentdata. Anecdotal evidence suggested wider discounting than in the past, even of some luxury brands.It was difficult to tell if the lower-than-expected outturns for some of the components of the retailprices index provided evidence of a larger-than-expected squeeze on retail margins than had beenassumed in the fan chart published in the November Inflation Report. It might simply reflect adifferent timing for the reduction in margins than had been assumed. Alternatively, it could be dueto other cost factors that had yet to be observed in the data or simply month-to-month volatility withno implications for the short-term profile of inflation. 16 The weakness of goods prices was clearly reflected in the retail sales deflator, which was basedon the same data. The CBI/Deloitte & Touche survey balance for prices of consumer services hadalso fallen recently, but this survey had only been running for a few quarters. However, the CIPSservices prices index had risen on the month. 17 The continuing pressure on retail margins did not seem to reflect weak demand, but other factorssuch as increased competition. If so, increased competition and reduction in retail margins mightspread along the supply chain. Even given the recent rise in input prices, it might be difficult formanufacturers to pass on higher costs, at least in the short term. It was possible that aggregatesupply had become more elastic for reasons other than a structural increase in competition in the UK,such as increased investment elsewhere by international companies and a corresponding increase inworld supply. More efficient and cheaper access to information about domestic and overseasproduction and prices, and associated scope for cost savings on inputs, might also have an effect onUK prices. For example, the more widespread use of the Internet might affect UK productivitygrowth and pricing behaviour. There was little evidence that this was, as yet, having a big effect,

mpc: 6 though its use could grow rapidly. In the first instance one might see greater effects on inter-firmtrade than at the retail level. One much quoted example of internet activity was in books, but recentmovements in prices looked small when compared with the impact of the ending of the Net BookAgreement some time ago. It was therefore important not automatically to assume that the effects ofthe Internet would be bigger than previous shocks. 18 Most of the evidence of weaker prices related to goods prices; indeed the gap between serviceand goods price inflation was as large as at any time since 1992. That gap was much larger thancould be explained by differences in trend productivity growth between the sectors. 19 Within the RPI there were large rises in a few components of the services index ­ notablyinsurance and foreign holidays. Some members thought it was useful to gauge the underlyingbehaviour of RPIX by looking at sub-indices which excluded volatile items, such as fuel prices, orpossibly other `exogenous' items such as seasonal foods and taxes. Other members thought thatsimilar arguments could be applied to changes in prices resulting from exchange rate movements;and that other than in the very near-term, when nominal rigidities existed, it was not useful toexclude items from the index. 20 There had been some recent announcements on water, electricity and gas pricing. Taken togetherthe effect on the short-term profile for retail prices was broadly in line with what had been assumedin the November Inflation Report projections.

The world economy

21 The likely pace of growth in the United States remained subject to various uncertainties. Thelatest stock market rise added weight to the factors underpinning strong consumption growth, but hadalso added to the risks of a sharper slowdown if equity prices were to fall. Inflation had risensomewhat over recent months, although not on the core measure. In any event, the US situationappeared to be rather different from the conjuncture of events in the UK, because of the muchstronger recent productivity growth there. 22 The Japanese GDP data had been heavily revised on release of the Q3 estimate. GDP wasrecorded to have fallen by 1% in the third quarter, but this followed an upward revision to the secondquarter. Overall the new data and revisions back to 1997 left the level of GDP broadly in line withwhat had been expected at the time of the November Inflation Report. The markets had seeminglytaken the view that as much weight should be placed on the upward revisions to the earlier data, ason the weaker Q3 growth number (which was subject to future revision). But if one placed more

mpc: 7 weight on the momentum implied by the recent numbers, then that would imply a less sanguine viewof prospects for the Japanese economy. 23 The news on activity in the rest of the industrial world, including the euro area, was either in linewith or slightly more positive than expected over the past month. There seemed little news inemerging markets over the past month that would materially affect the prospects for UK inflation. 24 Turning to the pattern of exchange rates, the recent German data had contributed to a slight risein the euro over the days leading up to the Committee's meeting, though over the past month as awhole the euro had weakened further against the dollar. Sterling's effective exchange rate had notchanged much since the time of the November meeting, but was around 1% above the fifteen-dayaverage used as a starting point for the projections. That strength was primarily against the euro,rather than against other currencies, reflecting more buoyant views of prospects for activity in the USand UK. Relative movements in yield curves over the period could account for some of sterling'sappreciation against the euro. Markets had now had a chance to adjust to the interest rate increasesin the United States, the euro area and the United Kingdom over the past few months. There hadbeen no immediate major effects on financial markets. Compared with some people's priorexpectations, market activity still seemed to be quite well sustained as the end of the yearapproached.

The labour market

25 The latest labour market data had by and large been as expected. There was, however, someconcern among some of the Bank's regional Agents' contacts about prospects for the coming payround. The Agents were also finding more widespread evidence of skill shortages, which had beenbuilding since the spring. But neither this nor any of the other survey evidence on skill shortages hadyet fed through to the observed earnings data. It was possible that the sentiment being picked up bythe Agents was rather more about a general desire for more highly trained and educated workersrather than a cyclical point about tightness in the labour market. 26 Some data presented by Bank staff on employment transition probabilities - flows out of variouscategories of inactivity - had highlighted a large number of people flowing from the category of `notwanting a job' into employment within a short period of time. At first sight this was rathersurprising, but might be explained by people seeing new job opportunities become available whichthey had not previously expected. More research was needed in this area, but so far it suggested thatthe pool of people who could move into employment was greater than might be thought based onmore traditional and more narrowly focused measures of potential labour supply. It might help toexplain why labour tightness had not fed through to earnings as fully as might have been expected.

mpc: 8 27 No new official whole economy productivity or unit labour cost data were yet available for Q3,but manufacturing productivity had shown signs of picking up and had consequently begun to reducethe rate of increase of unit labour costs in that sector. The latest GDP data suggested thatproductivity growth for the whole economy might have risen to around 2% at an annualised rate inthe third quarter, which would be close to its long run average. This was important, as the centralprojection in the November Report had assumed a recovery in whole economy productivity growthover the forecast period. The fall in profit margins could not go on forever, and required acontinuing intensification of competitive pressures. Unless there was an even stronger recovery inproductivity growth than had been assumed, earnings growth would need to turn out as low asassumed for the central projection of inflation to hold. Even then, unless productivity growthimproved further, inflation would tend to rise at some point in the future, as the short-run downwardprice effects - of lower margins and a strengthening exchange rate - wore off and the effect of theunderlying strength of demand began to dominate. 28 Members of the Committee reflected on the November central projection for real earningsgrowth. The earnings profile in the central projection was somewhat lower than the average ofoutside forecasts made for 2000, and given the evidence of skill shortages it was possible that therisks to this projection were more clearly on the upside. But some members retained doubts aboutthe interpretation of the data, and as yet there was no firm evidence that the outturn for earningsgrowth was diverging markedly from the central projection made in November. This question wouldneed to be re-examined in the context of the February forecast round.

Tactical considerations

29 There were several tactical considerations discussed this month. First, the Committee concludedthat there was no reason to change the view outlined in the minutes of the September meeting thatthe Y2K period should not constrain UK monetary policy setting. However, there was a relatedconsideration concerning the data. The uncertainties about the size of millennium-related effectsmight make it more than usually difficult to interpret the data relating to the period covering theyear-end; particular examples were for consumer spending, investment, money and earnings. Ineach case there would be difficulty in reaching a clear interpretation of the data immediately after theturn of the year. But it was always difficult to interpret the data for, say, retail sales aroundChristmas and the New Year. It was not clear that this fog would be that much thicker than usual,and it would not preclude making a change in interest rates in either December or January.

mpc: 9

The immediate policy decision

30 Some members thought that the balance of arguments was in favour of a rise in interest rates of25 basis points this month, but attached different weights to the various supporting factors. First, forthose members who had contemplated voting for a 50 basis point increase in interest rates inNovember (rather than 25), there did not need to be much news on the month to justify a further25 basis point rise in rates now. Second, and related to this, the latest data did on balance suggest thecentral projection for inflation would be a little higher than thought a month ago. The latest evidenceon activity was at least as strong as expected a month ago, and suggested that the pace of outputgrowth might not slow as envisaged in the central case. In particular, world demand now seemed tobe strengthening, and some of the influences on UK consumption growth - wealth, real incomes andborrowing - seemed to have remained strong or to have strengthened over the past month. On prices,there had been signs of a compression of retail margins, and sterling had strengthened which wouldput further downward pressure on the price level. But the oil price had risen further. Third, theposition of the UK economy looked very different from when the last reduction in interest rates hadbeen made in June. It was unclear whether the 50 basis point rise in rates since the summer would besufficient to slow the economy to a more sustainable pace, and a rise in rates now might helpfullyaffect expectations. Fourth, the projection for inflation made in November had an upward slope atthe two-year horizon. The passage of each month meant that other things being equal the projectionfor inflation would be higher two years ahead. So even if there were no news on the month therecould be a case for a rise in interest rates. Fifth, in the view of some members, the risks of making abig monetary policy mistake were greater on the upside, and stemmed from the potential for thetightening labour market to begin to feed through rapidly to wages and prices, which suggestedraising rates sooner rather than later. Overall, the case for raising rates this month, rather thandelaying, was not a matter of urgency, but on balance the evidence pointed to the need for a furtherrise in the repo rate of 25 basis points. 31 A second view was that there was no need for a change in interest rates this month. On this view,the news on demand and output was broadly as expected at the time of the Inflation Report, albeitwith the risks perhaps now slightly more on the upside. The news on labour market quantities wasalso slightly tighter than expected. However, the likely short-term profile for inflation was ifanything a little lower than thought a month ago. Indeed, the downward effects on prices fromreductions in utilities prices, removal of fuel and tobacco escalators and increasing competition couldwell be larger than had been built into the published central projection in November. Overall, neitherthe news on demand nor that on prices this month was conclusive. The data continued to highlightthe same short-run trade-off puzzle between output and prices discussed last month in the Report andthe minutes of the Committee's previous meeting. Looking forward, the prospects were still forrising inflation at the two-year horizon, so a further tightening would probably be needed at some

mpc: 10 stage if the economy continued along the path outlined in the published November central projection.In particular, given the latest news on labour market quantities and skill shortages, it was possiblethat the central projection for earnings growth in 2000 made at the time of the November InflationReport was now too low. But there would be more hard evidence on earnings fairly soon as manysettlements were agreed early in the New Year, and the risk of waiting another month or two wouldbe unlikely to give rise to a big policy mistake given the saucer-shaped projection for inflation.There was time to wait and see if signs of earnings and price pressures emerged. There wastherefore no need to change interest rates this month. 32 A third view also led to a conclusion that no change in interest rates was needed this month. Onthis view, the latest indicators were consistent with a slowdown in the rate of growth of domesticdemand and output through the fourth quarter of 1999 and beyond, and the short-term profile forinflation was at least as weak as in the central projection in the Inflation Report. The exchange ratehad appreciated compared with the central assumption, and was closest to the constant exchange rateprofile described in the November Inflation Report (see pages 48 and 58). A higher exchange ratewould put more downward pressure on prices than implied by the November central projection. Onthis view it was prudent to wait and gauge the impact of the two previous increases in interest ratessince the summer, and to see whether there would be a further intensification of competitivepressures. A rise in interest rates now would risk pushing inflation further below the target. 33 The Governor invited members to vote on the proposition that the Bank's repo rate be maintainedat 5.5%. Six members of the Committee (the Governor, David Clementi, Charles Goodhart,DeAnne Julius, Ian Plenderleith and Sushil Wadhwani) voted for the proposition. Mervyn King,Willem Buiter and John Vickers voted against, preferring a rise in rates of 25 basis points. 34 The following members of the Committee were present: Eddie George, GovernorMervyn King, Deputy Governor responsible for monetary policyDavid Clementi, Deputy Governor responsible for financial stabilityWillem BuiterCharles GoodhartDeAnne JuliusIan PlenderleithJohn VickersSushil Wadhwani Gus O'Donnell was also present as the Treasury representative.

mpc:

ANNEX: SUMMARY OF DATA PRESENTED BY BANK STAFF

A1 This Annex summarises the analysis presented by the Bank staff to the Monetary Policy Committee on 3 December, in advance of its meeting on 8-9 December 1999. At the start of the Committee meeting itself, members were made aware of information that had subsequently become available, and that information is included in this Annex.

I

The international environment

A2 Growth in the United States in Q3 had been stronger than expected, but GDP had fallen in Japan, and there had been a mixed picture in the euro area, with activity stronger in France than in Germany. The latest data from the US and Japan had suggested that growth in gross trade volumes had remained at relatively high levels, though might have fallen back slightly in recent months. Oil prices had risen again in November, reflecting stronger global demand prospects, high compliance rates with the agreement among OPEC members to limit supply, and lower oil stocks in OECD countries. World food prices had fallen. A3 US GDP growth in Q3 had been revised upwards to 1.4%, largely reflecting higher contributions from net trade and stockbuilding. Annual growth in industrial production had risen to 3.3% in October. Although the index of capacity utilisation in the National Association of Purchasing Managers' survey had been stable over the last year, the survey had been pointing to lengthening delivery times and a rising backlog of orders, consistent with a rather stronger outlook for activity. Consumer confidence had risen in November, on the back of strong income expectations. Unemployment had continued to fall, reaching 4.1% in October, and had remained unchanged in November. Producer prices had fallen by 0.1% in October, reflecting falls in food and energy prices. Annual consumer price inflation had remained unchanged in October, at 2.6%. A4 German GDP had risen by 0.7% in Q3, and growth rates in Q1 and Q2 had been revised upwards. French industrial and consumer confidence indicators had grown by considerably more

mpc: 2 than those in Germany in October and November. Annual growth in industrial production in the major three euro area economies had fallen slightly to 0.9% in September, reflecting growth of 3.1% in France, 0.3% in Germany and ­0.7% in Italy. But a strong rise in German manufacturing orders in October and a rise of industrial production of 1.4% in the year to October had suggested a possible future pick up in German activity. The euro area broad money measure M3 had risen by 6.0% in the year to October. Annual inflation in the euro area had risen slightly in October to 1.4%, mainly reflecting the rise in energy prices over the year. Survey measures had suggested that inflation was expected to rise further in coming months. A5 Japanese GDP had fallen by 1% in Q3, a steeper fall than market expectations, although the level of GDP in the previous quarter had been revised up substantially. Industrial production had fallen by 2.3% in the month to October, but it had still been 1.7% higher than a year ago, and overtime hours had risen in recent months. Inventories had continued to decline, falling by 8.8% in the year to October. The unemployment rate had remained unchanged at 4.6% in October. Base money growth had slowed to 5.5% in the year to October, but growth in broader measures had picked up slightly, and bank lending had continued to recover from the low in August. The consumer price index had fallen by 0.7% in the year to October, but excluding food had fallen by only 0.1%. The Economic Planning Authority had released details of a further ¥18 trillion economic revival package.

II

Monetary and financial cond itions

A6 Narrow money had grown strongly in November. After adjustment for seasonality and the introduction of the new 50p and £2 coins, notes and coin had increased by 1.4% on the month, and the twelve-month growth rate had risen to 8.9%, the highest since 1980. The usual practice of current updating of seasonal adjustment meant that part of the rise in the seasonally unadjusted data in November had been treated as a seasonal effect, depressing the seasonally adjusted figure relative to an alternative measure based on non-updated seasonal factors. Part of the pick up in November had reflected special effects, such as a temporary increase in cash holdings relating to government Winter Allowance payments. But even after adjusting for an estimate of the quantitative impact of these factors, narrow money growth had still been unusually strong, consistent with evidence from other sources of robust growth in retail sales.

mpc: 3 A7 The stock of M4 had risen by £4.2 billion (0.5%) in October. The twelve-month rate had picked up slightly to 3.0%. M4 lending (excluding securitisations) had grown by £7.6 billion (0.8%) in October, and the twelve-month rate had risen to 7.2%. Robust lending growth had been driven by the continuing strength in lending to households and a rise of some £3 billion in lending to other financial corporations (OFCs) on the month. A8 Households' M4 deposits had risen by £2.2 billion (0.4%) in October. M4 lending to households (excluding securitisations) had remained strong, rising by £4.1 billion (0.8%) in October. The twelve-month rate had picked up to 8.8%, the highest rate since 1991Q3, as net secured lending had continued to grow strongly. Within total lending to individuals, net secured lending had risen by £3.6 billion (0.8%) in October, and the twelve-month growth rate had increased to 7.9%. The value of loan approvals had also been strong (at £10.4 billion). October's net secured lending and loan approvals data, combined with the historically high level of the stock of loan approvals, were consistent with continued housing market strength. Total unsecured lending had also been strong, growing by £1.0 billion (0.9%) in October, but growth had weakened over the past year. According to provisional Bank estimates, mortgage equity withdrawal had risen sharply to £2.4 billion in Q3, leaving `total borrowing for consumption' (defined as mortgage equity withdrawal plus unsecured lending) at its highest level as a percentage of disposable income since 1990. A9 The M4 deposits of private non-financial corporations (PNFCs) had risen by £1.7 billion (1.4%) in October, and the twelve-month rate had risen sharply to 6.9%. M4 lending to PNFCs had grown more slowly, rising by £0.4 billion (0.2%) in October. But PNFCs' total borrowing, including foreign currency borrowing and capital market finance, had remained strong. A10 OFCs' M4 deposits had risen by £0.3 billion (0.2%) in October, although the twelve-month rate had remained very weak at -7.2%, reflecting the sharp rundown earlier in the year. OFCs' M4 borrowing had been much stronger in October, rising by £3.1 billion (1.6%) on a month earlier, and by 6.0% on a year earlier. A11 Short-term nominal forward rates at maturities up to five years had fallen following the 25 basis point rise in the repo rate and the publication of the Inflation Report in November. But market rates had risen again following the release of the MPC minutes, most markedly at

mpc: 4 maturities of around one year. The falls in long nominal rates seen in late October had continued in the first week of November. This had partly reflected a global bond rally, but the fall in UK gilt yields had been greater than in other countries. Corporate bond yields at comparable maturities had also fallen. Long real interest rates derived from the index-linked gilt market had shown a similar pattern to long nominal rates. Although market expectations of inflation in the short to medium term (derived from nominal and index-linked gilts) had fallen immediately after November's MPC meeting, the decline had largely been unwound later in the month. Survey measures of inflation expectations for 2000 had been broadly unchanged on the month. A12 Quoted retail rates data had suggested that the November rate rise had been almost fully passed through to standard variable mortgage rates. Fixed-rate mortgages had risen by less, reflecting movements in swap rates. A13 The FTSE All-Share index had risen by 6.2% since the November meeting. Small capitalisation stocks had outperformed the FTSE 100 and All-Share indices, rising by 12.2%. A14 The sterling effective exchange rate index had appreciated by 0.3% since the November MPC meeting, and by 0.8% compared with the starting point for the November Inflation Report projection. Over the month, sterling had appreciated against the euro, but had fallen against the dollar and yen. Monetary news had appeared to explain much of the move in the euro-sterling bilateral rate.

III

Demand and output

A15 Quarterly GDP growth at constant market prices had been unrevised, at 0.9%, in Q3. The annual growth rate had also been unchanged at 1.8%. Total industrial production had risen by 1.2% in Q3 and by 0.3% on a year earlier. Manufacturing output had grown by 1.0% in Q3, the first time since 1995 Q2 that manufacturing output growth had been higher than services, though the level of manufacturing output had still been 0.2% lower than a year earlier. Services output had grown by 0.9% in Q3, revised down from 1.0% in the first release, with annual growth at 2.4%. Construction output had grown by 0.5% in Q3.

mpc: 5 A16 The expenditure breakdown of GDP had shown domestic demand growing by 0.7% in Q3. Changes in inventories had contributed 0.2 percentage points to growth, so final domestic demand had grown by 0.5%. A17 Private consumption had grown by 0.6% in 1999 Q3, and annual growth had been 4.4%. This had been stronger than expected at the time of the November Inflation Report, partly because of an unexpectedly small fall in vehicle expenditure. The seasonal pattern of vehicle expenditure in 1999 had been affected by the change in registration dates for new cars. The ONS had included an estimate of the size of this seasonal shift in their estimate of total Q3 spending. But a similar adjustment had not yet been made to Q1 or Q2 in light of the latest data. Total new car registrations over the first eleven months of the year had been 1.8% lower than a year earlier. A complete breakdown of consumption growth had not yet been published, but retail sales had grown by 1.2% in Q3. Government consumption had risen by 0.5% in Q3, down from 0.6% in Q2. A18 Total investment had grown by 0.2% in Q3. Business investment had fallen by 1.3%, though it had remained 4.3% higher than a year earlier. Within this, manufacturing investment had fallen by 5.2% in Q3, and service sector investment had fallen by 0.6%. The fall in business investment had been puzzling, but the investment data had tended to be revised upwards in recent years. The gross operating surplus of corporations had risen by 2.3% in Q2, and the annual rate of decline had moderated to -5.0%. A19 Inventories had made a positive contribution to GDP growth in Q3, reflecting a slowdown in the pace of decline. Including the alignment adjustment, inventories had fallen by £0.7 billion in Q3, compared with a fall of £1.2 billion in Q2. The CBI monthly survey in November had reported that manufacturers still perceived their stocks to be more than adequate. A20 Net trade had contributed 0.2 percentage points to GDP growth in Q3, the second consecutive positive quarterly contribution. Total exports of goods and services had grown by 6.2%, and imports had grown by 4.9%. The rise in goods exports had been more broadly based than in Q2, with exports to non-EU countries growing by 10.1% and exports to the EU rising by 7.0%.

mpc: 6 A21 The Bank's regional Agents had conducted a survey of their contacts to help assess the nature of the export recovery. A sample of mostly manufacturing firms with export-related business had been selected. Of the 121 firms responding to the survey, slightly more than half had indicated that their export volumes had not grown over the past six months. Demand had been reported to be strongest in North America, followed by Asia and other markets. Europe remained the weakest market, but had been recovering slowly. A large number of exporters to Europe had noted that lower profit margins had been a factor affecting export volumes in the past six months. A22 On the assumption that the exchange rate remained unchanged, the expected growth in export volumes over the next six months had varied across regions. Around 40% of respondents exporting to Europe expected volumes to fall, with an additional 30% expecting growth to slow but remain positive. Those exporting to regions outside Europe had been more optimistic, with around three-quarters expecting positive export growth. A23 Turning to indicators of Q4 activity, manufacturing output had risen by 0.1% in October and retail sales volumes had risen by 0.5%. Retail sales volumes had grown by 1.2% in the three months to October compared with the previous three months, and by 4% compared with a year earlier. Growth in retail sales volumes had continued to exceed that of retail sales values. The CBI Distributive Trades survey had shown a balance of +24 respondents reporting higher activity in November, and further growth had been expected in December. The MORI measure of consumer confidence had risen to +4 in November, but the GfK confidence index (which was less volatile and had a bigger sample) had fallen to ­1.5. A24 Housing data had been mixed in November, with the Halifax house price index falling by 0.5%, but the Nationwide measure rising by 1.9%. Indicators of housing activity had also been mixed, but generally pointed to continued high levels of activity. Particulars delivered had risen by 4.1% in October and had been 17.4% up on a year earlier. But the seasonally adjusted net balance of respondents reporting an increase in net reservations compared to a year ago had fallen to +29 in the October House Builders' Federation survey, from +45 in September. The Royal Institute of Chartered Surveyors seasonally adjusted sales balance had remained high at +32. Private housing starts had risen by 4.5% in the three months to October compared to the previous three months, and by 9.0% compared to a year earlier.

mpc: 7 A25 The latest survey-based estimate produced by Bank staff had suggested GDP growth of 0.8%-0.9% in Q4. The estimate of three-month on three-month GDP growth produced by the National Institute of Economic and Social Research had been 0.9% in October and projected to be 0.8% in November. Turning to the individual surveys, the output expectations balance in the CBI Monthly Trends survey had fallen back a little to +6 in November from +12 in October. The total orders balance had been above its average at -16, but export order books had been at ­33, below their long run average. The October CBI Industrial Trends survey had shown that manufacturing investment intentions had weakened as the balance of expected capital expenditure on plant and machinery had fallen to ­11 in Q3 from ­8 in Q2, well below the long-run average. A26 The headline index of the Chartered Institute of Purchasing and Supply (CIPS) manufacturing survey had risen slightly to 54.2 in November. The output index had been 57.4, above the no-change value of 50 for the ninth consecutive month, and the highest level since February 1997. The survey had suggested rather less buoyant export growth than the official data, with the index at 53.1. Stocks of finished goods had fallen in November, after rising in October for the first time since July 1998. The CIPS services survey had strengthened to 59.5 in November, the highest level since June 1997, with business activity rising throughout the sector. The CIPS construction index had also been strong at 57.5 in November, with housing and commercial orders remaining robust. But construction new orders had fallen by 5% in the three months to September compared with the previous three months, and had been 16% below their level in the same period in 1998. A27 In its latest Pre-Budget Report, published on 9 November, HM Treasury had revised down its forecast of Public Sector Net Borrowing in 1999/00 by around £6 billion. Forecast receipts had been revised upwards, reflecting an improved cyclical outlook. Projections for total public sector spending out to 2001/02 had been little changed, although a £0.7 billion underspend in 1998/99 had been carried forward into 1999/00. A28 The Chancellor had announced that the automatic road fuel and tobacco duty escalators would no longer be applied: decisions on any real increases in these duties would be made by the Chancellor in subsequent Budgets. Removing the escalators from the fiscal projections had reduced projected receipts by £1 ¼ billion in 2000/01, rising to £6 billion in 2003/04.

mpc: 8 A29 The lower overall projections for borrowing had not reflected a tightening of the Government's desired fiscal stance, which would only be reassessed in the next Budget. Projections for the structural deficit over the next three years had been little changed. The Treasury had published a separate document prior to the Pre-Budget Report assessing the outlook for trend output growth over the next five years. This had suggested that trend growth could be 2 ½% per annum compared with the 2 ¼% per annum assumed in recent Treasury fiscal forecasts. Nevertheless, the fiscal forecasts in the Pre-Budget Report had continued to assume trend growth of 2 ¼% per annum, as would those in the next Budget.

IV

The labour market

A30 According to the Labour Force Survey (LFS), there had been a sharp increase in employment of 110,000 (0.4%) in Q3, following growth of 0.2% in Q2. After making an adjustment for the difference in average hours worked by full and part-time staff, employment growth in full-time equivalent terms had been the strongest for more than a year. Total hours worked had risen by 0.3% during the quarter. The proportion of the working-age population in employment had risen slightly in Q3, but had remained below the spring 1990 peak. A31 CIPS survey measures of employment growth had been little changed in November. The surveys suggested that manufacturing employment had again remained stable, services employment had continued to rise moderately, and growth in construction sector employment had remained robust. A32 Unemployment had continued to fall in Q3, by 38,000 on the LFS measure and by 57,000 on the claimant count measure. Almost the entire fall in the LFS measure had been accounted for by a reduction in the number of short-term unemployed (those unemployed for fewer than six months). The pace of decline in claimant count unemployment had appeared to slow in the most recent data (the claimant count had fallen by 11,000 during September and October, compared with a total of 54,000 in the preceding two months). However, the ONS were reviewing the seasonal adjustment of these data.

mpc: 9 A33 According to the Federation of Recruitment and Employment Services (FRES) survey of recruitment agencies, staff shortages had continued to intensify in November. The Bank's regional Agents had also reported evidence of rising shortages of both skilled and unskilled staff, particularly in the South. A34 The headline measure of annual growth in average earnings per head, a three-month moving average, had fallen from 4.9% in August to 4.7% in September. Within the total, the service sector headline measure had fallen by 0.4 percentage points to 5.0%, while the manufacturing measure had risen 0.3 percentage points to 4.0%. The Reward index had risen by 0.1 percentage points to 3.5% in October. According to the FRES survey, there had been little change in the rate of earnings growth of either permanent or temporary staff in November. A35 There had been little new information on settlements. The Bank's AEI-weighted twelve-month mean settlement had remained at 3.5% in October. Further details had become available on the proposed three-year deal at Ford. It had been reported that the first year would combine a pay increase of 4.0% with a cut in basic hours worked. This would be above the average rate of recent manufacturing settlements in the Bank's database. A36 Historically, much of the variation in nominal earnings growth had reflected changes in inflation expectations. The latest data had suggested that real average earnings growth, after adjustment for the surveyed inflation expectations of trade union leaders, might have levelled off at around 2%.

V

Prices

A37 The Bank's oil-inclusive commodity price index had fallen by 1.0% in October, taking the annual inflation rate from 13.2% to 12.9%. The monthly fall had reflected price falls in crude oil and all the other major components of the index. The non-oil index had fallen by 1.3% in October, but had risen by 1.5% on a year earlier. Metal prices had fallen for the first time since June, and domestic food price inflation had remained weak. A38 Seasonally adjusted manufacturing input prices had fallen by 0.1% in October, but the annual inflation rate had increased to 6.3% from 5.8%. The latest CIPS manufacturing survey

mpc: 10 input price index had risen to 57.1 in November, the highest for four years. Seasonally adjusted total output prices excluding excise duties (PPIY) had risen by 0.1% in October, to give an annual inflation rate of 0.9%. The seasonally adjusted CBI output price expectations balance had risen to ­7 in November, the highest since February 1998. In November's CIPS services survey, the average prices charged index had risen to 53.2, the highest for two and a half years. A39 Prices of exports and imports of goods had risen by 1.5% and 1.0% respectively in the three months to September. Stripping out the oil component, export prices had fallen by 0.7% and import prices had risen by 0.2% over the same period. A40 The annual rate of change of the GDP deflator had been 1.8% in Q3, the same as in Q2. Larger rises in the household expenditure and investment deflators had been offset by smaller increases in the government consumption deflator. The retail sales deflator had fallen by 0.9% in the year to October, the lowest inflation rate in this series on record. The stronger decline in recent months had reflected slower food price inflation. A41 RPIX inflation had risen by 0.1 percentage points to 2.2% in October. The gap between services and goods inflation had widened to its highest level since September 1992. RPIX service price inflation had risen to 3.9%, largely as a result of higher insurance premiums. RPIX goods price inflation had fallen to 0.4%, mainly reflecting lower food prices. In the latest CBI Distributive Trades survey, the balance of retailers' expected prices had fallen substantially to ­1, the lowest figure since the survey had begun in 1983.

VI

Reports by the Bank's regional Agents

A42 The Bank's regional Agents had reported a continued moderate recovery in retail sales, though not as strong as the official data had indicated. Strong demand had continued for computers and mobile phones, and sales of household and electrical goods had risen, reflecting stronger housing activity. But turnover in the clothing sector had remained weak, and demand for new and used car sales had been below expectations. It had been reported that customers might be delaying purchases because of expected price reductions in the New Year. Millennium-related stockbuilding had been only moderate, with the exception of food and drink,

mpc: 11 and pharmaceutical goods. Consumer demand for Millennium events had also been weaker than had earlier been expected by contacts. A43 Manufacturing investment had continued to be targeted at productivity improvements and opportunities for overseas production. There had been mixed reports regarding IT investment. Some firms had reported a pause ahead of the New Year, but it was widely expected that IT investment would pick-up strongly from January. Reports of increases in input prices had become more widespread, but manufacturers had continued to find it difficult to pass these price increases on, and margins had been squeezed further. This had been particularly true for export-orientated firms. Downward price pressure had continued in most parts of the retail sector. A44 There had been evidence of a further tightening in the labour market. Skill shortages had continued to increase, particularly in the southern regions of England. In many service sector industries, forthcoming settlements and other pay awards were expected to be higher than a year ago, reflecting skill shortages and greater profitability. But in other sectors, such as engineering, where profits had been relatively weak, settlements were expected to be rather lower than in services. It was widely expected that any upward pressures on wages would have to be absorbed by the employer.

VII

Market intelligence

A45 Market expectations of, and uncertainty over, future interest rates in the UK, the US and the euro area ­ as measured by the implied rates and implied volatilities on short term interest rate futures respectively ­ had fallen in early November, following the rises in official interest rates. The fall in implied volatilities had been greater than that usually seen following monetary policy announcements. A number of possible reasons for these declines had been suggested in the markets, including a view that interest rates may not be changed again until the New Year, and improved confidence that liquidity needs across the Millennium would be met. A46 Interest rates implied by UK short sterling futures had risen again later in the month. This had reflected renewed uncertainty about near-term rate prospects following the publication of the MPC minutes, reported rises in house prices, upward revisions to international growth forecasts,

mpc: 12 and a growing view that the Millennium date change need not deter the MPC from changing policy in December. Market prices, surveys and anecdote had suggested some expectation of a December rate rise, although it was seen in the market as less than an even chance. A47 Merger and acquisition news had been supportive for sterling at some points during the month. Changes in risk reversals traded in the foreign exchange options market had been small over the month, but had suggested that the markets perceived some risk that sterling would rise further against the euro, but fall against the dollar.

mpc: