GNP, M&A, EBITDA, P/E, NASDAQ, Econo-thread Part 10

[Roland]

"Does anybody have a basic explanation of what productivity is, from an economics standpoint?"

Well what it is is simple: output/input. How to measure it and why it changes are the interesting problems. When we talk about hourly labour productivity we need the aggregate of hours worked. And that based on stats that often err up to 5 % on the nr of people EMPLOYED.

"Almost like some place for economists to put things that they can't explain fully."

No, that would be total or multi factor productivity.

"I feel at a disadvantage of understanding when discussing the numbers immediately following WW2."

Well maybe an artificial post WW2 boom ended ? There is no cast-in-stone trend productivity growth.

"So these number seem to break the 20-year mold in their duration and strength."

You also have unprecedented imbalances in the economy. Inter alia an investment boom overhang. And investment drives productivity. Malinvestment drives productivity temporarily.

"Do you have a feel for what pet projects the Brits as a whole are trying to get through?"

Brits are on board with this, it's some capital and labour market reforms and some tech projects (internet in schools, the university stuff etc). Lissabon agenda, essentially.

"That one was scary."

Which one ?

"Probably not for the US. We have a pretty clean growth trend that is supported by known variables."

No. US fertility rate is now 2, it may stay there or drop to 1.5. Immigration is maybe 0.4 % per year - could increase to 1 % if you open the floodgates, could drop to zero if you get into some serious econ trouble. Out to 2050 it is just a guess.

[Colon™]

Dan, productivity is a measure of how much input is required to produce a given amount of output. The input could be labour, capital or land or multifactor productivity, which is supposed to measure the technological factor and is obtained by statistical black magic.
Labour productivity is preferable because labour-incomes constitute the lion-share of incomes, because it has a direct impact on wages and because it’s measured most easily by far.

Productivity is determined by numerous factors, of which the significance of many or most is hard to measure, not to mention the quality of a factor itself. (eg: Is the US education system good or bad? How much does the quality of the US education-system matter vs labour migration or quality of corporate training-programs? Etc)

You cannot assume that the rate of a given period (like 48-73) is more normal than another, and that there will be a return to when a certain factor changes (unknown in this case), because the entire environment is different now from then. There isn't a "normal" time-period, when the environment was normal, so it’s hard to me to believe in a normal trend-rate or a return towards it.

Another reason for me to stand sceptical towards trend-rates are the time-frames we use. For instance, you gave the average growth rate through 2001 starting from ’48, but why not include the war and pre-war periods (that would give you an average of about 2% IIRC), as there’s little reason to assume WW2 was an economic tabula rasa.

[pchang]

Question

Is productivity adjusted for inflation? All things being equal, output is measured in $$ only, but input is measured in $$ and labor hours. Labor hours don't increase due to inflation but $$ for input and output do. Is a writer in modern times really that much more productive than a writer in Roman times?

[Roland]

You take real output growth for productivity growth, yes.

Deviding nominal labour cost growth through productivity growth gives unit labour cost growth.

[faded glory]

Guys, what are the chances now that japan is getting worse and worse. They might call in there 1.1 trillion dollar debt?

[el freako]

Some interesting new data out from the OECD.

About this time last year the US had seemed to vastly outperform the EU over the period 1995-2000 (GDP per head in the US rose from 142% of the EU level in 1995 to 152% in 2000).

However, the release of newer data has changed this significantly.

First there was the adjustement to US GDP which lopped over 1% off the 2000 figure.

Then there was the census data for 2000 when an extra 7 million people were found in the US.

Finally has come the 1999 PPP figures from the OECD - PPP's are measured every 3 years and are adjusted for the intervening years by price deflators - these again showed that EU inflation would be much lower if measured using the hedonic deflator system the US uses.

This new data shows that GDP per head in the US was 141% of the EU level in both 1995 and 2000 (there was a small rise but it was less than 1%).


If you add in the expected growth rates for 2000-2002 (2.3% overall for the US and 2.9% for the EU) and use the average gap caused by US hedonic deflator methods for 1995-2000 then in 2002 the US has a GDP per head of 137% of the EU level.

The previous low was 139% in 1991.

It also alters the picture for productivity.

Using the new data the EU's GDP per hour worked went from 93.0% of the US's level in 1995 to 93.3% in 2000 - so even during the US's 'productivity miracle' it still perfromed (slightly) worse than the EU.



How many people can claim that the US still outperforms the EU?

[DanS]

I might note that the revised population figures throw things off substantially for the last several decades. It's not like an extra 10 million people arrived in the US just in the last 5 years. Rather, they were never counted previously, so who knows when they arrived, or even how many times they arrived. This is further complicated by the possibility that they might leave in bad economic times.

Secondly, these "found" people are mostly in the low productivity growth informal economy. This can't help but moderate the productivity growth number, whenever they arrived.

fg: Earlier in the thread I said what I thought about Japan. It doesn't look good at all. I'm sad for them.

[el freako]

DanS,

You may be correct that those 'classes' of people were not counted in 1990 or 1980 etc.

But the extra population only accounts for around a third of the revision in relative incomes - half comes from the 1999 PPP rates.

I have several points to make concering your statement "Secondly, these "found" people are mostly in the low productivity growth informal economy."

1. If by 'informal' you mean area's like the black or grey economies then this would have the opposite effect of actually raising productivity - as these people's hours of work would not be recorded but their spending would be.

2. In the 1990's the US's activity rate actuall fell compared to strong rises in the 1970's and 1980's - If the 'found' people were entering the economy then you would not expect this to happen.

3. If these 'found' people are only recently being included in the mid-year population estimate then you would expect either there to be a big jump in population one year in the 1990s (which doesn't happen) or, if the extra people are included over the decade you would expect population growth for 2001 to be more like the 0.9%-1.0% expected before the 2000 census compared to the 1.2% for 1990-2000, population in 2001 rose by 3.5m or 1.2%.

I therefore think that the 'jump' in population was due to real factors not the counting of people who were not counted before.

[DanS]

I don't know how it would affect the numbers, but when I say "found" informal workers, I mean an immigrant from Mexico who works in Washington, DC as a housekeeper, for instance. She probably gets paid in cash some of the time and sends a good portion of her income back home. She may slip in and out of the country many times in her life. She may be waiting for an amnesty so that she can go legit. The job that she fills may or may not be filled if she weren't there (low value work). She will be counted differently for different things or not counted at all and for various levels of government. Lots of different scenarios.

With the last census, the gov't said that they wanted an accurate population count above all, and that niceties such as immigration law would be forgotten more than is usual. I expect that the 2000 Census was the most accurate enumeration in quite a while.

edit: looks like we did a cross-post/edit. Quickly, in 2000/2001 there were a couple of green card amnesties that went on. For instance, there was an El Salvador (?) amnesty that gave out 100,000 or so green cards. Not 3.5 million, but...

[Roland]

El F: interesting numbers. Was about time to get back to realism over the US cooked up numbers...

Dan: What I don't understand is - how do you count illegal immigrants in the census ? We estimate about 100-300.000 here on 8 million people, but in the census, it is virtually perfectly in line with what official migration stats would suggest....

[DanS]

We count illegal immigrants in the census just like everybody else. That is, if they reply to the census, as they were encouraged to do in 2000. This was due to the Republicans having the power to quash statistical sampling in the census and the Democrats having control over the actual enumeration. This was a good political circumstance.

Re how we could be so wrong on the numbers, there are tons of factors. Much of it has to do with the fact that counting and then tracking illegals hasn't been a high priority item. When it gets to be a big issue and when there is political will, we do amnesties to reconcile the books and make everybody honest residents and then citizens.

[Roland]

"We count illegal immigrants in the census just like everybody else. That is, if they reply to the census, as they were encouraged to do in 2000."

But why would they reply ? Is there a guarantee that this data won't be used against them ? Or are they simply aware of the efficiency of US bureaucracy ?

[DanS]

The word was sent out through the immigrant community that the info would not be used against them. Then there was an active outreach (ads in Spanish and the like).

[DanS]

In a companion piece to Wolf's column referenced earlier, he goes through the per head figures much like ef did a couple of days ago but comes to different conclusions (used the 149% number). Interestingly, he delineates the per head figures by country.

http://news.ft.com/ft/gx.cgi/ftc?pagename=View&c=Article&cid=FT3CNFF4WXC&live=true&useoverridetemplate=FTD1OUN2DNC&tagid=FTDNE3BOBNC&SectionTag=na/column&PageTag=2comawo&imgID=FTD8OHRBONC

I'm really happy to see Ireland successful. Underdog makes good. Compelling economic story.

Here's a funny quote...

"Like many Europeans, I find the US addiction to ceaseless work appalling."

If he only knew.

[Roland]

Here's an even funnier quote: "...citing Business Week..."

LMAO.

Anyhoo, those comparisons that obsess about 0.5 %points of productivity growth are pretty lame. Most of GDP is in the service sector, most of that is not tradeable, so PPPs come with a huge uncertainty factor. And GDP captures only a part of the economy. Then we have labour productivity... no one knows the nr of hours worked. Even the nr of people employed is a wild guesstimate. When Germany changed social security laws, 2 million part time jobs "showed up", and the nr of people employed "rose" from 36.5 to 38.5 million. The BLS uses some rather daring fudge factors, too.

Is a US waiter more productive than an Austrian one ? A high school teacher ? That would be interesting micro things to investigate...

Btw, what does he mean by conference board ? The US conference board ?