Summary of Quarterly Bulletin
August 2000

Research work published by the Bank is intended to contribute to debate, and is not necessarily a statement of Bank policy.

Age structure and the UK unemployment rate (47k)
(by Richard Barwell, formerly of the Bank's Structural Economic Analysis Division).
Most models of the labour market assume a short-run trade-off between unemployment and inflation; this assumption is at the heart of the Phillips curve relationship and the expectations-augmented models that followed it. So recent developments in the labour market have puzzled economists: in August 1999, for example, the number of people out of work and claiming benefit fell to a 19-year low and yet the RPIX inflation rate was at its lowest level for more than five years.

Part of the explanation of this puzzle may be that the natural or equilibrium rate of unemployment has fallen, enabling the actual unemployment rate to fall substantially without generating a pick-up in inflation. Explanations for the fall in the natural rate have tended to focus on supply-side factors, such as the decline in union bargaining power, reduced generosity of unemployment benefits and increased deregulation of the labour market. This paper examines another supply-side explanation, which has received less attention in the United Kingdom: that the natural rate has fallen partly because of changes in the composition of the labour force. Youths tend to have higher unemployment rates than adults, and presumably have higher natural unemployment rates as well. The proportion of youths in the labour force almost halved over the past decade, so we would expect the aggregate unemployment rate and the natural rate to have fallen as a result.

Most of the existing literature investigating the impact of demographic change on the unemployment rate has looked at the US labour market. Katz and Krueger (1999) find that changing age structure accounts for about half of the fall in US unemployment between 1979 and 1998; Shimer (1998) finds that the effect is even larger, at about 70%. This paper provides a comparable estimate of the fall in UK unemployment that can be accounted for by the decline in the youth share of the labour force.

The first section presents two key stylised facts, which together suggest that demographic change could play a significant role in explaining recent developments in the UK labour market. First, that the youth share of the labour force has fallen dramatically over the last decade, and second, that youths always have higher unemployment rates than adults; the latter is analysed in the second section. The third section explains the 'shift-share' methodology developed in the literature, and uses it to provide a range of estimates of the impact of demographic change in the labour force on the unemployment rate. The fourth section discusses two alternative approaches that control for changes in the labour force participation rates of each age group.

Quantitatively, demographic pressures do indeed appear to explain part of the change in actual unemployment. Although this change is sensitive to the precise measure used, particularly the assumption made about the base year, it appears that about 0.55 percentage points, or 10%, of the fall in the unemployment rate between 1984 and 1998 can be accounted for by changes in the composition of the labour force. There is no robust evidence, however, that youths became less likely to become unemployed, through generational crowding effects, as their share of the labour force declined.

However, demographic pressures were not the only forces that affected the composition of the labour force over the period; changes in the participation rates of different age groups will also affect the unemployment rate. Controlling for these, demographic change explains less of the change in the unemployment rate over the period. However, it appears that the shift in the composition of the population caused by the baby boom and bust still explains about 0.45 percentage points of the fall in the unemployment rate over the period.

Finally, on the basis of current projections, it appears that future shifts in the composition of the labour force will have little effect on the unemployment rate over the next decade.

Financial market reactions to interest rate announcements and macroeconomic data releases (47k)
(by Andrew Clare and Roger Courtenay of the Bank's Monetary Instruments and Markets Division).
At 11.00 am on 6 May 1997 the new UK government announced that it had granted the Bank of England operational independence with respect to the implementation of monetary policy, subject to an RPIX inflation target of 2½% per year. This change was designed to improve the credibility and transparency of the monetary policy process. The Bank of England's aims were stated clearly, and the voting record and discussions of the members of the Monetary Policy Committee (MPC) were to be published shortly after interest rate decisions had been made.

Our aim in this study is to investigate whether there has been a systematic change since Bank independence in the way that market participants incorporate information from monetary policy announcements, and from other important macroeconomic data announcements, into financial prices. We use intra-day price data (rather than daily data, which are sometimes used in this context) because markets receive many different pieces of news throughout the trading day, and so the impact of a particular announcement may be obscured by using daily price series. We therefore concentrate on the minutes immediately preceding and following these announcements.

Our study uses high-frequency data on short and long-term LIFFE interest rate futures contracts, on the LIFFE FTSE 100 stock index futures contract, and on the dollar/sterling and Deutsche Mark/sterling exchange rates. We monitor the behaviour of these financial prices around the times of interest rate announcements and key macroeconomic data releases over two periods: from January 1994 to 6 May 1997 (pre Bank independence), and from 7 May 1997 to June 1999 (post Bank independence).

Our empirical results do not yield simple definitive conclusions about whether monetary policy is now better understood by financial market participants as a result of Bank independence. The total (cumulative) reaction of the LIFFE contracts and exchange rates to interest rate decisions appears either unchanged or lower in the post Bank independence period, depending on the market observed. This supports the idea that the news content of monetary policy announcements has fallen. However, while the total reaction supports this view, the immediate reaction in the first 5 minutes is larger in all of the markets studied here. With respect to interest rate decisions, it appears that the news contained in the decisions is incorporated into financial prices more quickly than in the pre Bank independence era. One possible explanation for this is that pre-positioning in the financial markets ahead of the decision has become more sophisticated since Bank independence, with the publication of a clear, unambiguous timetable for the announcement of interest rate decisions.

Looking at exchange rate responses, there is evidence to support the idea that FX market agents now pay more attention to macroeconomic data announcements. This evidence appears to suggest that the underlying economic data have become more important in these markets relative to the key monetary policy announcement.

A different picture emerges when we consider the impact on the LIFFE contracts of the same set of non monetary policy related announcements. For the short sterling and long gilt contracts these reactions are lower in the post Bank independence period. Since the total impact of interest rate announcements is also lower following May 1997, it is difficult to make any clear statements about the relative importance of monetary policy for LIFFE fixed-income market participants. We can say that all announcements now appear to have a lower impact upon the two interest rate contracts that we consider. Finally, there is a significant decline in FTSE 100 volatility around the set of macroeconomic announcements.

The empirical analysis presented is based on a relatively short sample, including a period when the markets will have been learning about the new monetary policy framework. The results can only be suggestive rather than the basis for firm conclusions. Nevertheless, there is some evidence that interest rate announcements have become less important for some financial markets, and no more important for others, since May 1997.

Common message standards for electronic commerce in wholesale financial markets
(59k)
(by Bob Hills of the Bank's Market Infrastructure Division).
An important aspect of electronic commerce is the potential for market participants to automate transaction processing fully, from the point of trade to final settlement. Such ‘straight-through processing’ could make wholesale financial markets more efficient, and lower the costs and risks that participants face. But it requires participants to use common message standards to exchange transaction data electronically. This article describes some of the initiatives to establish common message standards.

The development of common standards is central to the move towards automated processing of trade data and the wider adoption of electronic commerce in wholesale financial markets. This automation is expected to bring significant efficiency gains, as well as a reduction in costs and risk. Initiatives led by market participants to establish common standards have made considerable progress.

But it remains the case that too many trades in today's financial markets are still processed using fax or incompatible electronic networks. Standard-setting bodies continue to face difficulties in their efforts to gain widespread adoption of common and compatible message standards over the life of a trade.

Competitive pressures may force common standards to be adopted more widely if they are associated with new technologies that give market participants new ways to reduce costs or improve services.

The impact of XML, in particular, could be considerable. It has the potential to address some of the traditional failings of standards-that they are either too rigid, and do not reflect the needs of a particular market, or else that they are so flexible that they barely constitute a standard. It may also facilitate technological progress, by reducing firms' switching costs and so lowering barriers to entry and barriers to change. But this is likely to happen only if market participants work together to ensure that the XML-based standards that they create are inter-operable.

The precise ways in which electronic commerce and the development of common message standards will affect market structure in the medium term are difficult to predict. But it is clear that changing technology has the potential to bring about significant changes: to the ways in which markets operate and to the roles of market participants.