Mhux disponibbli bil-Malti
Jan-Oliver Menz
- 22 April 2024
- WORKING PAPER SERIES - No. 2930Details
- Abstract
- We study how millions of granular and weekly household scanner data combined with machine learning can help to improve the real-time nowcast of German inflation. Our nowcasting exercise targets three hierarchy levels of inflation: individual products, product groups, and headline inflation. At the individual product level, we construct a large set of weekly scanner-based price indices that closely match their official counterparts, such as butter and coffee beans. Within a mixed-frequency setup, these indices significantly improve inflation nowcasts already after the first seven days of a month. For nowcasting product groups such as processed and unprocessed food, we apply shrinkage estimators to exploit the large set of scanner-based price indices, resulting in substantial predictive gains over autoregressive time series models. Finally, by adding high-frequency information on energy and travel services, we construct competitive nowcasting models for headline inflation that are on par with, or even outperform, survey-based inflation expectations.
- JEL Code
- E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation
C55 : Mathematical and Quantitative Methods→Econometric Modeling→Modeling with Large Data Sets?
E37 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Forecasting and Simulation: Models and Applications
C53 : Mathematical and Quantitative Methods→Econometric Modeling→Forecasting and Prediction Methods, Simulation Methods - Network
- Price-setting Microdata Analysis Network (PRISMA)
- 17 July 2023
- OCCASIONAL PAPER SERIES - No. 324Details
- Abstract
- The coronavirus (COVID-19) pandemic caused a deep recession globally, as well as in the euro area, accompanied by a steep decline in inflation rates in 2020. This paper reviews some of the main challenges created by the pandemic for inflation measurement and provides micro price data analysis of how price setting has reacted to the strong COVID-19 shock. For this purpose, we use three different, but complementary, microdata sources for specific countries and sectors: micro price data underlying the official consumer price indices in Germany, Italy, Latvia and Slovakia; (scanner) data from German and Italian supermarkets; and online (web-scraped) prices for Poland. A common finding of the micro price studies in this paper is that state dependence significantly contributed to the price-setting response to the COVID-19 shock. Nevertheless, the extent and degree of responses varies widely by sector and even country, also depending on the severity of the pandemic situation.
- JEL Code
- D4 : Microeconomics→Market Structure and Pricing
E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation
- 17 July 2023
- OCCASIONAL PAPER SERIES - No. 323Details
- Abstract
- This paper provides an extensive literature review and analyses some open issues in the measurement of inflation that can only be explored in depth using micro price data. It builds on the analysis done in the context of the ECB’s strategy review, which pointed at directions for improvement of the Harmonised Index of Consumer Prices (HICP), including better quantification of potential biases. Two such biases are the substitution bias and the quality adjustment bias. Most analyses of substitution bias rest on the concept of the cost of living, positing that preferences are stable, homogeneous and homothetic. Consumer behaviour is characterised by preference shifts and heterogeneity, which influence the measurement of the cost of living and substitution bias. Climate change may make the impact of preference shifts particularly relevant as it causes the introduction of new varieties of “green” goods and services (zero-kilometre food, sustainable tourism) and a shift from “brown” to “green” products. Furthermore, PRISMA data show that consumption baskets and thus inflation vary across income classes (e.g. higher-income households tend to buy more expensive goods), pointing to non-homotheticity of preferences. When preferences are heterogeneous and/or non-homothetic, it is important to monitor different experiences of inflation across classes of consumers/citizens. This is particularly important when very large relative price changes affect items that enter the consumption baskets of the rich and the poor, the young and the old, in very different proportions. Another open area of analysis concerns the impact of quality adjustment on measured inflation. Evidence based on web-scraped prices shows that the various implicit quality adjustment methods can produce widely varying inflation trends when product churn is fast. In the euro area specifically, using different quality adjustment methods can be an overlooked source of divergent inflation trends in sub-categories, and, if pervasive, shows up in overall measured inflation divergence across countries.
- JEL Code
- E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation
- 17 July 2023
- OCCASIONAL PAPER SERIES - No. 319Details
- Abstract
- This paper documents five stylised facts relating to price adjustment in the euro area, using various micro price datasets collected in a period with relatively low and stable inflation. First, price changes are infrequent in the core sectors. On average, 12% of consumer prices change each month, falling to 8.5% when sales prices are excluded. The frequency of producer price adjustment is greater (25%), reflecting that the prices of intermediate goods and energy are more flexible. For both consumer and producer prices, cross-sectoral heterogeneity is more pronounced than cross-country heterogeneity. Second, price changes tend to be large and heterogeneous. For consumer prices, the typical absolute price change is about 10%, and the distribution of price changes shows a broad dispersion. For producer prices, the typical absolute price change is smaller, but nevertheless larger than inflation. Third, price setting is mildly state-dependent: the probability of price adjustment rises with the size of price misalignment, mainly reflecting idiosyncratic shocks, but it does not increase very sharply. Fourth, for both consumer and producer prices, the repricing rate showed no trend in the period 2005-19 but was more volatile in the short run. Fifth, small cyclical variations in frequency did not contribute much to fluctuations in aggregate inflation, which instead mainly reflected shifts in the average size of price changes. Consistent with idiosyncratic shocks as the main driver of price changes, aggregate disturbances affected inflation by shifting the relative number of firms increasing or decreasing their prices, rather than the size of price increases and decreases.
- JEL Code
- E3 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles
E5 : Macroeconomics and Monetary Economics→Monetary Policy, Central Banking, and the Supply of Money and Credit
- 6 February 2023
- WORKING PAPER SERIES - No. 2773Details
- Abstract
- How much does quality adjustment matter in measuring consumer price inflation? To address this question, we use different sources of micro and macro price data for Germany and the euro area. For Germany, we find that quality adjustment applies to a large range of goods and services but, on average, price adjustments due to quality changes reduce headline inflation only by 0.06 percentage points, which is balanced out by an increase due to quantity adjustment (e.g. a smaller package size) of the same amount. For the euro area, we assess the impact of heterogeneous quality adjustment methods by deriving the distribution of member states’ cumulative inflation rates for typical quality-adjusted products. Our macro-based estimate makes up to ± 0.2 percentage points for headline HICP inflation and ranges between± 0.1 and 0.3 percentage points for core inflation, when controlling for income differentials between member states. [...]
- JEL Code
- E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation
C43 : Mathematical and Quantitative Methods→Econometric and Statistical Methods: Special Topics→Index Numbers and Aggregation - Network
- Price-setting Microdata Analysis Network (PRISMA)
- 17 June 2022
- WORKING PAPER SERIES - No. 2669Details
- Abstract
- Using CPI micro data for 11 euro area countries covering about 60% of the euro area consumption basket over the period 2010-2019, we document new findings on consumer price rigidity in the euro area: (i) each month on average 12.3% of prices change, which compares with 19.3% in the United States; when we exclude price changes due to sales, however, the proportion of prices adjusted each month is 8.5% in the euro area versus 10% in the United States; (ii) differences in price rigidity are rather limited across euro area countries but much larger across sectors; (iii) the median price increase (resp. decrease) is 9.6% (13%) when including sales and 6.7% (8.7%) when excluding sales; cross-country heterogeneity is more pronounced for the size than for the frequency of price changes; (iv) the distribution of price changes is highly dispersed: 14% of price changes in absolute values are lower than 2% whereas 10% are above 20%; (v) the overall frequency of price changes does not change much with inflation and does not react much to aggregate shocks; (vi) changes in inflation are mostly driven by movements in the overall size; when decomposing the overall size, changes in the share of price increases among all changes matter more than movements in the size of price increases or the size of price decreases. These findings are consistent with the predictions of a menu cost model in a low inflation environment where idiosyncratic shocks are a more relevant driver of price adjustment than aggregate shocks.
- JEL Code
- D40 : Microeconomics→Market Structure and Pricing→General
E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation - Network
- Price-setting Microdata Analysis Network (PRISMA)
- 11 October 2021
- WORKING PAPER SERIES - No. 2604Details
- Abstract
- Those of professional forecasters do. For a wide range of time series models for the euro area and its member states we find a higher average forecast accuracy of models that incorporate information on inflation expectations from the ECB’s SPF and Consensus Economics compared to their counterparts that do not. The gains in forecast accuracy from incorporating inflation expectations are typically not large but statistically significant in some periods. Both short- and long-term expectations provide useful information. The professional forecasters expectations seem to help to correct the upward forecast bias in the low inflation period and to make the model forecasts more robust, in particular in the environment of high volatility. By contrast, incorporating expectations derived from financial market prices or those of firms and households does not lead to systematic improvements in forecast performance. The analysis is undertaken for headline inflation and inflation excluding energy and food and both point and density forecast are evaluated using real-time data vintages over 2001-2022.
- JEL Code
- C53 : Mathematical and Quantitative Methods→Econometric Modeling→Forecasting and Prediction Methods, Simulation Methods
E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation
E37 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Forecasting and Simulation: Models and Applications
- 21 September 2021
- OCCASIONAL PAPER SERIES - No. 264Details
- Abstract
- This paper summarises the findings of the Eurosystem’s Expert Group on Inflation Expectations (EGIE), which was one of the 13 work streams conducting analysis that fed into the ECB’s monetary policy strategy review. The EGIE was tasked with (i) reviewing the nature and behaviour of inflation expectations, with a focus on the degree of anchoring, and (ii) exploring the role that measures of expectations can play in forecasting inflation. While it is households’ and firms’ inflation expectations that ultimately matter in the expectations channel, data limitations have meant that in practice the focus of analysis has been on surveys of professional forecasters and on market-based indicators. Regarding the anchoring of inflation expectations, this paper considers a number of metrics: the level of inflation expectations, the responsiveness of longer-term inflation expectations to shorter-term developments, and the degree of uncertainty. Different metrics can provide conflicting signals about the scale and timing of potential unanchoring, which underscores the importance of considering all of them. Overall, however, these metrics suggest that in the period since the global financial and European debt crises, longer-term inflation expectations in the euro area have become less well anchored. Regarding the role measures of inflation expectations can play in forecasting inflation, this paper finds that they are indicative for future inflationary developments. When it comes to their predictive power, both market-based and survey-based measures are found to be more accurate than statistical benchmarks, but do not systematically outperform each other. Beyond their role as standalone forecasts, inflation expectations bring forecast gains when included in forecasting models and can also inform scenario and risk analysis in projection exercises performed using structural models. ...
- JEL Code
- D84 : Microeconomics→Information, Knowledge, and Uncertainty→Expectations, Speculations
E31 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Price Level, Inflation, Deflation
E37 : Macroeconomics and Monetary Economics→Prices, Business Fluctuations, and Cycles→Forecasting and Simulation: Models and Applications
E52 : Macroeconomics and Monetary Economics→Monetary Policy, Central Banking, and the Supply of Money and Credit→Monetary Policy