I am very glad to welcome this guest contribution by David Lizoain on how to measure inflation correctly in the Eurozone. It is difficult, and at times dry topic, but I encourage you all to read David’s excellent summary of the issues below.
Mario Draghi and the rest of the ECB Governing Council paid us a visit in Barcelona last week. After witnessing the overwhelming police presence and the temporary suspension of the Schengen treaty, I can attest that these guys are serious about maintaining their price stability.
The Governing Council defines price stability “as a year-on-year increase in the Harmonised Index of Consumer Prices (HICP) for the euro area of below 2%”. What I would like to discuss in this post is not the inflation target of 2% but rather the HICP. I’d like to argue that by targeting a flawed variable (the HICP), the ECB is generating flawed policy.
The HICP has one enormous deficiency: owner-occupied housing (OOH) is presently excluded from the index, primarily because of the methodological inconsistencies that existed between the statistical methods of the different countries.
Omitting OOH means that the HICP is not as complete of an indicator of the price level as it could be. Moreover, on account of differential rates of home ownership across Europe, the weight of the omission varies from country to country. This makes it difficult to compare HICPs across countries and calls into question the appropriateness of combining the different indicators to generate a single European index. (For a nice example of the non-comparability of the HICP across countries, take a look at page 67 of Eurostat’s technical manual on OOH).
In the key paper on this subject, “Housing Prices and Inflation in the Euro Area”, Boris Cournéde pointed out the following:
The recent experience of strong house price inflation in several euro area countries begs the question as to whether the exclusion of owner-occupied housing costs might have driven a wedge between the HICP and the cost of living. The presence of any such wedge clearly matters because many important economic decisions, such as wage settlements and consumption choices, are directly influenced by changes in living costs.
The use of an incomplete HICP makes it much harder to measure and therefore correct imbalances across Europe. This factor should be taken into account when the issue of distributing inflation across Europe is raised.
Very gently, Eurostat explains that the ECB might be getting everything wrong:
From the outset, it has been considered unsatisfactory to exclude OOH from the HICP since this may give a misleading picture of the inflationary pressures present in the economy. Hence, the exclusion of OOH may impinge on the ability of the HICP to meet its primary objectives, which are, on the one hand, price convergence assessment in the EU and, on the other, the monitoring of price stability in the euro area. For this reason, the treatment of OOH in the HICP has been given the highest priority at Eurostat.
The Eurostat draft elaborates how OOH might be incorporated into the HICP. There are various methodological options; Eurostat is opting to use a net acquisition approach.
This approach is expected to generate relevant differences in the measured rates of inflation. If a country has a housing bubble, incorporating OOH into its HICP would be expected to raise its measured rate of inflation; a housing crash would produce the opposite effect.
In the case of Spain, my assumption is that if OOH had been incorporated in the HICP during the bubble years, higher levels of inflation would have been registered – with all the consequences for monetary policy. By excluding OOH, the ECB was allowed to conduct monetary policy that was more appropriate for Germany than for Spain. Including housing would have shown with even greater clarity how competitive imbalances were building up across Europe.
A higher rate of inflation in Spain would also have implied a fall in Spanish real wages during the boom years. Coordinating wages across Europe is already difficult; if the HICP does not reflect the true evolution of the cost of living and if the HICP should not really be compared across countries, this task becomes even more complicated. (It’s worth taking a look at Andrew Watt of ETUC’s Golden Rule on this subject).
If we added OOH to the HIPC, we would expect Spain to be showing lower rates of inflation now that its housing market has crashed (which in turn would bring down the European rate). This deflationary tendency was also present when the ECB irresponsibly chose to raise interest rates in 2011. However, the HICP neither reflected the housing bubble nor the housing bust. Just as the country HICPs may not have been appropriately capturing the divergence in inflationary pressures, now they may not be capturing tendencies towards convergence.
I disagree with the ECB’s refusal to publish the minutes of its Governing Council. I disagree with the ECB’s choice to define a 2% target. But I am perplexed as to how the ECB could let such massive housing bubbles emerge and burst without incorporating OOH into their price index.