Smaghi, member of the board of the European Central Bank, gave a bad speech....

from Merijn Knibbe

On 11 July Lorenzo Bini Smaghi, member of the executive board of the European Central Bank, gave a speech about the present state of the economy for ‘The Ruling companies association’ (I don’t make that up, M.K.) in Milan. He’s proud of it, as he put a link on the homepage of the ECB. http://www.ecb.int/press/key/date/2011/html/sp110711.en.pdf?061f56720e9a2d1fd1286e24b3b0b5cc

Should he be proud? No. Among other things, the ‘profit margins’ metric he uses implicates that profit margins increase when oil prices go up…. Generally, the mistakes he makes have three sources: 

1. Arithmetical impurities
2. Bad, bad graph reading
3. Conceptual flaws

Below, I’ll point this out in more detail.

Ad A. Twenty percent down is more than twenty percent up.

Suppose that I earn 2.000,– a month.  As I’m such a loyal employee, my boss promises to give me a one year pay rise of 20%, which will be followed by a decrease in pay of 20%. The increase will be € 200,– but the decrease will be larger: € 240,– and I will end with a wage of €1960,–. Down is more than up, when one uses percentages of economic variables. The economists who construct the detrended cyclical indices which Smaghi uses extensively (slide 11 and18) adapt their formulas to account for this unwanted effect.  Economists who use well defined metrics like these indices should have basic knowledge about the construction of these metrics. Smaghi does however not seem to have this, as he makes the mistake of confusing increases on a one to one basis with decreases and presenting this information in a table in a way that suggests that production increase in for instance Germany is larger than the post 2008 decrease (slide 21), even though the very graph he uses shows that it isn’t. In Germany, industrial production is still below pre-crisis level (only some transition countries have surpassed the pre-crisis level, by the way). There is no way that Germany has recovered from ‘2008’ – an important fact. Smaghi does not seem to understand this – but probably would have done so if it had been about his salary and not about the state of the economy.

Ad B. A weak and noisy relation is a weak and noisy relation and can’t be used to prove a point

Slide 4 states ”The recipe for recovery is competitiveness”. But the only thing that the graphs on slide 4 and 5 indicate is that if Unit Labor Costs (ULC) of a country increased 10% more than the EU average between 1999 and 2010 that might have been bad for short term recovery (compared with Germany) – smaller changes clearly do not show any influence. This is of course also not to be expected when one only looks at 2008 – 2011 growth, I mean, there is something like a business cycle and in the short run cyclical developments tend to trump structural developments. Smaghi should have taken a longe rperiod. But taking a longer period will show that Germany, Smaghi’s shining example, did not do too well up to 2005…. And the present (far from spectacular) upturn in Germany might as well be caused by rather low indebtedness of households and companies (the German government does have quite some deb, in factt). ULC may be important – but the graphs used by Smaghi do not show this (up to the -10% point). Note: as ULC are defined as the labor share in GDP you can’t decrease them forever, a point missing from all discussions about this metric.

More bad graph reading: slide 7, wrongly titled “The international environment remains favourable”. Considering  unavoidable ‘noise’ in the data which are used to construct the PMI-index (Purchasing Managers Index) of this slide the title should be: ‘The international environment might still be expanding, but even if it is, this expansion is much less than six months ago’. But I do admit – that’s a bit long. Also, according to a site on the PMI-index, “As with many other indicators, the rate of change from month to month is vital. A reading of 51 (expanding manufacturing industry) coming after a month with a reading of 56 would not be seen favorably by the markets, especially if the economy had been showing solid growth previously : http://www.investopedia.com/university/releases/napm.asp#ixzz1SAgnJThW. And lo and behold – the index recently did decline from 56 to 51 (well, 52), according to slide 7…

And even more: higher inflation rates in emerging countries are, according to Smaghi, also due to negative real interest rates in some of these countries (slide 17). Well, that’s theory. In the real world, Brazil, with a +5% real interest rate, has the same inflation rate as Turkey, which knows a -5% real interest rate according to Smaghi’s own graphs. What Smaghi tells is us is just not in the graphs.

Ad C. Did the ECB never hear about oil price increases, rentier incomes and implicit and explicit subsidies?

Smaghi shows calculations of the profit margin. This margin is defined as: “profit margins is defined as the growth rate of the GDP deflator at basic prices minus that of unit labour cost”. Now, that’s quite ‘bonkerish’, in my opinion. As, for some very bad reasons, Eurostat defines ULC as the labor share in GDP while Smaghi uses (mainly) Eurostat data, this means that:

1. When oil prices rise, leading to an increase in the GDP deflator, this profit margins metric rises
2. When land rents increase, for instance because of subsidies on so called ‘Green energy’ (the present German situation), this profit margins metric rises. The same for other rentier incomes.
3. When the number of self-employed increases because of a shift from wage labor to self-employed activities, this profit margins metric rises
4. When the government decreases wages, this profit margins metric rises

The very least Mr. Smaghi should have done is to account for changes in the terms of trade (point 1), to make a distinction between profits and rents, including high prices for natural gas and coal (point 2), to make a distinction between ‘mixed income’, labor income and genuine profits (point 3) and to limit the concept of profit to the business sector of the economy (point 4).

Another point: he does look at loans to households and non-financial corporations (sheet 44) – but only looks at it on a cash basis and does not look at it from the balance sheet point of view or the profit and loos point of view, an amateurish mistake which at least my students of (business) economics are NOT allowed to make (first year students, and not just my first year students, that is).

There is some interesting information in the presentation (government debt in Ireland went up from 20% of GDP (and falling) in 2007 to 100% (and rising) in 2010, those kind of things). And it’s important that ECB officials discuss this at meetings. Smaghi, however, does not yet seem to be up to this task.

About these ads