Brad DeLong's Semi-Daily Journal (2004)

The plotted value has a numerator and a denominator -- the compensation figure is probably based on some establishment survey (?) and can be considered reasonably accurate, but I'm more doubtful about GDP, which is subject to significant price indexing, so it is at best a "virtual", not a real entity.

At the same time, I have a gut feeling that the trend that the graph is suggesting is definitely there -- how can businesses revenues grow at a higher rate than GDP (my assumption?) if productivity increases are fully shared with workers?

But there is not any fundamental reason why the productivity increase, or in Marxist terms, inreased surplus-value, should go fully or proportionately to workers' wages and salaries. Part of it could be put in society's coffers to redistribute to worthy activities like education, research, healthcare, infrastructure enhancement, etc. But then I would suspect that is not happening to the extent most people would like, but instead it is just circulating in various investment devices. Many tech companies for example do regular stock buybacks to spruce up the stock price, or do mergers by stock swap, and some companies pay dividends. There are various ways of disbursing profits that don't show in employee compensation.

This relates to an earlier conjecture I made about declining real wages (in terms of "constant" production, whatever that means). In the end all of those things -- productivity, social security, living standards, are somehow connected.

That plunging red line? It's a graph of the systematic and intentional destruction of the middle class.

And this is definitely, absolutely taking place in the software development industry today. Developers are being paid less than they were during the bubble, certainly, but employers are also trying to force salaries down below their pre-bubble values.

Employers' principal argument for this is "Developers just graduated from third-rate universities in Elbonia with less than 1000 hours of real computer time under their belt will work for 10 cents an hour and are completely interchangeable with mid-career first-world developers from top-tier schools who have been programming since elementary school!"

This is spreading from sector to sector like wildfire too. It won't be long before American economists, analysts, writers, editors etc. are seen as "too expensive." I give it two years at the outside. The only jobs left will be the rare middle to upper management jobs (reserved for those born into privilege) and minimum wage service industry jobs.

I would like to see the same graph for compensation paid to workers in the nonfarm business sector, divided by total value added of the nonfarm business sector (instead of GDP).

Anyone still sanguine about the erosion of the manufacturing sector?

When we lose factories we also lose the associated infrastructure which is critical to manufacturing startups - not just the machine shops and toolmakers, but also the parts suppliers and the sales force, the culture and the know-how.

We won't be the ones to make the next generation of widgets, because we won't be able to build them.

Hedonic pricing indices and chained dollar calculation do effect the GDP numbers to the upside (how much? can someone point me to GDP numbers that exclude hedonic pricing and make them comparable with European GDP numbers?) and are doing this with exponential increases. That effect can probably explain a large part of the drop in that chart. It also would explain the "productivity miracle" (productivity = GDP / working hours and many other numbers based on GDP comparisons.

Bernhard

Re: net capital income...

But couldn't some of the difference be also due to increased government consumtion? (esp. in the military). Also, given the growing size of the current account deficit, isn't it possible that a relatively large portion is due to net capital income of _foreign_ firms?

It is a steep fall, beginning in 1992. I am curious about the haspe of things during the ten years prior to 1992.

I think the broad trend is correct, but Mathew's point is the key one - self-employment has been growing reasonably strongly and you'd have to take their output out of the denominator as well as the numerator (same for the government sector of course)

The chart implies that profit margins have probably risen. And they have, but not by the same amount as the decline in the labor share of value added might imply. The difference is that capital consumption has surged, in response to a deepening of the capital stock, particularly in high-tech goods that tend to depreciate rapidly.

To the extent that capital deepening implies an ever rising share of capital consumption in GDP, ALL sources of income should tend to decline as share of GDP. In the case of labor income, this is evident in the chart. In the case of profits, a cycical revival obscures it. Admittedly, the a portion of the rise in profits/wages may also be structural. But I don't think you can infer that from the chart.

I'm with Bernhard. There are a number of data mysteries which seem to have too-rapid GDP growth at their heart. The productivity/jobless recovery question becomes a good bit more manageable if we just give less credit to quality gain in goods output. The disparate performance of US and Canadian economies (which have more overlap than most) becomes less disparate if we give less credit to quality improvements.

I'm assuming that it is the quality issue, where attribution of new "units" seems rather arbitrary, that is the most likely new source of distortion in GDP measurement. Perhaps GDP is skewed upward by some other factor. Still, just whittle GDP down a bit and a number of other things begin to make more sense.

FYI:

Study: Jobs shifting to lower-paying industries

In 48 out of 50 states, jobs are shifting from higher-paying industries such as manufacturing and information to lower-paying industries such as retail and hospitality, according to a report released today by the Economic Policy Institute.

In California, for example, the average pay in industries that are growing is $34,742, 40 percent less than the average wage in industries that are shrinking, $57,800. The study analyzed data from November 2001 to November 2003.

Michael Ettlinger, one of the economists who did the study, said government policies, especially regarding international trade, could address this shift.

The Economic Policy Institute is a non-profit group that was created to make sure the concerns of low- and middle-income workers were included in economic policy discussions.

By Margaret Steen Mercury News http://www.mercurynews.com/mld/mercurynews/7763001.htm

There may be many factors involved, but capital income is certainly part of the explanation; the 1998-2001 increase in your compensation/gdp index also corresponds to the 1998-2001 earnings slump, for example, and we are currently seeing earnings recoveries in other economies, not just this one.

As for the middle class being destroyed, the sharp increase in my net worth over the last 15 months or so owes a lot to the earnings recovery.

All the data in this chart is nominal data, so
price indexing or inflation would not distort them.

MacDonald's comments seem to be on target more than anything else I can think of.

In a normal cycle there is about a one year window
when profits per unit explode because unit labor costs growth falls below price increases so
margins expand. that window occured in 2003
this cycle so in 2003 profits share of the pie rose sharply.

the other side of the story is productivity. Virtually all the gains in productivity this cycle have gone to capital as the drop in hours worked means that labor income has fallen.
this shows up as zero growth in real percapita income since Bush came into office.

Perhaps there's a new sector of farmers using Rolls-Royces as tractors and bathing their cattle in champagne?

Scams like the recent $400 billion expansion of "medicare," i.e. subsidies for the medication of the aged, are one of the things coming out of "non-farm business compensation's share of GDP.

Minimally this may represent an increase in the welfare of the aged -- though on the average this is a group who are better off than the rest of us, so it is not immediately clear why we should be adding to their transfer payment income. On the other hand "medicare" payments are very largely just increased price pressure, or sloppy money spraying, in the direction of higher prices and margins for doctors, pharmaceutical companies, HMOs, and the usual suspects.

The financial rip-off apart, I wonder how much metaphysical damage is done to American culture by the endemic Republican notion that everybody is sick, and the sick are entitled to a place at the front of the line for government money.

The short answer to Brad's question is yes: net capital income has grown by more than 25% over the last three years. In fact, from its trough in 2001:Q2, profits for non-financial corporates are up 77%, from $376 to $667 billion. Meanwhile, value added by non-financial corporates is up just 8%. Thus, over the period, the ratio of profits to valued added for the sector has jumped from 7.4% to 11.7%.

None of these calulations depend on hedonic pricing, as they are based on nominal values. (The is, though, a second-order effect, as the price indexes affect BEA's estimates for economic depreciation.) However, there is a common misconception that hedonic pricing is somehow suspect. In fact, well constructed conventional price indexes yield similar results.

Also, the idea that hedonic pricing yields overestimates of U.S. growth relative to European growth has no basis, for two reasons. First, most Euroepan national accounts are constructed using fixed weight indexes; the U.S. national accounts rely on chain weights. The effect of a fixed weight index is to magnify the reported growth contribution from sectors where prices are falling rapidly: computers and the like. Chain weighting avoids this magnification. Thus, for GDP growth, what hedonic pricing giveth, chain weighting taketh away. In fact, you'd get the highest growth estimates by coming hedonic pricing with a fixed weight index, which a few countries outside the U.S. in fact do.

Also, most countries in Europe import the bulk of their tech equipment. The main effect of moving to hedonic pricing would be to raise real investment growth, while also raising also real import growth. The effect on GDP growth would be close to wash.

I'm staring at that graph.
Here are factors I would want to correct for:

Does compensation include stock options, retirement plan matching, medical care, and other perks? How much did this mix change over the 90s?
How much of this can be traced to the housing market (people increasing consumption with equity from their house)?

Basically, I can believe the trend, but I think correcting for those factors would reduce the signifigance by several percentage points.

See? I told you so.

You should stick in a separate graph, small if you like with a scale starting at zero.

Matt,

Accepting that price indices are not the problem, it remains true that a lot mysteries can be made less mysterious by assuming a slower GDP growth pace than has been reported. I don't know of any other single change that makes sense of as many things. Are we so sure that GDP accounting is on target, and that all these other oddities have some other explanation?

K

Brad: I'm not sure that this graph is surprising. It seems perfectly consistent with a jobless recovery -- output is growing while employment is not. This may simply be another manifestation of that phenomenon.

Of course, that still begs the question of WHY output is growing without higher employment...

Brad

i've been tracking some labor share numbers myself, and, this *has* been a remarkably biased recovery in terms of wage/profit splits.

yet, i can't seem to replicate your figure - could you post the table number from the BEA that you're working with? i can find a 'corporate sector' sheet, but, am not positive how you're exactly defining 'nonfarm business compensation'. or, are these possibly per capita figures? some guidance would be hugely appreciated.

thanks

joshb

Chris Hanson: "Developers are being paid less than they were during the bubble, certainly, but employers are also trying to force salaries down below their pre-bubble values."

Don't forget that Y2K remediation efforts forced employers to vastly increase compensation to programmers. Post Y2K, there is no competition for IT people and employers in large companies are paying less for consultants across the board.

I'm perplexed by this 7% drop in real worker compensation. I'm looking at a historical Employee Cost Index study issued by the BLS on 11/26/03. It says that between September of 2000 and September of 2003 the index rose, in constant dollars, from 106.8 to 112.3 (with 100 being June 1989). That's an increase of 5%. Even if we're measuring worker compensation per capita (so that all the people who are now not working need to be included) the decline in employment hasn't been large enough to turn a 5% increase into a 7% decline. Or am I missing something obvious and important about these numbers?

"Take total compensation paid to workers in the nonfarm business sector, and divide it by GDP. Make the result into an index, with 1992 = 100. That is the red line in the figure above."

If Brad's plot is true - and the numbers strike me as just as odd as they do James - changes in income to investors can account for only a miniscule fraction of the change. The obvious explanation would be that more individuals are becoming firms -- and salaries are getting replaced by contractual income -- so that income shifts between accounts even though the persons involved are roughly equally well off. This would tie in with the employment statistics, esp. the discrepancy between firm payrolls and household employment. Still, that is only a discrepancy on the order of 1-2%.

Another point is that it would be nice to see a longer time series. Clearly, the equilibrium value of nonfarm business compensation to GDP should be less than 100, as nonfarm workers aren't the only people who share in the fruits of GDP. So another question is, why was the ratio so large 1992-2000?

It's hard to draw any conclusion from a single graph, but I'll give it a try. Hey in my Physics lab as an undergraduate eons ago, I actually plotted graphs from two data points, as the evenings grew darker and I got hungrier.

Since business revenue must go somewhere, doesn't this graph suggest an increasingly larger share of GDP of nonfarm businesses being classified as retained earnings? In particular, a larger share of wealth is accumulating under the control of top-level managers of capital. So even though manager's compensation is included in that sinking (stinking?) red curve, the assets that these managers ultimately control is not.

As Chris Hanson suggests

"American economists, analysts, writers, editors etc. are seen as "too expensive." I give it two years at the outside. The only jobs left will be the rare middle to upper management jobs (reserved for those born into privilege) and minimum wage service industry jobs."

Throw in University Professors, particularly in the California state university system...

But this only will be elucidated by a better understanding of the ownership and control structure of the US economy e.g. a large scale
graph of who owns what and who controls what.

Karl Marx said salaries would always stay low because capitalists would always shift jobs to the country that would pay the lowest wages. Lenin said The only country a capitalist is loyal to is the one his feet are in when he makes a deal. I'm not a communist, but it looks like they were right about some things.

Well, isn't this consistent with what opponents of free trade said would happen? The losers are labor. Some jobs go overseas, and labor are (informally / illegally) imported to do jobs which can't be sent away. Free trade.

The argument has been made (e.g. by Tom Friedman) that busting labor (he said unions, but millions of actual flesh and blood workers **were** affected) was a good thing, freeing up money for defense, tech innovation, and investment. Marxist economists whose names are forbidden (Sraffa, Joan Robinson) say that there's a strict reciprocity here, and that someone has to lose.

And then there's the "reserve army of the unemployed" idea which I've tried to float here and there. Until we have global full employment, seemingly wages will fall indefinitely, especially in unskilled labor. Wasn't that one of the goals? The US has no comparative advantage in unskilled labor, and increasingly less in semi-skilled and skilled labor. Just a location advantage for some workers.

Conservatives **do** argue that labor (and the unemployed) are far overcompensated in Europe, leading to stagnation of various sorts. Maybe they're right. Others argue that American labor is still overpaid while third world labor is starving. There's a point there too.

The liberal freetrader argument that everyone will be better off with free trade is true in the long run and on the whole, when we're all dead, but not for large groups of individuals during their actual lifetimes.

It's too soon to tell, but when the Democratic Party made free trade its flagship issue under Clinton (same as the Republican Party had done under Reagan and Bush), by alienating and weakening a big part of its base it may have damaged its long-term prospects as much as it did by promoting civil rights in the 60's. This doesn't prove that the Democrats was wrong in either case, but now that the Republicans have decided to go berserk they're facing a terribly weakened opposition party, and Rove is talking about reducing the Democrats to permanent irrelevance.

My vote:

#1 - Tax cuts, home refinancing has substituted for income allowing for increased consumption.
#2 - Soft labor market restraining/reducing domestic income.
#3 - Capital investment consuming a greater percentage of GDP as the economy shifts to rapidly depreciating capital assets. Farmers avoid this "problem" through the structural nature of their business (land doesn't tend to depreciate).

Greenspan wrote about long term assets last April.

"Before World War I, [...] Intellectual property--patents, copyrights, and trademarks--represented a far less important component of the economy, which was mainly agricultural"

"Over the past half century, the increase in the value of raw materials has accounted for only a fraction of the overall growth of U.S. gross domestic product. The rest of that growth reflects the embodiment of ideas in products and services that consumers value".

http://www.federalreserve.gov/BOARDDOCS/SPEECHES/2003/20030404/default.htm

http://www.epinet.org/content.cfm/webfeatures_snapshots

Jobs shift from higher-paying to lower-paying industries

In 48 of the 50 states, jobs in higher-paying industries have given way to jobs in lower-paying industries since the recession ended in November 2001 (see map). Nationwide, industries that are gaining jobs relative to industries that are losing jobs pay 21% less annually.1 For the 30 states that have lost jobs since the recession purportedly ended, this is the other shoe dropping—not only have jobs been lost, but in 29 of them the losses have been concentrated in higher paying sectors. And for 19 of the 20 states that have seen some small gain in jobs since the end of the recession, the jobs gained have been disproportionately in lower-paying sectors.

Just what modern trade theory predicts. What is so surprising?

http://www.epinet.org/content.cfm/webfeatures_snapshots

States with particularly worrisome job markets—where there have been substantial declines in the number of jobs and where the difference in wages between job-gaining and job-losing industries is particularly high—include:

Delaware which has lost jobs since the recession ended and where job- gaining industries have wages 43% below those in job-losing industries.

Colorado, which has lost almost 2% of its jobs since the end of the recession and where job-gaining industry wages are 35% below the wages in job-losing industries.

West Virginia, which has lost 1.7% of its jobs since the end of the recession and where wages in job-gaining industries are 33% below wages in job-losing industries.

The shift in jobs from higher-paying industries to lower-paying industries has affected nearly every state. This dynamic has the potential to significantly slow the growth of living standards for working families.

Table: Average wages in growing and contracting industries, end of recession to November 2003

Nationwide, industries that are gaining jobs relative to industries that are losing jobs pay 21% less annually.1

1. Note that this is a larger shift than the 13% change that has been reported for hourly wages, suggesting that jobs are both shifting from higher hourly wage to lower hourly wage industries and from industries that employ workers for greater numbers of hours to industries that employ workers for fewer hours.

K Harris -

"Accepting that price indices are not the problem, it remains true that a lot mysteries can be made less mysterious by assuming a slower GDP growth pace than has been reported."

If this supposition is correct, and I am leaning to it, then we can more properly understand how weak job creation and wage and benefit "gains" have been since 2001.

On the stock option as compensation aspect, the Dec. 2001 paper by Hamid Mehran & Joseph Tracy in the NY Federal Reserve's Economic Policy Review is an interesting read of how the data is compiled as well as their analysis of wage trends in the late 90's. Mattew's comment (the 3rd one I think) is a good one and I wish I knew how to implement. But I see the rightwing spin on this: they might claim the nonfarm business sector has fallen relative to the entire economy, which is why they can trust the household survey employment numbers more than the payroll employment numbers. Now, I have no idea if this spin would have any validity given I have not figured out how to read the data in the way mattew suggests. Any else?

The data seems to be conspiring to turn me into a marxist!! But seriously, what else can you expect when capital is mobile and labor is not. That is a power differential that can only be addressed by a strong democratic political system functioning to balance the interests of workers against owners through policy. Which describes the us of a today -- NOT. What we have today is a chief executive who is nothing but an representative for the energy industry.

"Not to mention what such a fall in worker compensation would imply for capital income. The money has to go somewhere, after all. Could net capital income possibly have grown by 25% over the past three years?"

Productivity is rising nicely, job creation is poor, wages and benefits are rising quite slowly and not close to keeping up with productivity. So, the return to capital is fine. Corporate earnings and compensation for upper management are rising rapidly. We surely could have experienced a significant shift in return to labor and capital since 2001.

Pardon me if this has already been answered, but:

What's the source for either the graph or the raw data in question?

(As I'm trying to kick my caffeine habit, I figure being scared witless about America's economic future might wake me up in the morning.)

Thanks.

Add Brad's chart to Angry Bear's data showing that the largest group exiting the work force is college educated males, and the EPI survey showing that lost jobs are replaced by ones paying much less, and it is all of a piece.

Sure makes hash of the idea that worker re-education is any kind of panacea....

Paul Krugman
12/30/03

"So if jobs are scarce and wages are flat, who's benefiting from the economy's expansion? The direct gains are going largely to corporate profits, which rose at an annual rate of more than 40 percent in the third quarter. Indirectly, that means that gains are going to stockholders, who are the ultimate owners of corporate profits. (That is, if the gains don't go to self-dealing executives, but let's save that topic for another day.)"

Fourth quarter profits are likely to rise at 26%. Profits are rising rapidly, wages and benefits are not.

Trickle down economics. Or is it deluge up?

This graph is the first step in the investigation of something. If your hypothesis is that the average compensation of professional workers has fallen by 7 percent since the year 2000 then that should probably be investigated directly in a separate graph. As things stand the results can be interpreted differently depending on ones view of what is happening in the economy.

Nah -- the difference has merely been added to executive salaries and political contributions, so that the amount is probably closer to 18% growth.

I confess to not really understanding this idea that the "problem" -- if one exists -- is that "capital is mobile but labor is not." In the first place, the reason most unions have traditionally been hostile to immigration -- i.e., labor mobility -- is that they believe it lowers wages. In the second place, what difference does it make in the grand scheme of things whether capital goes to where the workers are or the workers come to where capital is? Take the "offshoring" of software-programming jobs to India. Imagine a scenario in which, instead of offshoring the jobs, companies instead just brought Indian workers to America and paid them the same as they would in India. The labor would be mobile, but how would the outcome for American workers be any different?

Anne

You mentioned Krugman's NYTimes OpEd article about gains to stockholders. I mentioned retained earnings, but these gains to stockholders could also be in the form of dividends. Obviously sales revenue for any firm goes into either wages, dividends, retained earnings or other costs (possibly including suspicious payments, such as funny depreciation charges). The answer to the riddle of the sinking red graph lies in carefully disentangling these accounting items. But the presumption should be that we will discover some structural change in who wields economic power in the United States.

Remember, the capital gains and dividend tax rates are 15%. For shareholders who are wealthy enough to have a significant income from the combination or, even better, a significant income from dividends alone, this represents a terrific tax break.

D. Barnes and others:

Re: business investment

I have been wrestling with the issue of what is a "normal" level of investment as a % of GDP. Certainly, there are tax-induced incentives for businesses to invest (at least until they expire in Dec. 2004), and it seems that there was a rather large surge of investment in the late 90s related to IT equipment, optics, etc. The chart I produced shows real non-residential equipment & software as a %ge of real GDP. This ratio starts out at about 3% in 1947, grows to 4% by the late 70s, is stable at about 5% in the 80s and starting in 1992 rockets up at a 45 degree angle peaking out at 9.4% in 2000. The number currently is 8.7%. My question is what is a "normal" ratio? Certainly it may be that due to a higher proportion of business investment being allocated to IT, that those assets like computers have a 3-4 year replacement cycle and therefore quicker depreciation rates, and that has somehow distorted this ratio. However, if more dollars are being allocated to IT, then less dollars are being allocated to other things, like fax machines and steel plants, so shouldn't the "normal" ratio be closer to the pre-1992 level?

monte

http://www.morganstanley.com/GEFdata/digests/20040112-mon.html

Stephen Roach -

There can be no mistaking the important implications of this jobless recovery. Lacking in job creation as never before, it follows that there is equally profound shortfall of wage income generation. Normally, at this juncture in a US business cycle expansion, private wage and salary disbursements — fully 45% of total personal income and easily the largest component of household purchasing power — are up by 8% (in real terms). Yet 24 months into the current expansion, this key slice of income is actually down nearly 1% — the functional equivalent of about a $350 billion shortfall in real consumer purchasing power.

Hm. Did I say a bad word?

Monte-- on the ratios you are working with this is a time when using nominl or real data makes a big, big, big difference. If you look at nominal capital spending data in the 1990s it
does not show a capital spending boom.

I've cited this data before, but two years into the recovery real wage & slary income is still
below where it was at the economic bottom. The increase in average hourly earnings , wages, has
been offset by the decline in hour worked.

In cycles before the 1990 recession this series would almost increase 10% over the first two years of recovery.

In the December employment report real weekly earnings were unchanged from a year
ago.

I'm starting to think the bond market is telling us the economy is a lot weaker than we think.
On the capital spending side we had a large tax incentive for firms to shift capital spending back to 2003. Has that happened and maybe it is the reason the sharp drop in computer orders in December was not a fluke?

>I confess to not really understanding this idea that the "problem" -- if one exists -- is that "capital is mobile but labor is not."

I think Stephen Roach does a better job of data analysis and summarizing the issues than I can do. At http://www.morganstanley.com I suggest
taking a look at
/GEFdata/digests/20040112-mon.html
and
/GEFdata/digests/20031006-mon.html
and
/GEFdata/digests/20031212-fri.html

Recoveries lose traction when profits don't
generate income and job growth. Note that when we
say 'recovery' we are assuming a *national* economy which is recovering. So if American companies make record profits by off-shoring high wage jobs to India, but that profit doesn't improve living standards in the United States, the sustainablility of the American recovery is at risk.

>Take the "offshoring" of software-programming jobs to India. Imagine a scenario in which, instead of offshoring the jobs, companies instead just brought Indian workers to America and paid them the same as they would in India. The labor would be mobile, but how would the outcome for American workers be any different?

Is this a serious question? As American voters, we have a level of control (including how many Indian software engineers are allowed to come into the country) that is entirely missing in the case that the jobs go to India. (We also have the protection of American intellectual property laws, but that is another discussion.)

It's worth noting that while I.T. jobs in America are 1% below where they were at the *trough* of the recession, the software industry in India grew 26% last year.

"If Brad's plot is true - and the numbers strike me as just as odd as they do James..."

I just looked at the numbers for every year since 1980 (the 3rd-q figure for every year through '03).

The graph sets the all-time high, not the average, as its measure of "100" and "normal", and counts the fall from there.

According to bls.gov compensation hit 59% of GDP as a peak as shown (100 on the graph) and also in 1982. That's out of 24 years. Coincidentally there was a recession all three times. Or perhaps no so coincidentally -- compensation continued to grow for a while as recession hit boosting it to a "top" as pct of GDP. That's what it looks like, anyway.

Over the full 24 years the median pct for comp/gdp was 57% (with six years 57%)

The pct for 3rd Q 2003 was 56.0% (actually .5595) Setting the 24-year average of about 57.5% as "normal" or "100", I'd say we're down to about 97, not 90. At least according to the numbers I saw.

In an election year I guess we're going to see a lot of measuring losses from all-time highs.

Camille, I think you're missing the point of my response. You said in an earlier post that bad things for workers are what we can expect to happen "in a world where capital is mobile but labor is not." That implies that if labor were mobile, things would be better for workers. My question was: What difference does it make if capital goes to the workers or the workers go to capital? Statistics about the impact of capital mobility don't help answer this question. And when you say "we have a level of control" over the number of jobs that go to Indians in the U.S., what you mean is that we can restrict the labor mobility of Indians. You think this is a good thing, so it doesn't really make sense for you to say that things would be better if labor were mobile.

The point is that your problem is not with the combination of capital mobility and labor immobility. Your problem is with mobility, period. You don't want foreign workers emigrating to the U.S. in unrestricted numbers, and you don't want "American" capital leaving the U.S. in search of foreign workers.

Of course, the other part of your argument I don't understand is that the U.S. has been a massive importer of capital in the past decade, which is to say that far more foreign money has been invested in the U.S. than American money has been invested abroad. How can it be true that, on the whole, Americans have been hurt by the mobility of capital?

If anyone wants to try and understand "the large and variable statistical discrepancy between national income and national product" here is the source.

http://www.bea.doc.gov/bea/dn/nipaweb/TableView.asp?SelectedTable=43&FirstYear=2002&LastYear=2003&Freq=Qtr

James, maybe Americans are hurt by investment in the US because they either now have no jobs or not very good jons, and furthermore they don't own the company any more.

I think that economics would be a lot stronger if it were not possible to refute a statement by pointing out that it sounds Marxist.

I ran the numbers for 24 years from the BEA for compensation/GDP through Excel to check my eyeball, but then the Professor changed to using NI, so I had to run those *too* to see if what I saw before remained. ;-(

But it does -- and more so.

The first thing that leaps out is that the highest -- purportedly "best" -- numbers for compensation share come with recessions. The 24-year highest comp/NI ratios were in the 1980-2 recession years, measuring on the Professor's 100-pt scale 103.4, 101.7 and 102.7 compared to an average for all 24 years of 100. Remember the 10% unemployment of those days? And the two "next best" comp shares arrived with the other two recessions, 1991-2 and 2000-1.

So it hardly seems that the "best" numbers are really all that desirable! They sure look like they result from NI (and GDP in the prior graph) falling faster than compensation, rather than from rising compensation. Which is entirely logical, as wage setting lags profitability.

Secondly, we *have* seen drops like this before, and more. Since the recent 2001 peak comp/NI has fallen 3.96%. But coming out of the 80s recession it fell more: 4.34%. And that "low" was reached during 1984 when GDP grew at over 7%! A real whole lot like now.

Thirdly, since the peaks of 2001 and 1992 are above the average -- and caused by recessions, which are not generally recommended as policy -- the apparent fall from them is exaggerated. The fall is to 3% below the 24-year average, as opposed to an apparent 5% from the graph peak and 10% reported in the prior post.

So we really have seen this before if we look back far enough. Comp/NI surges to a "good" long-term high as the economy plunges! Then it collapses to a bad long-term low as the economy shoots up from its bottom. As in 1982-84, so in 2001-3. It's hard for me not to believe all this is at least in large part simply an artifact of the whipsaw effect of compensation trends lagging sharp turns in the economy.

Well, we can always be optimistic. Maybe the recovery will stall out, the economy will plunge backward, and comp/NI will happily shoot up again!

I'm not sure who has been refuting arguments by saying that they sound Marxist, but it isn't me.

In any case, Zizka, I don't follow the logic of your response. Much of the foreign investment in the U.S. has been of the direct nature, that is, it has built factories, etc. That investment clearly creates jobs in a direct sense. The foreign investment in American stocks or bonds helps the U.S. economy by keeping capital cheap and accessible. Now, we can argue over whether we'd be better off if we weren't reliant on foreign capital to keep us flush, but I don't think we can deny that we are reliant on it. So restricting capital mobility will hurt more Americans than it will help. You can't talk about IBM sending jobs to India without also talking about Siemens setting up operations here that employ 60,000 people or Mercedes building its SUVs in Alabama. If you want to use the government to stop the first, then you run the risk of losing the second. The U.S. has more jobs and better jobs because we live in a world in which capital is mobile than the U.S. otherwise would. That may change in the future, but it is true now and has been for a long time.

"Camille, I think you're missing the point of my response. You said in an earlier post that bad things for workers are what we can expect to happen "in a world where capital is mobile but labor is not." That implies that if labor were mobile, things would be better for workers."

Why would I be interested in a 'what-if' scenario, just because it's the flip-side of my assertion? No, my interest is towards what produces prosperity for the American wage-earner. Period. Arguing about the characteristics of hypothetical mobile labor is a distraction.

In our historical moment American capital has mobility and American workers do not. This means that the Great American Job Machine is alive and well and producing jobs by the hundreds of thousands -- in India and China, primarily. Jobs produced directly by American companies or by subcontractors. An unprecedented number of these are replacing high wage American jobs. I am concerned with the implications of this fact. If you had read the links I suggested (I won't participate in this exchange any further unless I can tell that you have done so) you might be concerned as well.
Here is a quote from one of my links:
There can be no mistaking the important implications of this jobless recovery. Lacking in job creation as never before, it follows that there is equally profound shortfall of wage income generation. Normally, at this juncture in a US business cycle expansion, private wage and salary disbursements — fully 45% of total personal income and easily the largest component of household purchasing power — are up by 8% (in real terms). Yet 24 months into the current expansion, this key slice of income is actually down nearly 1% — the functional equivalent of about a $350 billion shortfall in real consumer purchasing power."

And that is why this recovery lacks traction.

"The U.S. has more jobs and better jobs because we live in a world in which capital is mobile than the U.S. otherwise would. That may change in the future, but it is true now and has been for a long time".

What is "a long time"? Since before free trade (NAFTA, etc.) or just since free trade?

"The U.S. has more jobs and better jobs". Well, we're talking about the jobless recovery right now. As for the better jobs -- better than what? The perception is that labor hasn't been doing so well recently.

If you're saying that you know that there's a net gain for labor -- for people who depend on labor for their living -- because you've asked the question and gotten an answer, bring it forth. If you're just saying so because it's axiomatic and known to be true -- well, I don't accept that.

Note that in my original statement that I allowed for the possibility that American unskilled and semi-skilled (and recently skilled) labor used to be overpaid and was draining resources better put into investment and/or into relieving third-world poverty by providing jobs. That's basically what Friedman said. I realize that politically that would be a hard sell, but if that's what's happening, let's say so.

And even if we're profiting right now from foreign capital investment, what is the long term prospect?

At this point I don't have a specific policy axe to grind. I just think that free trade cheerleaders have been evasive to the point of dishonesty. Couldn't we just make a table of the winners and the losers based on what's happened so far and what can be expected to happen?

Camille, I've read Roach's pieces -- I've even commented on them before in different threads on this site -- and I don't agree with them. I think the evidence for his conclusions is flimsy and anecdotal, and does not fit with the hard numbers on either employment or American investment abroad. There just has not been anywhere near enough offshoring to account for the trends in U.S. wage and employment growth (or lack thereof). It's possible that ten years from now things will be different. But they are not now.

On two particular points, the comparison between the rate of growth in the IT industry in the wake of the 1990-1991 recession and in the wake of this recession is preposterous because the employment rate in the IT industry was clearly inflated in the late 1990s by the explosion in cheap capital. The $600 billion or so that went to fund the telecom buildout, for instance, certainly employed a lot of IT professionals. But it wasn't sustainable, since the return on investment on most of that $600 billion was negative. (We could create as many jobs as you'd like by borrowing money and using it to pay people to write useless code.) The fact that IT employment is below where it was in 2000 is therefore not surprising. What Roach doesn't mention is the absolute number of American IT workers today compared to 1991 (though he's careful to mention the absolute number of Indian IT workers). I don't have the exact numbers to hand, but there are a lot more Americans working in the IT field today than were a decade ago.

The second point is that Roach has no explanation for what happened in 2001 to make this offshoring suddenly occur. American capital was no more mobile in 2001 than it was in 1999. Indian labor was not dramatically more productive in 2001 than it was in 1999. The pressures to be profitable were no greater in 2001. So what happened? If American companies could have been making bundles more in 1999 by shipping jobs abroad, why were they hiring so many workers that the U.S. unemployment rate was 4%?

Finally, the point about capital mobility is that it's not just American capital that's mobile. It's Chinese capital and German capital and Japanese capital. It's this capital mobility that's helped keep the U.S. economy going for more than a decade now. So if you want the U.S. to be a self-enclosed island, things are going to get a lot worse for American wage-earners before they get better (and actually I don't think they ever would get better).

Zizka, I am assuming that there is some connection between the level of investment in an economy and the number of jobs in that economy. That may be axiomatic, but it seems relatively uncontroversially axiomatic. Therefore, if a substantial percentage of the investment in an economy comes from foreign capital (as it does in the U.S.), it seems safe to assume that the benefits to that economy of capital mobility are widespread.

In any case, here's a recent interview with Peter Drucker http://www.fortune.com/fortune/subs/article/0,15114,565912,00.html in which he makes the same point more concretely, saying "We import twice or three times as many jobs as we export. I'm talking about the jobs created by foreign companies coming into the U.S. . . . We are exporting low-skill, low-paying jobs but are importing high-skill, high-paying jobs."

I don't really think the "free trade" period started in 1993. Outsourcing of manufacturing to East Asia occurred long before that, and Americans had been buying German cars and Japanese cars and TVs and electronic devices for two decades or more. I will say, though, that I always find the idea that NAFTA was somehow bad for America bizarre, since it was followed by the best sustained period of economic growth and employment and income growth since the 1960s. That wasn't the result of NAFTA, but I'm hard-pressed to see how we can say free trade led us astray in the 1990s.

Maybe free trade cheerleaders have been evasive to the point of dishonesty. I think the evidence is pretty straightforward: free trade hurts workers in certain industries, helps workers in others, and helps essentially all consumers by keeping prices down and fostering competition (as well as making the global economy more productive). Any rational tallying up of the costs and benefits to the winners and losers would show that the overall gains to Americans far outweigh the losses (although the gains don't necessarily come by significantly boosting GDP growth). That doesn't mean free trade is a panacea (for developing countries or the U.S.) or that the costs aren't real. But it does mean that we should not be setting up tariffs or restricting the ability of capital to enter and exit this country.

James, I think you are making good points and I kind of agree with you in some ways. But it's worth pointing out that you can't tell what advantage mobility gives in wage bargaining just by looking at flows. You need to know labor supply elasticity in a particular market vs. labor demand elasticity. If capital has more options than labor, then capital can hold down wage demands by threatening to leave. Flows of foreign capital matter if foreign capital is actually giving American workers options -- "if you don't raise my wage then those Chinese people will just build a factory here and employ me instead." If foreign capital flows into the financial markets here and ends up in the hands of American capital holders who can then invest it based on whereever they find low wages, but they happen to decide to invest it in the U.S. because the U.S. has low wages for an advanced country, then that has not necessarily increased worker bargaining power. The capital will just leave if workers try to use its presence to get higher wages. The direction and size of capital flows does not *necessarily* tell you whose bargaining power has increased. Although for recent levels of trade and investment it probably does -- this goes back to some of Krugman's and Freeman's points in their debate with Ed Leamer about trade back in the 90s. But the past may not be a good guide to the future when it comes to trade.

Also, I think it is legit to argue that certain firms, particularly in manufacturing, have made very impressive leaps recently in the capacity to use information technology to coordinate production across national boundaries. So the world is different than 1999 in some industries. The Wall Street Journal (news section, which is fantastic on econ, not editorial!) had some excellent pieces on this.

You are right that U.S. capital likes labor mobility from foreign countries. It increases capitals bargaining power by giving capital more options relative to labor. But the mobility of workers from some foreign countries is irrelevant to U.S. labor mobility. We find it difficult to go to another country if we get a bad wage offer here.

Does anybody know whether this time series includes health care costs as part of compensation? Crucial question right about now...

The ECI does include benefit costs.

As far as the capital question goes, you may be right about the fact that the impact of capital flows on wages is indeterminate before the fact. But Roach's focus has been on "the jobless recovery," and I just think it's implausible to argue that capital mobility has been responsible for destroying more American jobs than it's created. More substantively, I'm skeptical that more direct investment in a country is ever bad for its workers (well, with obvious exceptions like real-estate bubbles). If "American capital holders" decide to take the capital from foreign investors and invest it in the U.S. -- even if they're doing it because the U.S. has lower wages than other developed countries -- doesn't that have to be more in American interests than if they decide not to invest in the U.S.? (With appropriate modifications made for impact on consumers if capital is invested in low-wage countries instead.)

I agree that the process of figuring out how to use IT to coordinate production efficiently is an ongoing one, and that companies are making great strides every year, which is one reason why I believe the productivity numbers. But in terms of the outsourcing of IT jobs -- which doesn't involve complicated manufacturing schedules or especially elaborate organizational arrangements -- I just don't see any evidence that the ability of, say, software firms to use code written by Indian programmers was significantly greater in 2001 than it was in 1999.

Finally, my capital/labor mobility point was just a simple one: American workers want American capital and foreign labor to be immobile, and foreign capital to be mobile. So the "capital is mobile, but labor is not" diagnosis, which you hear surprisingly often, doesn't cut it.

James (and Camille): I don't know in which part of the country you guys are each living, but there are regional differences in employment patterns. If on a national level the outsourcing of R&D jobs (with the public's focus mostly on the computer industry) is small compared to other factors (which it is), in some high-tech regions like the SF bay area the impact is larger than the national average.

You wouldn't believe how many companies are creating or extending presences in India or China, and sending people with US experience there to lead ad build teams.

"But in terms of the outsourcing of IT jobs -- which doesn't involve complicated manufacturing schedules or especially elaborate organizational arrangements -- I just don't see any evidence that the ability of, say, software firms to use code written by Indian programmers was significantly greater in 2001 than it was in 1999."

James, Roach has clearly said that this outsourcing wave has been enabled by the new telecommunication technology (emblematically the Internet). This is one factor for why after dot.com the trend increased. Another factor that I heard anecdotally (but the number of anecdotes is ever increasing) is this:

(1) Increasing numbers of Indians who came here during dot.com want to go back and do go back, mostly for cultural and family reasons, but also due to shifting opportunities (perspective of career/business growth in India vs. grunt for life in the US, to state it bluntly).

(2) These people came here with technical expertise, and gained experience in how to do US-style business, which is often underappreciated. Not only is it important to be able to solve technical problems and develop products, but also to know what problems are important, what product features and solutions will appeal to potential customers, how to deal with business partners, business procedures, market development, how to sell effectively, what does the market look like into which you are selling, how to manage teams effectively, etc.

(3) This enables many of these people to become effective managers/business people, introduce US-style business methods to India, and the whole thing creates a positive feedback loop in which the return of those people creates an environment where more want to return, and more outsourcing opportunities for US companies are enabled.

I'm sure the same thing holds for China, only I happen to know more Indians than Chinese.

Also in 1999 there was more money for the industry to play with (no cost squeeze of today's proportions), the remote management and logistics were not well figured out, and also people wanted to come to the US, the mothership of capitalism and democracy. Today they are fingerprinted, (more) despised, and perhaps the shine of the US has faded a bit.

If the high-tech business in Asia gets more traction, and the societies are following suit by improving infrastructure and lifestyle (at least regionally), we may well see a relative exodus of high-tech professionals. Where I work quite a number of Indians are mulling over thoughts of going to Bangalore. But don't hold too much hope of their jobs being replaced locally.

CM, I think the "people wanted to come to the US in 1999" idea is an interesting one that I haven't heard before. I'm not sure it can account for whatever's happening, but I'd like to see someone write more about it.

I don't think the "cost squeeze" explanation will fly. Companies -- especially the kind that are at the forefront of outsourcing today -- are always presumably trying to maximize profits, so the fact that they had more capital to play with in 1999 shouldn't have made them more willing to waste it. And in any case, there is no "cost squeeze" for companies looking for workers. Everything we've been told -- especially on this blog -- is that you can get American programmers and IT guys for significantly less these days. So I can't see that sending jobs abroad is more urgent than four years ago.

I still don't really get the idea that "remote management and logistics" were not well-worked out in the late 1990s. American companies have been running foreign operations in developing countries since the early decades of the 20th century. They've been using EDI communication networks for at least two decades. The Internet has certainly made things cheaper and more efficient, and will continue to do so as it becomes more integrated into company operations. But I don't see the organizational or management mystery that got unlocked in the past few years. The one concrete improvement I can see is that I assume the technology infrastructure in India and China has been improving steadily. But it's not like these places were technological backwaters until 2001.

In that sense, I think Roach's analysis of the impact of "global labor arbitrage" on the U.S. job market is like his analysis of the impact of underreported hours worked on U.S. productivity. He's recognizing an important long-term trend but using it to explain short-term changes in a way that the data just can't support.

I have no doubt that companies are investing abroad and that India will become a major center of software development in the future. I have enormous doubts that this explains anything but a minuscule part of the "jobless recovery."

I think I tentatively agree with this statement. Too often a perfectly valid analysis leads to embrace of an illogical or harmful course of action. Regarding the ‘losers’ in a transition (to a globalized labor pool) that is, overall, beneficial for both Americans and foreigners, why can’t we establish a domestic ‘revolving fund’ to provide financial assistance to Americans subject to the stress of an industry going south. This cannot be blamed on any defect of character and, in some cases (although I would wager very few!) can it even be blamed exclusively on corporate mismanagement or malfeasance.

(Although I still think the steel industry management is culpable. George Bush gave that industry a two-year reprieve to prepare for the open market. And what did they do with that time? Did they attempt to diversify? Provide a transition fund for soon-to-be-unemployed workers? No. What they did do is under fund the pension plan, which has since defaulted to the PBGC insurance group, which, in turn, is now experiencing such stress that it too is considering bankruptcy.)

It is one thing to think ‘in the aggregate’, but it is another thing to provide some form of relief, in the name of civilized equity, perhaps, for the individuals who are experiencing the financial burden of being in the minority.

Post Script: What is the effect of the 10 to 15 million unregistered immigrants on the chart?

CX: I omitted the statement with which I agreed:

That doesn't mean free trade is a panacea (for developing countries or the U.S.) or that the costs aren't real. But it does mean that we should not be setting up tariffs or restricting the ability of capital to enter and exit this country.

To add my two cents to this free trade debate:

"Zizka, I am assuming that there is some connection between the level of investment in an economy and the number of jobs in that economy. That may be axiomatic, but it seems relatively uncontroversially axiomatic."

This is hardly controversial. Foreign investment allows the U.S. to invest and spend more than it can otherwise. Take away foreign capital and the economy will contract.

Frankly, it seems clear that the bubble induced overinvestment in the tech sector. At the same time the sector was also experiencing a temporary employment boost in Y2K. When the bubble popped following Y2K, the employment market for the sector collapsed.

Employment demand for this sector is therefore unusually low and will take longer than usual to work out as the investment overhang is worked out. Since wages are sticky on the downside, most of the impact from the change in demand is being seen in unemployment. Likewise, many workers who were induced into the labor market have again exited. War related spending is helping defense related parts of the industry.

I>t's too soon to tell, but when the Democratic Party made free trade its flagship issue under Clinton (same as the Republican Party had done under Reagan and Bush), by alienating and weakening a big part of its base it may have damaged its long-term prospects as much as it did by promoting civil rights in the 60's. This doesn't prove that the Democrats was wrong in either case, but now that the Republicans have decided to go berserk they're facing a terribly weakened opposition party, and Rove is talking about reducing the Democrats to permanent irrelevance.

Speaking as an armchair economist, I couldn’t agree more. Your analysis is spot on. The Democratic Party has a death wish that is being played out in the public spotlight. The decade of the 1990’s exposed a ‘wealth’ of deficiencies in the capitalist market system that could have been used by the Democrats to redefine themselves, while the opposition was scurrying to high ground in red-faced embarrassment. Iwhat did the Part do instead? It remained joined at the hip with stale domestic fiscal policy, oblivious to the trends of capital relative to income (as discussed in this thread), and, in it’s commitment to free trade, demonstrated an astonishing insensitivity to the real problems of dislocated people who are, in fact, the labor pool, as James Emerson points out. (The 200,000 to 2,000,000 new and ‘better’ jobs that were going to materialize after NAFTA never did appear. Business yawned. They were simply delighted that academia found a clever way to sell the introduction of a cheap labor pool to the American public.)

I would not be at all surprised to discover that Karl Rove is breeding Democratic economists in a barn somewhere.

CX: “Iwhat did the Part do instead? “

What did the Party do instead?

Between a smarty pants word processor that keeps changing my words and my own mistakes, I am using up band width. I’m out of here, but good insight, James Emerson.

KLA, I think the idea of some kind of trade assistance is a reasonable one. (I thought there was something like this put in place for NAFTA, but I could be wrong.) But I do have a question regarding your last post.

Bill Clinton is the only Democrat to win the presidency in a straight which-candidate-do-you-like-best election in the past thirty-five years. (Carter's election is inseparable from the backlash against Watergate.) He won twice, both times going away. (Perot, it seems clear, drew more votes from Clinton than from Bush or Dole.) During his presidency, when as you put it the deficiencies in the capitalist system were being exposed, 20 million new jobs were created, the U.S. enjoyed the lowest unemployment rate in the developed world, and we saw the first sustained period of wage gains for average Americans in three decades. And Al Gore, one of the worst presidential candidates in history, nonetheless won the 2000 election. (And I have little doubt that had Gore run toward Clinton's domestic policies, instead of away from them, he would have won convincingly). So where is the evidence that a) Clinton's domestic policies were bad for the country and b) that they hurt the Democrats at the ballot box?

RE: So where is the evidence that a) Clinton's domestic policies were bad for the country and b) that they hurt the Democrats at the ballot box?

I am just out the door so I cannot fully engage, but I did not mean to suggest nor do I think I said that Clinton’s domestic economic policies were uniformly bad in themselves nor bad for the country. I had, at the time, and still have, objections to the way NAFTA was implemented (even Mickey Kantor is a little reticent about the alleged ‘environmental’ constraints that were watered down into a few clauses that were ‘risibly nugatory’ in clout, to borrow some vocabulary from the philosophers.)

Since I am pressed for time, this is my main objection. Democrats can move in the right direction on occasion but they have a tin ear and a blind eye for consequences and consequences matter. NAFTA should have been implemented incrementally instead of the shock therapy approach that worked so well in, oh, say, Russia? There is so much doggerel in the modern world about the moral turpitude of a society premised on immediate gratification. Is it not appropriate to suggest that global transitional changes, such as those stimulated by NAFTA, be implemented with constraints that reduce short-term harm for a few while preserving long-term benefit for the many?

Final Post Script RE: Democrats and Free Trade

I think I am starting to understand. An infantry battalion is actually a group of ‘our boys’ but a labor pool is an economic construct. Got it! ;)

Well, some think that losing Congress was bad for the Democrats. Open up your eyes and look around a little. The hard right controls all three branches of government. These are not my father's Republicans. Barry Goldwater would be a moderate today.

I totally agree that free trade has been good for some people and bad for others. What I disagree with is the lumping and averaging method, characteristic of utilitarians and economists, which simply wraps the whole story up by asserting, based on general principles and cherry-picked data, that "everyone is better off". I'd just like to see who is and who isn't.

If a job moves from Germany to Alabama at the same time a job moves from Detroit to Japan, from a non-nationalist point of view what's happening in both cases is that a high-paid job is replaced by a low-paid job. IE, on the net, labor generally is worse off. The people getting jobs are better off, and the people losing jobs are worse off. This is just the normal competition that freetraders want, bringing th cost of a "product" (labor) down.

The reason it looks bad to me is that the supply of docile unskilled labor (some of which is easily trainable upward) is essentially infinite. When wages started rising in Mexico, jobs started leaving Mexico too. So to me the net wage-lowering dynamic probably will be stronger than the wage-raising dynamic until, as I said, the whole world is employed.

My point about the Democratic Party is that a pure freemarket conservative (with Social Darwinist tendencies) finds nothing to object to in this picture. That's what he wants. But traditionally the Democratic party has been the political representative of the people most likely to be hurt (i.e., unskilled and semi-skilled labor in the older age brackets), but they have essentially ignored that part of their constituency. This cannot strengthen the party.

By and large, the various ways proposed to reduce the harm to such people were not put in effect, mostly because Republican votes were what rammed the program through.

I mostly am objecting to the skewed and self-serving descriptions of freetrade by its advocates, especially the liberals among them. I am not actually supporting Gephardt or proposing a specific counterproposal. In general, though, I would support somewhat less free trade than we've got, slowing down the pace of progress, interfering with the market to the benefit of low-paid workers, some compensation of the victims, and heavier investment in retraining. After all, if tariffs were at 15% (average) 20 years ago and we were actually doing tolerably well, and they're at 3% now, is it outrageous to suggest that perhaps we'd be better tweaking what we've got so that tariffs ended up averaging 5% or 8%? But anybody who says anything whatsoever to this effect is immediately portrayed as a Smoot-Hawley Luddite Kim Il Sung clone.

Everybody knows why Gore was defeated, but nobody agrees about it. Glad that you have little doubt. Doubt is bad for people.

A factor noted by Democrats is that voters seem to support rather liberal Democratic policy proposals, but vote for Republicans or centrist Democrats who oppose these proposals. The two explanations given are usually either coded issues involving race and sex, or media bias. Based on what I've seen, Clinton did as well as he could have given the playing field he was on, but the results are as I've said.

So why is it hard for liberal Democrats to win? The Voters vote the way because of what the nof right-center media tells them, and in order to function at all and make media buys Democrats become slavishly dependent on donors. So it's not just the voters who are pushing the rightward shift. I don't have a quick solution for this, but the outcomes have not been not good. For example, health care coverage and free trade were two of Clinton's goals. He started out by compromising his healthcare plan to please the insurance companies, and then he lost the compromise too. But he did get free trade. There's a big skew there.

James mentioned that American labor wants foreign capital to come in but not foreign labor. Well, every country with any strength is gaming free trade to its advantage. China, India, Taiwan, Europe, almost all of them. The only countries playing by all the rules are helpless third world countries who have no choice. To propose that the US might game the system a little to the benefit of American labor is not absurd or outrageous. I understand why the Republicans wouldn't try to do that, but why didn't the Democrats?

In a lot of ways, ordinary Americans are the mules of the New World Order. We pay high drug prices so others can pay low drug prices. We fight wars to do whatever it is we're doing in the Middle East. We buy, run deficits and go into personal debt to juice up the world economy. Short-term that part is fun, OK, but there's a long-term down side and the other players are playing a different game.

"Companies -- especially the kind that are at the forefront of outsourcing today -- are always presumably trying to maximize profits, so the fact that they had more capital to play with in 1999 shouldn't have made them more willing to waste it."

Excuse me? Of course they are not willfully "wasting" it, but wouldn't you think the willingness to fund experimental developments (in the sense of "let's see whether this works", not necessarily "let's see what comes out of this") depends on how easily the momey is coming in? R&D and marketing organizations are always presuuring the purse string holders for funds, and when more money is there, they get more. Also more money means more accomodation of demand for raises, bonuses etc. If your family/kids have been pestering you for this $500 whatnot item that always looked a tad too expensive for what it's worth, and you get a, say, $10,000 surprise bonus, the odds that you finally give in to buying it will not rise? Why do you think corporate finance works differently?

"Everything we've been told -- especially on this blog -- is that you can get American programmers and IT guys for significantly less these days. So I can't see that sending jobs abroad is more urgent than four years ago."

What is "significantly less"? There is a limit to the lowballing, and there is the notion of "overqualification". With companies expecting (or at least not ruling out) a business improvement right around the corner, they don't want to give people grossly unfair deals (in the aggregate that is, there are always aberrations). Believe me that. Most know that these guys will be the first out the door when things turn around, and even if not, you will to some extend get what you pay for.

Also if you have a choice of hiring here or in Asia (e.g. you already have an office there), even significantly reduced rates here (10%? 20%?) may not match the cost savings there; there are hidden costs like healthcare and other insurances and variable costs of business besides nominal salaries, and don't forget there is a strategic element as well.

"American companies have been running foreign operations in developing countries since the early decades of the 20th century."

Software engineering and other knowledge work are different from manufacturing. The delay loops are much shorter, smaller lead times, and you need more interactive communication and faster turnaround. With manufactured goods, it is well-accepted that it will take between weeks and months from order to shipping, but with software and knowledge services your customers will expect response in the range of a week, days, or even hours. Those time frames will also be dictated by your internal deadlines, like when splitting R&D efforts between continents. This is where cheap and efficient communication/computer net infrastructure makes the difference between feasible and infeasible.

"The Internet has certainly made things cheaper and more efficient, and will continue to do so as it becomes more integrated into company operations."

"The one concrete improvement I can see is that I assume the technology infrastructure in India and China has been improving steadily."

Yes, and the outsourcing is slowly and "steadily" increasing as well. It is not that things have startedd when the recent wave of attention appeared, and the pace at which this goes is often exaggerated. It is a matter of slow shifting and compounding, but the effect is already visible.

"I have enormous doubts that this explains anything but a minuscule part of the "jobless recovery.""

Well, it explains some of it, and the IT/knowledge based services are not the whole economy. Again, there are differences between sectors, and nationwide averages need not correctly reflect trends in specific sectors.

Regarding "minuscule": if this contributes let's say 5-10% to the IT labor market calamity and you think it's litte, how about a "small" 5-10% paycut for you? When I asked people this question about similar quantitative judgements, it usually worked.

CM, I don't mean to hold you responsible for what other people argue, but then I wasn't really arguing with the picture of a "slow shifting and compounding." What I'm arguing with is the assertion, by Stephen Roach among others, that "global labor arbitrage" should be at the "top of the list" of explanations for the jobless recovery and stagnant wage growth. Roach is talking about the impact of offshoring on employment in the economy as a whole.

As far as the calamity question goes, Roach says IT jobs are 1% below where they were two years ago. Given how much IT employment was inflated by the bubble, is this surprising? Is it really evidence that something fundamental has changed? Is the feeling in the Valley any different from what it was in the mid-1980s, after the PC bubble burst?

Zizka, gaming free trade to the benefit of American labor (that is, that small percentage of American labor that's exposed to direct foreign competition) means gaming free trade at the expense of American consumers. Tariffs represent a transfer of money from all American consumers to a small number of protected American workers. And tariffs on the kinds of products that are traditionally protected -- textiles, shoes, autos, steel -- are regressive taxes. I don't see how it's a liberal Democratic position to ask someone who works forty hours a week at Wal-Mart or tending bar to subsidize someone who works in a textile factory or a steel mill.

I need to clarify the position in my original post to separate the confusion from the honest disagreements of opinion and fact.

I mixed two types of argument - a general critique of Democratic economic policies and a specific critique. John Emerson then introduced the economic policies of a specific Democratic president, namely, Bill Clinton, so there is a three-way overlap. My general critique of ‘standard’ economic policy is the inappropriate emphasis on fiscal policy to the exclusion of some newer trends that are becoming increasingly important, namely the growth and interplay of capital and income, as discussed in this thread. The standard line of tax the rich and give the poor a break inevitably leads to accusations of class politics and populism, at which point the dialogue breaks down, so any modern fiscal position must be placed within a broader context that includes the relationships between capital and income wealth. And yet this dialogue is almost non-existent.

My specific critique, which is not so much partisan as it is economic, is the free trade issue. I do not oppose free trade. I believe the expansion was inevitable and, well, we all know some version of how to deal with the inevitable. My specific objection is two-fold. I hate the way it was sold to the American public. If people want to make an issue of presidential ‘lying’, consider the promise of some wildly indeterminate number of ‘new and better’ jobs that were promised in exchange for opening the labor market. (The new jobs claim was based on a single study that was not repeated and predicted a net growth of 170,000 new jobs stateside - a number that was rounded to 200,000 and later expanded to 2,000,000.) There was so much wink-winking, and elbow pushing going on that I’m surprised we didn’t have an increase of Elvis sightings. (Yes, jobs did increase under Clinton, but none of that increase can be attributed to NAFTA. The ink was no sooner dry on the agreement when the Mexican economy sank, and Salinas departed in a hurry.)

The second objection I have to the NAFTA-led introduction of free trade is the failure to acknowledge, let alone deal with, the jobs issue. The response? Arnold Kling is advising displaced Americans to stop whining and consider a second career in home health care. I can only see a very bleak future for a Party unable to provide a civilized and equitable compensation for workers displaced during economic transitions, especially one that was initiated by the Democratic Party.

So, getting back to Bill Clinton. He is definitely squarely in the middle. I do not object to his fiscal policies, and as, James Emerson points out, his numbers are pretty good. But Clinton sponsored the NAFTA effort - for the greater long term good. I personally do not believe it is productive or accurate to ’blame’ one person - usually it is a president in the cross-hairs - for patterns that have long-term implications. The process is very, very complicated. The criticism I have is directed more at the leadership within the Democratic Party that is responsible for strategizing and defining tactics. You people - if you pardon my expression - are on life support.

Just for the sake of disclosure. I am what used to be called a Scoop Jackson Democrat or what I guess is now called a liberal hawk. I am a strong supporter of the Bush effort in the Middle East, but I am very wary of an unregulated market and unrestrained growth. I think the Democrats introduced some market forces that they do not seem prepared to acknowledge, let alone understand and mitigate. That is why I am now a moderate Republican. You need to get your economic house in order because the Party has lost a huge chunk of people who vote and think exactly as I do.

James, why do you choose Walmart workers for your example. You don't give a shit about them. Why don't you use yourself as an example of the consumer? Or the people you work for?

There are other things I would do for Walmart workers having to do with labor law, minimum wage, etc., more significant than what you're doing with tariffs.

I don't accept your claim that the proportion of American workers hurt by globalization is small. (I file it with your assurance that you know why Gore lost). I don't know what that number is, but I wish I did. You don't either, and you don't want to. (If you did know it, you should have told me). In any case, globalization goes along with a large number of other areas in which the Democratic Party has failed to represent the sector of society I'm talking about.

Globalization isn't really my issue any more. I basically dropped it a few years back. I'm semi-pro-free trade. I just get sucked back into it because of the arrogant, cocksure, absolutist assurance of pro-free-traders who don't seem to recognize any nuance at all.

What are the reasons, incidentally, WHY we won't see wages for unskilled labor on a race to the bottom? One of my main points.

You respond very selectively according to your agendas and what you assume are mine -- not to what I've been saying.

RE: effect of IT on business and labor trends

The cyclical nature of IT investment is surprisingly subtle. The fact that IT investment is currently somewhat lower is undoubtedly due in part to a major ‘paradigm’ restructuring within the IT industry. The Industry began offering a highly centralized processing environment that did not ‘fit’ the needs of many businesses without extreme restructuring of the traditional business operations. Much of the early yelling and screaming came from trying to put an elephant into a tutu and teaching it ballet. In response, IT retooled the approach to provide a more distributed architecture. This was an improvement but left companies without the centralized access required for decision-making in exchange for the benefit of providing local solutions to local problems. The third and present stage is a compromise between centralized and distributed architectures - known as hybrids. It is not at all unexpected that the transitions - marked by internal course changes - will have a phenotypic expression of an investment lag. That ’lag’ is covering a lot of activity.

The following from Steve Roach’s paper “The Global Labor Arbitrage” which I took the liberty of posting because I think he is right on the money:

“The asymmetrical impacts of the global labor arbitrage lie at the heart of the great political debate now swirling in the United States and elsewhere in the developed world. The resulting tensions underscore the distinct possibility that jobless recoveries may remain the norm in high-cost developed economies for some time to come. That’s not something to take lightly. The threat to traditional sources of job creation strikes right at the heart of economic security. That has given rise to a political backlash in the US Congress that could prove quite destabilizing for the global economy.

In the end, the choices are stark -- to look inward and protect the “old way” or to look outward and encourage the “new way.” I would dare say this dilemma is quite comparable to what farmers faced in the late 1800s as the Industrial Revolution blossomed. The same would have probably been true of sweatshop workers when confronted with the assembly lines of mass production in the early 1900s. The “rusting” of Smokestack America in the early 1980s is a more recent example. At each of these earlier critical junctures, there were no clear answers at what would drive the next wave of job creation. America’s model -- one that fosters flexibility, entrepreneurialism, innovation, technological change, and investment in human capital -- always found that answer. The global labor arbitrage forces the US and the rest of the developed world to rise to the occasion once again.”

Actually, I wrote a whole response about the "race to the bottom" thesis and then cut it because I thought I was going on too long. Anyway, here it is:

Your picture of globalization as a "race to the bottom" isn't accurate. Look at the history of the Asian economies over the past four decades. In just about every one of those countries, wages have risen steadily over time. Old jobs have left (Korea started out making shoes, then moved up to semiconductors and now high-end electronics), but new ones have been created. The picture of wages being lowered globally as a result of free trade just doesn't fit with the facts. We can debate whether this means developing countries should open their borders to imports, but I don't think you can argue that the movement of capital into developing countries has led to net losses for labor. You add up what a Korean worker and an American worker made in 1962 and add up what they make today, and you'll see that there has been a cumulative gain.

Back to our regularly scheduled programming: Man, don't tell me who I do and don't care about. I didn't use myself as the consumer because an extra dollar for a pair of sneakers or a t-shirt doesn't make that much of a difference to me. Does it make a difference to the person at Wal-Mart? Yes. One of the dubious things about tariffs from a social-democratic perspective is that they hit me and the Wal-Mart worker the same. Please, let's raise the minimum wage and get some form of national health insurance. But do not prop up industries that can only survive by charging more for the very same products than their competitors do.

More to the point, if you want to redistribute income to industrial workers, then there are lots of more progressive ways to do it than tariffs. Just raise the top income-tax rate by two percent and dedicate part of it to subsidizing companies threatened by foreign competition. (Actually, now that I think about it, that's probably illegal under the WTO, but you could probably give workers in select industries tax credits.) Regardless, the right way to do it is to make sure that the redistribution takes place down the income ladder -- rather than up, or sideways -- and that it doesn't constitute industrial policy. Let competitive markets work, then redistribute if the outcomes seem unjust or unacceptable. Otherwise you're going to get less growth, less innovation, and less productivity.

As for how many workers are affected by globalization, well, we know that just 15% of the American population is in manufacturing, which is presumably where most of the impact from foreign imports would come (since our imports of services are small). Robert Scott, who is as ardent an opponent of NAFTA as you're likely to find, estimated that it cost Americans 900,000 jobs: http://www.epinet.org/content.cfm/briefingpapers_bp147. I think his economics are absurd, and his paper is extraordinary for not mentioning the word "consumer" once, but I'm willing to accept the estimate for the sake of argument. I don't see those numbers as demonstrating that trade is responsible for devastating labor. And they have to be balanced against both the benefits we derive from foreign capital -- which employs a significant percentage of that 15% who work in manufacturing -- and from the lower prices and higher quality created by foreign competition.

I'm not sure why you're saying I don't see nuance. I've just spent hours arguing about this and offering you evidence in support of the case, rather than simply saying "free trade is good." But ultimately you have to decide if you think tariffs are a good idea or not. And saying that we should have 7% tariffs is not more nuanced than saying we should have 0% tariffs. It's just a different decision.

What about Mexico and Brazil? Korea and Taiwan had extraordinary advantages, in part due to being key military allies and poster children for capitalist development. Taiwan, when I was there in 1983, was remarkably protectionist; I don't know about Korea. They got their niches before globalization.

Thank you for the hypothetical things that might have been done. I can think of other hypothetical things that might have been done. Free trade has coincided with the destruction of unions, a slipping minimum wage, non-enforcement of labor law, uncontrolled low-wage immigration, general weakening of public benefits, and weakening and increasingly costly education. Republicans think that all these things are good, whereas Democratic free traders only think that one of them is good. However, Democrats have not effectively fought the others. And we've been hearing for all this time that Dems HAVE to focus on the middle class, as if the rest were non-existent.

15% is not terribly "only" to me. Especially because it used to be more than 15%. And especially because workers in all exportable areas (IT, help-desks) are now all competing both with overseas labor and with legally and illegally imported labor.

Incidentally, my "race to the bottom" argument is in part empirical and in part the kind of formal argument that economists make all the time. I guess such arguments are illegitimate if the likes of me makes them.

So anyway, I am not and never have been a real anti-globalizer, but the self-congratulatory and one-sided tone of what I hear (plus accusations that I want the third world to starve) offends me. Plus the offering of hypothetical things that might have been done as if they refuted my point. And including the opportunistic choice of Walmart workers as favored examples when they're not who the argument is about.

James: re dot.com bust vs. offshoring

You are arguing that offshoring is not on the "top of the list", but that the IT job loss is caused by the unwinding of dot.com hiring excesses. (Are you not?)

But these two things are not contradictory. I think nobody is seriously claiming that the recent job loss has been mostly, or even to a large extent, due to offshoring. The concern about offshoring, in my reading of what people are saying, is forward-looking; in other words:

(1) jobs have been lost mostly because the dot.com bust unwound a lot of ventures thus reducing demand on the computer and telecommunications industries, and sending forth ripples of reduced demand throughout the global economy

(2) with the recovery, to whichever extent it has been real, new jobs in the high-tech and knowledge-based service sector are created domestically only at a slow pace, and this is due to (a) increased pressure on workers to deliver "more with less", (b) increased IT-enabled productivity, and (c) increased IT-enabled offshoring, not necessarily in this order

(3) there is a "job multiplier" effect: a delta in white-collar jobs leads to a delta in other jobs and vice versa -- office construction, building maintenance services, office supplies, catering, gastronomy (business lunches/dinners/parties), retail, city/county spending due to differences in local business taxes, etc.

It is not primarily the job loss of the past that was due to offshoring, but sub-par domestic hiring today that is impacted. In fact the white-collar offshoring happened already before and during dot.com, but at a slower pace as experience had to be gained and technology was not at today's level (and it was to a large extent put in place by dot.com activities), and it was partially masked by the dot.com hiring frenzy.

CM, I think when Stephen Roach says that offshoring is No. 1 on his list of explanations for the jobless recovery, then we have to assume that he's saying offshoring is the main reason for the lack of new jobs in the IT industry between 2002 (or whenever the recovery ended) and today. That is contradictory to the argument I'm making.

The point about the impact of the dot-com hiring boom is not simply that when the bubble burst, people lost their jobs. It's that there was far more hiring than there was work in the post-2000 era. So the main reason, at least in my mind, why we have seen a jobless recovery in the IT industry is because the IT industry already had more than enough capacity to handle the increase in demand once the recession ended. It's not because of offshoring (at least not most of it). Greater productivity has undoubtedly had an impact. But in large part I think it's because there were simply enough workers to handle the work.

And Zizka, on the outside chance you're still reading this, Wal-Mart workers are exactly who an argument about tariffs are about. Tariffs are taxes, and like sales taxes they are regressive. If you want to increase tariffs to help protect manufacturing jobs, you are taking money from most Americans (including a huge number of working-class and working-poor Americans) to keep intact the jobs of a few (and many of those few, like steelworkers, earn a hell of a lot more than a Wal-Mart worker or a waitress at Applebee's does). It may be that in some cases the math for this makes sense, but I'm hard-pressed to see how it can be considered just.