Summary of Quarterly Bulletin
August 1999

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

What makes prices sticky? Some survey evidence (66k)
(by Ian Small and Tony Yates of the Bank's Monetary Assessment and Strategy Division).
It is now widely accepted that price stickiness-the tendency for prices not to adjust immediately to changes in market conditions-is an important feature of the transmission mechanism of monetary policy. In September 1995, the Bank of England conducted a survey of price-setting behaviour by UK firms to find out just how sticky prices were. This article-based on the data that were collected in that survey-tries to shed light on what makes it more or less likely that prices will be sticky, ie that they will not respond immediately to changes in market conditions. This article uses the Bank's price-setting survey to investigate what might make prices more or less sticky. It discusses the impact of competition; whether price changes are prompted by cost or demand shocks; if price stickiness is related to the characteristics of firm's customers; whether price changes vary if goods are sold abroad or into the domestic market; and, finally, whether prices are more sticky downwards than upwards.

There are several findings from the survey. The more competitive are firms' product markets, the greater is the propensity to change prices in response to a demand shock; but market structure does not affect the responsiveness of prices to changes in costs. Second, there is some evidence that real rigidity (measured by the flatness of the supply curve) reduces the responsiveness of nominal prices to demand shocks. Third, there is evidence that the lower are search costs in product markets (at least measured by our proxy), the greater is the responsiveness of prices to changes in costs; although search costs seem to have no effect on the responsiveness of prices to changes in demand. Fourth, the export intensity of firms appears to reduce the responsiveness of prices to changes in costs; this supports pricing-to-market models based on rigidities in demand, as opposed to those based on the sunk costs of supply. Fifth, there are significant asymmetries in the responsiveness of prices to the direction of cost and demand shocks. Demand increases appear less likely to prompt price changes than demand decreases; but cost increases are more likely to prompt price changes than cost decreases.

The results in the article confirm some theories of price stickiness, but reject others. What weight should be placed on these results? The answer depends on how valid the research strategy presented is as a device for testing economic theories. One view that has much currency in modern economics is that economic theories should not be constructed from ad hoc relationships, but be judged by how well founded they are in microeconomics; so analogously it might be argued that, where possible, theories should be well founded in microeconometrics. So the evidence in the article is a useful complement to macroeconometric studies. Moreover, much of the empirical evidence on price stickiness, based on aggregate or individual firm data, faces the difficulty of identifying natural experiments that correspond to the theories being tested. The questionnaire, which asks firms what they would do in particular situations, sets up those natural experiments explicitly. Clearly, to place any weight on these findings assumes that firms would actually do what they said they would do, if confronted with the situations hypothesised in the questionnaire.

The use of explicit targets for monetary policy: practical experiences of 91 economies in the 1990s (50k)
(by Gabriel Sterne of the Bank of England's Centre for Central Banking Studies).
In June 1999 the Bank of England hosted its sixth Central Bank Governors' Symposium. This year the subject was 'Monetary policy frameworks in a global context', based on a report prepared by DeAnne Julius of the Bank's Monetary Policy Committee and Maxwell Fry, Lavan Mahadeva, Sandra Roger and Gabriel Sterne of the Bank's Centre for Central Banking Studies (CCBS). In this article Gabriel Sterne draws on one of the chapters of the report. The report uses a survey of 91 central banks to assess developments in monetary frameworks across a wide cross-section of economies. The final report, along with a selection of papers originally presented at a CCBS Academic Workshop in November 1998, will be published by Routledge in mid 2000.

A monetary policy framework comprises 'the institutional arrangements under which monetary policy decisions are made and executed' (McNees (1987), page 3). Following the breakdown of the Bretton Woods system of exchange rates, policy-makers have employed a variety of monetary frameworks in order to increase the credibility of monetary policy. Since the key characteristic of the framework is often an explicit target for monetary policy, the aim of the article is to assess the use of such targets in a range of economies in the 1990s. The analysis is based on data provided by 91 central banks that responded to a questionnaire on monetary policy frameworks circulated by the Bank of England in late 1998.

Explicit monetary policy targets have become more widely used in the 1990s than at any time since the Bretton Woods era. In the survey of 91 central banks, 96% (all but four countries) were using some form of explicit target or monitoring range in 1998. This contrasts sharply with 1990, when only 55% had an explicit target or monitoring range.

The article assesses in detail the use of explicit targets. The first section of the article argues that the choice of policy target rests not just on the likelihood and utility of hitting a single number. Other important roles for explicit targets may include defining informal or formal contractual relationships between institutions, and focusing analysis on particular economic indicators.

The second section goes on to examine which targets have been adopted in the 1990s by the 91 countries sampled, and the degree of flexibility with which they have been implemented. The announcement of an explicit target can represent full commitment to a particular outcome, or it may be no more than a benchmark used to explain deviations from the target. The sample provides extremes of experience that include rigidly fixed exchange rates on the one hand, and loose monitoring ranges for one or all of the exchange rate, money and inflation on the other. In the case of domestic monetary targets, the data used in the article relating to the deviations of outcomes from targets indicate that, in many cases, targets have been implemented quite flexibly.