Lies, damn lies, and statistics

Rarely have statistics been misused so much for political purposes as when recently the ECB published the results of a survey of household wealth in the Eurozone countries.

Thus begins a new column by Paul De Grauwe and Yuemei Ji, which points out, inter alia, that the median household wealth statistics currently being used by some German economists and commentators to justify future wealth grabs in the Eurozone periphery are in fact telling us something important about German inequality.

But it gets worse. Tim Worstall (H/T Eurointelligence) quotes the ECB report as follows:

This section shows how households save for retirement using voluntary private pension
plans and/or whole life insurance contracts. Public pensions and occupational pension plans
are not considered in this report, as the value of some public pensions and occupational pension
plans can be difficult for households to evaluate. Cross-country comparisons are challenging in the sense that institutional arrangements across countries with respect to the different modes of retirement savings, such as voluntary private versus public or occupational, can be quite substantial. A deeper analysis of these differences falls outside the scope of this report.

As Worstall says, this means that many households’ major asset is being excluded, essentially on the grounds that including them would be really rather difficult.

Time for the report to be consigned to the dustbin, surely, and for those people currently abusing it to spend even five minutes or so reflecting on what the likely political impact would be if their proposals were implemented.

43 replies on “Lies, damn lies, and statistics”

Dont blame the ECB (well, dont blame it all on them anyway). Blame the idiotic media which decided to blast the report out to anyone and everyone without seeking to explain or interpret it.

à la recherche du temps perdu

In particular

“If the private sector runs a financial surplus (an excess of income over spending), there has to be either a fiscal deficit or a current account surplus (or both). The bigger the private surplus, the bigger the fiscal deficit or current account surplus must be. If, in reverse, fiscal deficits are to fall, the private sector must spend more relative to income or the current account must improve. Evidently, this needs to happen with higher spending, not lower incomes, particularly after a deep recession.”

I would actually go one further than Kevin. It’s not just that this throws into question the conclusion reached in the paper (however innocently arrived at), it may actually by design give the *wrong* conclusion.

Economies where a greater proportion of households don’t own their own home are, almost axiomatically in the OECD, going to have more household wealth in assets such as pension funds. I will leave it to fellow readers to speculate on the correlation between household financial maturity and the narrowness of the typical portfolio.

Anyway, conspiracy theorists might be interested in the following graph:
Germany has the lowest proportion of household wealth in owner-occupied housing, while Spain has the highest of the EU-15, followed by Greece.

If I were a German economist, I might certainly be making a point about the importance of diversifying household wealth portfolios away from one massive, illiquid and bubble-prone asset. It is risible, however, that anyone with basic training in economics would jump to the conclusions that some seem to have.

Wolfgang Munchau had an interesting angle, noting the study’s implication about wealth based PPPs.

It should be said that the Cologne Institute for Economic Research criticised the methods of collecting data used in the Bundesbank study when it was published last month.

Sven Giegold MEP of the Green Party  pointed to the German wealth distribution: “The study really highlights the very unequal distribution of assets in Germany as well as in Europe,” Giegold said. “In Germany, the top 10% of households by net worth own 60% of the country’s wealth.”

Deutsche Welle, the German broadcaster, was one media outlet at least that gave prominence to the caveats including dubious comparisons with Spanish data that had been collected in 2008 before its household wealth dipped as well as difference in typical household sizes

Also last month when the Irish national accounts data for Q4 and 2012 was published showing a 15% jump in computer services exports that resulted in the reporting of a rise in GDP in 2012 rather than a shrinkage, it was all reported as fact. Ministers bragged about the rise in services exports and the fact the headline data showed that goods exports had been overtaken in 2012 for the first time.

A week later, the Central Bank claimed the jump in computer services exports resulted from competitiveness when the fact behind the fantasy was that the rise mainly related to tax dodging, unconnected with activity in Ireland.

Unless people are fools and cannot see what’s going on, the issue is the truth rather than exaggeration or official spin.

A far as I know, any economist who commented on the data went with the official view.

It reminds be of the bubble period when it was hard for a dissenter to break through the wall of self interest and consensus.

Most people here avoid the issue but as with much else, when there is a foreign angle, it’s another story.

Absolutely right to point out the value of pension “assets” / “legitimate expectations”, but it does seem to have taken German ones needing to be illuminated.

What is the value of a typical Irish occupational pension pot?

How does the value of a public sector one typically compare with a private sector one?

Has anyone done a comparison between those in Ireland or the periphery and Germany?

That would be the thing to do rather than telling everyone to just stop considering the question, surely.

Any report from the ECB on the mean household wealth of the unsecured bondholders they were so keen to protect?

So, who owns half the houses in Germany that the householders do not own. Perhaps it is the Greeks or the Cypriots or, oh schade!, those pesky Russians with the hot money.

And how does the ECB household wealth survey reconcile with GDP per capita in the EZ. Just wondering!

Whenever I read a statement these days my first point of call is who wrote it and what’s their agenda.

I believe this is the only way to determine what to read and what to discard. Any Government Spokesperson (or politician) will make the “facts” suit their cause.

I regularily end up wondering if there are any real journalists left in the country and then we see the BAI Proposals designed (probably by Rabbitte) to remove the last chance of a decent question.

I’d love to see Jeremy Paxman interview each member of the Dail. They wouldn’t last pissing time before they’d be running to Rabbitte asking him to shut down the press because they were asking questions…..

Looking at the bright side, this study is (i) obviously flawed and (ii) serves to underline the very large, and growing, disparity in wealth in Germany. To the extent that it was intended to display Germany as too poverty-stricken to bail out other countries, if this was the intention, it has clearly failed.

What the debate does illustrate, however, is that the rich in the creditor countries think that the rich in the debtor ones should be stuck with more of the bill. It is hard to find fault with this argument as many of the latter were major beneficiaries of various asset bubbles.

There may be a middle ground. Merkel has floated the idea of using the €6 billion made available to fight youth unemployment to fund early retiremenst in the countries most affected (on the model of an earlier such scheme in Germany). How serious the intent is to actuallly put up any cash remains to be seen. The current budget is short over €11 billion and the usual culprits (Germany and the UK) are blocking moves by the Commission to get countries to cough up.

@ Grumpy

“What is the value of a typical Irish occupational pension pot?”

Do you mean price?

@Joseph Ryan

The same house costs twice as much in France than in Germany .Between Berlin and Paris ,the ratio is much worse .
The owners of the rented houses are in the first 10% of the distribution of wealth .This affects the average wealth ,a lot less the median wealth.The ECB study refers to the median not the mean( see Paul De Grauwe and Yuemei Ji).

The US consumer has a very high income and a very low net wealth due to a minuscule savings rate .The relation between GDP and wealth is not as simple as you imply.

There are always caveats in international comparisons and usually authors and users of data miss some of them.

Yesterday in Amsterdam, Mario Draghi said: “…we have seen reductions in unit labour costs. Ireland has seen an 18 percentage point improvement relative to the euro area average.”

That claim was based on data but is misleading.

Paul De Grauwe and Yuemei Ji criticise the data used in the ECB survey while they go ahead and use OECD capital stock data. It’s a reasonable assumption that they do not themselves have an indepth knowledge of the 22 economies covered.

The Irish data they use is not reliable and that is likely the case with some other countries also.

In an economy like Ireland, so dependent on inward investment, the paper value of outward investment is greater! Accenture, the American management consultancy, is an ‘Irish’ company.

I certainly wouldn’t take the Netherlands data used by De Grauwe/ Ji at face value either.

So as the Queen says in Shakespeare’s Hamlet: “The lady doth protest too much, methinks.”

The Bible: “Let him who is without sin among you be the first to throw a stone at her.”

1. The ECB study data are reliable (at least for Germany) and well corroborated by other sources

2. The data are understandable, if you know German economic history and the local situation

3. The de Grauwe data are nonsense

And I will show all 3 points here.

3. de Grauwe

He takes very wild estimate data from 2001, based on lots of general extrapolations, pre Euro times, and then certainly does NOT “update this to 2012”, but makes wild extrapolations over more than 10 years, based on wild assumptions, how this develops, without any tax levels, unemployment, house price development, population, especially immigration, etc. Completely wild.

The underlying theory, Solow growth 1957 is well known to NOT work at all at this relative precision level. People have tried so many extensions to that, to make it somehow fit on a much rougher scale.

But as Goldman Sachs, you can use it to sell BRIC stocks (the famous O’Neill paper Nr. 99), to gullible customers, with much, much higher growth rates in play,

making a ton of additional assumptions, e.g. how exchange rates and productivity develop, for which the comparison with the real data later on just make you laugh.

The assumptions de Grauwe takes are also strange (3% depreciation).
I am not aware that anybody else does this kind of nonsense anymore.

But maybe Kevin O’Rourke can show me somebody, and how this person validated this wild thing with any kind of data?

2. german wealth development and distribution

Germany went 2 times in recent history through very different situations, after WWII and reunification, and the corresponding economic policies and situations are therefore different.

I do this not in the historic timeline, in order to get the easier to understand parts to you quicker:

a) the Capital stock in East Germany in 1990 was practically wiped out

Factories, Houses run down, accounts receivable to eastern countries not retrievable. To the west significant debt, which was of course also wiped out

b) a lot of the money flowing into the East went to consumption, and no invest, because, you know, if people are entitled to get a western social minimum, they will not stay for substantial less in the East

c) The approximately 100% of GDP had nevertheless to be squeezed out of the West, with high taxes on corporations and individuals (marginal 60% tax rate)

d) With in many eastern location real estate prices going up pretty rapidly with generous tax subsidies, and then crashing in 1998, buying the old crap simply did not make sense. Especially when people had to suffer one bankrupt employer after the other. They spent their money on travel, into the free west, and consumer goods, to the point that we had some kind of consumer credit problem in 1999.

And you know what, I would have probably made the same in their position.

@ john

I tried to motivate myself to read more than 3 pages of this.

But it is kind of “yawn”. I didnt take RR as 90 % a kind of magic threshold, and actually also didnt read it that way.

And of course the consequences depend heavily on how did you get there, what is the perspective, what is your net international position, and what are the alternatives for what volume of investment.

And now some folks act like they have found some big smoking gun.

@francis i provided a cheat sheet or cliff notes…..the spread sheet error ties in with the heading on the post above but agreed off topic,kinda fun though..will stop gloating soon… really i will.

Macroeconomics really is the servant of political dogma. Alesina was transparently flawed if you bothered to think as you read it – but it became a “classic”. If RR really have cocked up like that to reach the ‘right’ conclusion I think you can argue the whole subject, so far as it it used as the intellectual enabler for politicians, has little credibility left.

Meanwhile George Osborne is squeezing down transfer payments to £53 per week, or about €60. Wonder what the multiplier on that is?

@ grumpy

I could also contribute Cobb-Douglas and Solow to have it screwed up, even significantly more, anybody interested in discussion ? not necessarily today, but kind of until next weekend ?

We laughed here about the IMF multiplier discussion as well

Maybe I start this with one really extreme case, Einsteins General Relativity theory.

Special relativity was obviously true, the simplicity had considerable charm, something kind of special for physicists, but is it true?

It took basically just ONE data point , Solar_eclipse_of_May_29,_1919, that convinced most of the physicists: YES

It was a : if we have a chance to see it, than on that day in this place, and everybody understood this PREDICTION quantitatively

if the outcome of some economic analysis depends on just one data point in the past, like this hilarious IMF multiplier, which depended on GReece with the notorious data problems, its just political garbage,

one of the reasons people here do not believe the IMF anymore in general.

One of the reasons I came to macroeconomic was , that I found out , pretty quickly , that all these short term fits, efficient market crap, etc ist just that crap.

And I did for some short time also things similar to this Grauwe above : – )

And then I did go back to Cobb-Douglas, what did they do exactly,

crap, ok, but where could I find suitable datasets, aaaah, US Great Depression, and there is even this Solow thing, and then you look in the details of the analysis of data , which were on just one page !.

And it is crap too. And the followup papers did not fix this …., a story developing over years , and pretty systematic search and analysis on my side.

I can do this.

Therefore I knew very fast what is fundamentally wrong with de Grauwe.


Good link from Walter Russell Mead:
It is well written and very unbiased.

@ John

soo, coming back to you, and the RR thing, its kind of expected, if choosing a few data or not (and I would have actually excluded the post WWII for myself internally also) determines the outcome, yeah, as usual in economics: not hard


the topic of this thread was the ECB wealth survey.

KOR was arguing that Grauwe and the Baron Münchau(sen) gang eurointelligence would have shown anything, that the ECB report should “to be consigned to the dustbin”.

The opposite is the case.

The notorious dishonesty of Grauwe and Eurointelligence was again exposed.

And we had this already, all nations will finance their resuce packages by taxing their own rich. This summer.

@Francis indeed,and sorry have not had a chance review/read in detail the above.
Hence my reluctance to hijack a very interesting tread,but the RR controversy one could argue ties into above.
My proverbial bar stool awaits,will read the materials and reply appropriatly,regards as always.

@ grumpy

“Macroeconomics really is the servant of political dogma. Alesina was transparently flawed if you bothered to think as you read it – but it became a “classic”. If RR really have cocked up like that to reach the ‘right’ conclusion I think you can argue the whole subject, so far as it it used as the intellectual enabler for politicians, has little credibility left.”

Thinking aloud.

Alesina may have been ‘transparently flawed’ but not so much that our own IFAC has taken his work (decreasingly) seriously over the last few years. Because of the way the academic system works, he reminds me of the Greek story of Tithonus, who was granted the wish of living forever, but not eternal youth. So he aged and aged until he became a tiny cricket.

But there’s flawed and there’s flawed.

On R&R my scorecard reads so far:

* Well done Herndon, Ash, Pollin for taking the time to work through the paper. If Macro is to retain credibility it is through developing stronger peer review systems.

* Poor stuff on not releasing underlying data. Stephen Kinsella shows how it should be done via twitter link.

* More questions to come on the choice of examples and the weighting.

* Yes, I think their work had an impact and was taken seriously, see briefing of Republicans and letter from Rehn, etc. It had the quality of a meme.

* Rogoff in particular looking disingenuous about ‘influence’: see his briefing of Senators and further examples, eg: at DAVOS. He is the former head of the IMF.

* Poor initial response from R&R, see Krugman. They need a decent PR guy.

@ Grumpy

“If RR (Rogoff and Reinhart) really have cocked up like that to reach the ‘right’ conclusion I think you can argue the whole subject, so far as it it used as the intellectual enabler for politicians, has little credibility left.””

A lot of flannel around the death of Thatcher mentioned the late 70s and the UK and an intellectually exhausted Labour party that had run out of ideas. Thatch came along with the “Labour isn’t working ” slogan. Gotcha.
I think 35 years on we are in the same place with neoliberal orthodoxy. The memes are in large part banjaxed.

Take the Japan money supply story. Punters are supposed to invest in risk assets. Why? Earnings growth in Europe and the US is all sloping downwards. Why bother buying overvalued crap? A nice exit strategy for the well informed at the expense of Mrs. Watanabe. How many models are still based on efficient markets ?

So I wouldn’t blame the situation on 2 people even if they are Reinhart and Rogoff. System failure tends to have far more parents. As Michael Hennigan notes we don’t do accountability here in the West.

Private Eye on Terry Leahy, formerly of Tesco

“Like most great leaders he is modest but relentless in taking credit for success and modest to the point of reticence in not taking the blame for anything at all.”

A fascinating report with all sorts of insights into economic/cultural diversity across the EZ. What a great pity Ireland and even the UK weren’t included.

There are some very difficult to comprehend stats, an example of which are:

Average public pension wealth in Lux is $1.5M. We are talking social security here, albeit contributory. The equivalent in Ireland is about $300K.

Average participation in credit card debt 4.3%!! Really. I pay my CC off each month but during the month I enjoy the credit. We’re talking 4.3% of households not persons, I would have guessed 95%.

Median deposit €6K. Even for households with the main person over 55 this only rises to €8K. This makes the sacrosanct €100K look like the preserve of a very tiny minority. Something doesn’t figure here.

@ seafoid,

“Germany is working”, and we say, the target in Europe should be 60%, and it is in the treaties : – )

I dont know, whether you count Germany as part of “the West”, we certainly do.

We “do accountability here”.

the detail part of point 3.

You can be a Dr. Freiherr von und zu Guttenberg, economics and defence minister and all that, apparent heir to the chancellorship, with 75% approval ratings.

If you get found cheating on your PhD thesis, and dont fess up quick and complete enough, you are getting laughed out of the country.

How long would you like a list of similar cases of fatal missteps, people in many other countries would just laugh about?

Germany is no less corrupt than the UK. For the really big disasters nobody pays. The bundesbank wallahs who designed the euro- what are they doing now? Even post ww2 most Nazis were spared. I think it’s called Realpolitik.

Comments are closed.