Zero-Sum Foolery 1 of 4: Game Theory Gamesmanship
by Sandwichman
Zero-Sum Foolery 1 of 4: Game Theory Gamesmanship
It has become fashionable recently, in denunciations of the lump-of-labor fallacy, to appeal to the notion of a “zero-sum game” in addition to the customary allegation of a “fixed amount of work to be done.” In this manner, pseudo-intellectual poseurs can evoke the urgency and panache of mathematical game theory without knowing the first thing about it.
Here are a few examples:
The main reason the lump of labor theory is wrong is that it is based on the assumption that everything that is going to be invented has been invented, and that therefore economic competition is a zero-sum game, a fight over a fixed lump. – Thomas Friedman, The World is Flat.
The idea that increased participation of older workers will negatively affect employment for younger people is known as the lump of labor fallacy. This fear of displacement is grounded in the assumption of a zero-sum labor market in which every job occupied by an older worker is one less potential job for a younger person. – World Bank, Live Long and Prosper: Aging in East Asia and Pacific (with credit to Munnell and Wu, and Zhang and Zhao).
The argument that jobs taken by low-skilled immigrants are jobs forgone to native-born Americans rests on the erroneous assumptions that immigrant labor can easily substitute for native labor and that employment is a zero-sum game. The fallacy, known as the lump of labor.” was popularized in 1891 by the economist David Schloss, who described it as the erroneous belief that the amount of work was fixed and could be parceled out in various ways. – Susan K. Brown and Frank D. Bean, “Population Growth” in Debates on U. S. Immigration, edited by Judith Gans et al.
One misunderstanding we ought to dispel immediately is the so-called lump of labor hypothesis. This philosophy maintains that there is a fixed amount of work to be done—a lump of labor—so if the elderly can be encouraged to leave the workforce, there will be more jobs for the young. This zero-sum thinking is simply wrong. Economists treat labor as one of the primary inputs into economic output, and the more input, the more output. – George P. Shultz and John B. Shoven, Putting Our House in Order: A Guide to Social Security and Health Care Reform.
As far as I can determine, Paul Samuelson was the first to use this analogy in a 1978Newsweek column on the “Economics of Discrimination”:
Upon thoughtful analysis of the nature of the economic system, economists find that it is essentially not a zero-sum game. Economists call it “the lump-of-labor fallacy” to believe that in any period – 1933 or 1978 – there are only so many jobs: it is false philosophy of despair, economists point out, to insist on cutting down on each worker’s weekly hours in order to spread out an allegedly limited total of work and of income among as many people as possible.
Sandwichman has previously noted in passing the zero-sum allusions but hasn’t paid much attention to them. That is about to change.
Arguably, the most likely assumption of the unidentified non-economists (those nobodies presumed to commit the alleged fallacy) is not a “zero-sum game” but a repeatedprisoner’s dilemma. It is Samuelson, Shultz, Munnell, Friedman and their ilk who plunge zealously into what Anatol Rapoport described as a zero-sum TRAP – which is to say, the conceptual reduction of non-zero-sum games to zero-sum games in order to render them “solvable” as a predetermined type of “problem.” Attributing a zero-sum view to their opponents enables the propagandists to insinuate that their alternative is a bowl of cherries. If zero-sum is win/lose, then non-zero sum must be win/win, right?
The arguments supporting my critique of Samuelson et al.’s fraudulent game-theory gamesmanship are rather involved. So why would anyone want to spend time reading about the refutation of yet another bucket of boilerplate propaganda? Because this one bears not only on crap economic policy but also on crap arms race strategy and crap climate change policy, which I will get to in the second post in this series.
In his preface to Strategy and Conscience, Rapoport recounted an exchange he had with a strategist who had come to his university to talk about “Defense and Strategy in the Nuclear Age.” Overcome with revulsion at the speaker’s clinical detachment in addressing mass extermination, Rapoport asked the speaker, “how would he defend himself if at some future time he were a co-defendant in a genocide trial.” The speaker respectfully replied that “he would plead ‘partially guilty.'” But it was the response of many of his colleagues to his question that rattled Rapoport. They thought the very question was inappropriate and violated the standards of academic discourse. Somehow, even in discussing “the unthinkable,” some thoughts must remain taboo.
Unlike Rapoport’s strategist, the propagandists who recite the lump-of-labor, zero-sum fallacy catechism are unlikely to acknowledge even partial responsibility for promoting economic inequality and social injustice. After all, why should they? They were only repeating what they have heard and have been told to say. They didn’t know what they were talking about.
cross posted with Econospeak
Isn’t part of the problem the fact that at any given time available employment is limited? It can change, of course, but that change is likely to involve major developments, many of which are unpredictable, such as creation of whole new spheres of economic activity through inventions or creative entrepreneurship or, more likely, creation of major demand via governmental action.
Interesting.
Lately, I have been thinking about MAD (nuclear war style) in regards to the star wars, China, NK, Japan in NE Asia and US neocon tilting dreams South China Sea treating China like North Vietnam to get them to the negotiating table with cruise missiles (aka aircraft carrier and B52 raids).
Or India Pakistan nuclear “deterrent” over Kashmir (tomdispatch).
“Isn’t part of the problem the fact that at any given time available employment is limited?”
Not really, no. Do you work in your yard — raking leave, cutting grass, etc.? Fifty years ago, very few middle-class people would think of hiring a lawn-care service to do that stuff. You might hire the neighbor’s kid, if you didn’t have one yourself. If you could hire someone for fifty cents an hour to do your laundry and clean your house, you might seriously consider doing so. When the price is $15 per hour, you might not think it such a good use of your money. There is always work to be done. A limiting factor is the artificial lower bound on what one can pay for the work.
“It can change, of course, but that change is likely to involve major developments, many of which are unpredictable, such as creation of whole new spheres of economic activity through inventions or creative entrepreneurship or, more likely, creation of major demand via governmental action.”
Creating major demand via governmental action usually has an equal, often greater, destruction of demand in other areas. Look at the sugar price supports. The result of overpriced sugar was the creation of major demand for corn syrup via governmental action, and the reduction of demand for sugar cane even more. Why more? Because the artificially high price of cane sugar reduced the efficiency of the market. (Similarly, the artificially high price of unskilled labor reduced the demand for that labor and reduces the efficiency of the labor market.)
“Economists treat labor as one of the primary inputs into economic output, and the more input, the more output.”
Within the context of the quote this is the fallacy of argument from authority. Where is the data?
Sorry about the double post above.
Warren: I was thinking of things like infrastructure repair and improvement but I take your point.
“I was thinking of things like infrastructure repair and improvement but I take your point.”
There is plenty of work to do there, too. But we are already paying people subsistence-level income to do nothing. Why would they take a construction job that pays them about $1 more per hour than they get for doing nothing?
In another post, someone said we should take money from the Defense Department to pay people to do nothing. Of course, then the people working for the Defense Department will be out of work, too. At least they are producing something for the money we pay them. It may not be what you want them to produce, but it is still something.
What we should be doing is taking money out of the welfare programs and putting it to infrastructure repair and improvement. If you want money from the taxpayers, you must work for the taxpayers.
“But we are already paying people subsistence-level income to do nothing.”
Oh? And who is it that is on the dole in this day and age? And how do you come up with the $1.00 difference, even if you could identify these recipients of largess? Not too many years past there were major demonstrations by groups of the minority unemployed for jobs in the, then mostly segregated, construction industry in NY. And when did construction wage get set at $1.00 above “subsistence-level income”?
Who are we paying a subsistence level income to do nothing?
‘And when did construction wage get set at $1.00 above “subsistence-level income”?’
When the illegal immigrants started taking the jobs.
“Who are we paying a subsistence level income to do nothing?”
We do have long-term unemployment insurance well beyond the twenty-six weeks the premiums paid for. That is giving people a subsistence-level income to do nothing. Then we add in food stamps, TANF, housing subsidies, etc.
http://www.budget.senate.gov/republican/public/index.cfm/2012/12/total-welfare-spending-equates-to-168-per-day-for-every-household-in-poverty
Warren,
It matters how you view humans, if for you inanimate things are more important than people, then why should the pentagon be ready to nuke the world if we are going to be overrun?
That $168 per day per each household is a long ago disproved and rebutted red herring.
However, each US resident including you and I is/are paying $1100 for the production run of F-35’s. For the top 100 pentagon shovel leaning “systems” you and each US resident are pay $4400. That don’t mention your part of keeping those systems running another $8800. I used to be an ilsm and know these things about weapons systems.
Then you gotta pay for the soldiers, and dependents and retirees. Also the DVA needs about $180B a year.
In this argument the pentagon mess may be a strawman, but your “systems” money I related above is fabulously profiting a few who can buy congress. The poor are in the millions and not organized to buy congress with their family’s minute take of that $168/day.
Is everyone a moocher? Is so why do you think they mooch?
One of the reasons greed is a vice is because it means a human being idolizes inanimate items. That idolatry keeps the human motivation level at its basest level. Man should be able to expand his horizons and keep his soul!
Are the moochers the government is spending $168 per day more a millstone on the US than the war mongers getting around $1000B (1T) a year?
I think Sandwichman in this article is suggesting that models live and die on assumptions.
If the assumption are based on hypotheses and those hypotheses are not acceptable in the needed “risk”. Then the model is toast.
“Lump of labor theory”, “employment as zero sum game” etc. may be valid assumptions in the minute cases where the model lives in the same restrictions as the hypotheses that base its assumptions.
In the military system acquisition we do all kinds of “case analyses” relying on ground rules and assumptions. The intuitive analyst will determine what the colonel wants the study to find and base the ground rules and assumptions to get what the boss wants.
Same in economists’ papers.
“We do have long-term unemployment insurance well beyond the twenty-six weeks”
We did three years ago, not today.
Warren,
When I went to your Republican party link, I no doubt succumbed to my own biases in in deciding to look for where it was wrong, but there are a couple of obvious ones.
Most people think of TANF when they think of welfare. TANF was 2 percent of their “total”. Redefining terms is only good debating if you don’t expect people to notice.
Most of the programs relate to health spending. My health spending (and most people’s) is through employer provided insurance which is not included in median household income, so the headline is a blatant apples to oranges comparison.
If you are going to provide links as evidence at this site you are going to need to be more discerning.
Let me clarify what the issue is I’m looking at
It is the attribution to unidentified people, without evidence, of a “belief”. Often the subject of the sentence is the belief itself. There is no believer. It is literally, then, a “believerless” belief.
In these examples, allusion is also made to game theory AS IF the writers are basing their claims on some analysis. But that is a lie. These pundits have done no game theory analysis and they refer to none.
Samuelson et al. are aware of no game theory analysis that supports their claim because there is none. In fact, game theory would expose the claims they are making as fraudulent.
“That $168 per day per each household is a long ago disproved and rebutted red herring.”
Well, let’s look at the numbers, shall we?
The most recent data from the Census Bureau says there were 46.7 million people in poverty. https://www.census.gov/hhes/www/poverty/about/overview/
For FY14, we spent $456.4B on welfare. http://www.usgovernmentspending.com/year_spending_2014USbn_17bs2n_40#usgs302
That’s $9773 per person.
The poverty rate for a family of four in 2014 was $23,834.
http://talkpoverty.org/basics/
But we were spending $39,000 on welfare for that family of four!
33,283k of those people in poverty are in 9,733k families (or subfamilies), which gives an average of 3.45 people per household. So, 3.45 people per household, at $9773 per person is $33,500, which is 41% over the poverty line for a family of four. If you actually have a family of four, it’s $39,092 per year — 164% of the poverty line.
Now, according to the BLS, the median hourly wage is $17.40.
http://www.bls.gov/oes/current/oes_nat.htm#00-0000
At 40 hours per week for 50 weeks per year, that’s $34,800. Take out the FICA taxes (assume no federal or state income tax) at 7.65%, and that person is left with $27,596.40 for the year — only 82% of what we are spending on welfare per median household of 3.45 people.
But you are right — with 3.45 people per household, it only comes to $91.78 per day.
So, if you are a family of four, with a single earner making the median hourly wage, you bring in $27,596.40. If you are NOT working, taxpayers spend $39,092 on you and your family. (BTW, the median construction laborer makes $15.34 per hour.)
Warren is arguing that jobs are not zero sum because people don’t need to eat, therefore it is always possible for someone to find a job even if the pay is vanishingly small. In other words, we could have zero unemployment and no freeloaders as long as a certain number of people suffer and die from starvation, exposure and other such poverty related maladies every year. It’s a common argument, and popular, as long as the party starving, freezing and so on is someone else.
The thing is that at a given instant there are a certain number of people working. On a very short time scale, let’s say a few seconds, the odds are that there are only so many jobs and so many people seeking employment. Over a longer period, a day, a month, a year, the number of jobs and the number of workers can increase or decrease. That’s why the multiple-round prisoner’s dilemma game is a better analogy than an arbitrary zero-sum game.
In a way, it is like Zeno’s paradox. At the infinitesimal scale motion is impossible, but infinitesimals are smaller than any real numbers. In the real world, motion is quite possible. One wouldn’t expect the ancient Greeks to get this right.
Whether due to game theory, zero sum, or whatever, it appears that the income gap is severe and increasing and that wages generally are stagnant. The prospects for young people entering the job market are bland to discouraging. That is the problem regardless of what is causing it.
Appeal to authority is a well worn logic fallacy.
I see Fox News when I am on the treadmill at the gym.
They always have panels, all their panelists are authorities! Not!
I observed to my conservative brother that I use Fox News when I want examples of logic fallacy, they always supply!
Another tea party arguer I casually engage often uses the introduction to a point: “you have to agree…” which I often respond “no I won’t go there…”
Seems my response might be caught in the spam filter, since I have server links to sources for the numbers I use. Please check. Thanks.
Warren it wasn’t in Spam or Pending but in Trash. Why exactly I don’t know and why my attempts to bring it out have yet to be successful. But bottom line don’t try to put more than three links in a comment. That WILL get WordPress to put you into moderation of one type or another. Break it into multiple comments if you must.
Now it doesn’t seem to be anywhere. Sorry. You may need to break it into chunks.
Thanks. I’ll re-do it.
“Well, let’s look at the numbers, shall we?” Warren
“For FY14, we spent $456.4B on welfare.” Warren
“The most recent data from the Census Bureau says there were 46.7 million people in poverty.” Warren
“That’s $9773 per person.” Warren
Our boy, Warren, has taken a gross government spending figure and divided it by the numbers of people in a defined category or class. He implies that total gross “welfare” spending, however that may be defined, was paid as stipends to individuals. He has gone well beyond a misunderstanding of government spending statistics. He has passed through the world of logical fallacy and now demonstrates that he places an idea into drive before engaging his brain.
Warren,
Your numbers are both off topic and not credible.
I will just give one example of the absurdity. Included in the category of “welfare” by the website you cite are earned income tax credits and child tax credits. The Center on Budget and Policy Priorities reports that 9.4 million people were “lifted out of poverty” by EITC and CTC. So those 9.4 million would not be counted in your 46.7 million people in poverty.
That is just ONE example based on the first line item in the detailed breakdown. When you divide a number by another number, you have to know what those numbers mean for the answer to be meaningful. Just because the result of a calculation looks impressive doesn’t mean it means anything.
Sandwichman:
You have a lot of patience.
Someone is wrong on the Internet: “In a way, it is like Zeno’s paradox. At the infinitesimal scale motion is impossible, but infinitesimals are smaller than any real numbers. In the real world, motion is quite possible. One wouldn’t expect the ancient Greeks to get this right.”
Maybe it’s me (being wrong on the Internet), but I thought Zeno’s Paradox was about continuous motion being an illusion (like the motion seen on a screen at several frames per second), which Quantum Mechanics says it is, so the ancient Greeks (or at least Zeno) got it right. I am not a historian, but was pleased to see the historian who blogs at “Ex Urbe” agrees that this is what Zeno meant.
Sorry for the off-topic comment, but I hate to see Zeno dissed for coming up with the idea of Quantum (i.e., discrete rather than continuous) Mechanics over 2000 years ago.
“Included in the category of ‘welfare’ by the website you cite are earned income tax credits and child tax credits. The Center on Budget and Policy Priorities reports that 9.4 million people were ‘lifted out of poverty’ by EITC and CTC. So those 9.4 million would not be counted in your 46.7 million people in poverty.”
Excellent points. Thank you.
(But if tax credits to corporations are going to be considered “corporate welfare,” then we have to leave in the “Payment Where Earned Income Credit Exceeds Liability for Tax” and the “Payment Where Child Tax Credit Exceeds Liability for Tax.”)
So let’s add those people in. That’s 56.1 million people.
So now let’s divide up that $456.4B into 56.1 million people. That gives us $8,135 per person. For a family of four, that’s $32,542. That’s the equivalent of a $35,238 income subject to the 7.65% FICA tax. At 40 hours per week and 50 weeks per year, that’s $17.62 per hour, which is still more than the median hourly wage of $17.40 reported by the Bureau of Labor Statistics.
“He implies that total gross ‘welfare’ spending, however that may be defined, was paid as stipends to individuals.”
I made no such implication.
Warren,
“So let’s add those people in. That’s 56.1 million people.”
And if you multiplied it by sixty-six and divided by eight it would still have nothing to do with zero-sum games and economists’ bogus arguments from authority.
Quite so. We got a bit off-topic running down Jack’s “creation of major demand via governmental action” rabbit-hole.