Oligopoly is a market structure characterized by a small number of relatively large firms that dominate an industry. The market can be dominated by as few as two firms or as many as twenty, and still be considered oligopoly.
http://www.amosweb.com/cgi-bin/awb_nav.pl?s=awb
Our resident Economics Expert, Telerion, was claiming in another thread (http://www.timeforchess.com/board/showthread.php?threadid=94167&page=6) that the US economy, particularly those industries which were "immigrant rich" was competitive. He was also ridiculing my claim that "Few markets in the US are truly competitive; most have oligopolist structures and/or tendencies" by asserting that most economists think otherwise (probably true, but a fallacious argument nonetheless).
In researching his claim that the meat packing industry was competitive (it ain't), I found this from the US government: These measures are known as four-firm concentration ratios,
or CR4.
CR4 in steers and heifers is quite high—four firms account for
nearly 80 percent of purchases, in contrast to the average CR4 of
40 percent across all U.S. manufacturing industries. http://www.ers.usda.gov/publications/agoutlook/jun2000/ao272i.pdf
Does an economy where the average industry has four firms control 40% of the market look more like one dominated by competition or by oligopolies? And could this market structure be a major reason why the US economy is very resistant to price deflation (the Consumer Price Index has been rising virtually every year since 1941)?
Originally posted by no1marauderWith reference to the Australian banking sector and Oil industry I think the argument for oligopoly being good for business is a valid one. The behavior of our 4 biggest banks could best be characterized as a union of business interests, in that they have been very careful to avoid predatory tactics amongst each other. This banking oligopoly ensures a basement price for financial products and services that in the long run protects the market from cannibalizing itself into extinction. The oil companies over here (Caltex, Shell, Bp and Mobil) are also very effective in moving their pricing cycles in sync with each other. They seem all too willing to practice self control and not try and drive anyone in the "club" out of business.
Does an economy where the average industry has four firms control 40% of the market look more like one dominated by competition or by oligopolies? And could this market structure be a major reason why the US economy is very resistant to price deflation (the Consumer Price Index has been rising virtually every year since 1941)?
Originally posted by no1marauderCompetition
Oligopoly is a market structure characterized by a small number of relatively large firms that dominate an industry. The market can be dominated by as few as two firms or as many as twenty, and still be considered oligopoly.
http://www.amosweb.com/cgi-bin/awb_nav.pl?s=awb
Our resident Economics Expert, Telerion, was claiming in ano ...[text shortened]... price deflation (the Consumer Price Index has been rising virtually every year since 1941)?
Originally posted by no1marauderAre we really debating this? Or are we discussing the fact that you lost the debate to telerion - and were psychologically wounded - in the other thread at the same time as coming across as a shrill, stubborn, self-righteous plonker?
Our resident Economics Expert, Telerion, was claiming in another thread [blah blah blah] Does an economy where the [...] has four firms control 40% of the market l[...] one dominated by competition or by [...] could this market structure be a major [...] US economy is very [...] deflation (the Consumer Price Index has [...] every year since 1941)?
Which is it to be? Let's hone the terms of reference at the get go, shall we?
Originally posted by telerionSo when the government says that the CR4 in the average US manufacturing industry is 40%, that's "bad research"? Or is it "bad research" to claim an industry is competitive when its CR4 is over 80%?
Just as in the other thread, if there is a considerable amount of interest in the topic, I'd be happy to take it up. On the other hand, if it's just going to be more of no1 trying to piss me off with insults and bad research, then I'll save my blood pressure and skip it.
Originally posted by no1marauder"Recent concentration trends in meatpacking can be defined in terms of livestock procurement—the share of steers and heifers purchased by the four largest steer and heifer packers, and the share of slaughter hogs purchased by the four largest hog packers.
http://www.ers.usda.gov/publications/agoutlook/jun2000/ao272i.pdf
These measures are known as four-firm concentration ratios, or CR4.
CR4 in steers and heifers is quite high—four firms account for nearly 80 percent of purchases, in contrast to the average CR4 of 40 percent cross all U.S. manufacturing industries. Moreover, local market oncentration may be higher, because slaughter cattle usually are not shipped far and many producers may see buyers from only two or three nearby packers. The other striking feature of steer and heifer CR4 is the increase—from 36 percent in 1980 to 72 in 1990 and 78 in 1997. No other manufacturing industry shows as dramatic an increase since the U.S. Census Bureau began regularly publishing concentration data in 1947.
Hog slaughter is less concentrated—the top four hog packers handled 54 percent of slaughter in 1997. But CR4 in hog slaughter has increased sharply, from 32 percent just 12 years earlier. Like other livestock, hogs are not transported far to market, and as a result many producers may have more limited options locally, with a choice of buyers from only two or three packers. Meatpacking has also shifted sharply toward larger plants that annually slaughter at least 1 million hogs or 500,000 steers and heifers. Such large plants, which handled less than a fourth of steer and heifer slaughter in 1980, accounted for over threefourths just 15 years later. Large plants handled 63 percent of all hog slaughter in 1980, compared with 88 percent by 1997.
Shifts in plant size suggest that there may be economies of scale in slaughter, and that scale economies and the resultant shift to large plants may account in part for the increase in concentration. If there are scale economies, then increasing meatpacker concentration may lead to lower meat prices for consumers."
Originally posted by zeeblebotUSDA: If there are scale economies, then increasing meatpacker concentration may lead to lower meat prices for consumers."
"Recent concentration trends in meatpacking can be defined in terms of livestock procurement—the share of steers and heifers purchased by the four largest steer and heifer packers, and the share of slaughter hogs purchased by the four largest hog packers.
These measures are known as four-firm concentration ratios, or CR4.
CR4 in steers and heifers is qui ...[text shortened]... nomies, then increasing meatpacker concentration may lead to lower meat prices for consumers."
They could, but that is unlikely with four firms dominating the market (ask Tel why). In fact since that article was written in 2000, the price of ground chuck has soared from a $1.90 a lb to $2.82 a lb, a 48% increase. http://data.bls.gov/cgi-bin/surveymost
Originally posted by telerionI’ll take you up on this—just for play. 😉 I’ll likely be slow in response, though, since I’m not on here much these days...
Just as in the other thread, if there is a considerable amount of interest in the topic, I'd be happy to take it up. On the other hand, if it's just going to be more of no1 trying to piss me off with insults and bad research, then I'll save my blood pressure and skip it.
Tel, I’m doing all of this from a long memory reach (25 years)—
By "the value of their work" I mean the marginal contribution to the total profits of the firm from a worker's labor input (i.e. "marginal value of product of labor" ). This is a basic economic concept, not a phrase I'm pulling out of my behind.
The “basic economic concept” applies to perfectly competitive labor markets with homogeneous labor and negligible transaction costs. In cases where transaction costs are significant (especially training* and monitoring costs; and especially in industries where there is significant teamwork involved in generating a unit of output) it is more efficient for wages to be assigned to jobs rather than individual workers. (In cases where the learned skills are not fungible across labor markets, the worker may also make a considerable investment in acquiring firm- or industry-specific skills.)
In such cases, the standard w = mrp equation becomes a double summation (which I don’t know how to reproduce here) across jobs and over time.
Basically, the labor market is a barter market in which workers trade their time, efforts and skills for a compensation package—for simplicity, a wage. When wages are assigned to jobs, workers compete to acquire a job, from the performance of which they receive the assigned wage; workers compete not only at the hiring stage, but in the internal labor market of the firm as well. Firms hire workers not only on the basis of their ability to perform entry-level jobs, but also on their potential to learn higher-skilled jobs. Workers may also be searching for employment that allows them to advance. Thus, there are search and screening costs on both sides of the market (transaction costs reflecting heterogeneous labor and employment opportunities).
All of this is industry-specific. In the relevant industries, labor becomes a quasi-fixed factor of production (Oi); wage rates become job-rates (Thurow); and internal labor markets develop (Doeringer, Piore, Williamson, etc.). The connecting link between the otherwise parallel human capital theory (Becker) and internal (or dual) labor market theory is transaction costs (Stigler, Demsetz, Alchian).
Granted that the particular kind of industry you were thinking of in your reply to twhitehead likely does not fit the above description; low-skilled labor may much more reasonably be treated as homogeneous.
________________________________________
The (in)famous so-called “Cambridge Capital Controversy” is partly over whether or not the mrp of capital (K) can be actually calculated in any case. Or, what exactly is a “quantity of capital”?
The standard profit-maximization equation is X = PQ – wL – rK.**
However, as a practical matter, the quantity of K is measured by it’s monetary price (r). That is—unless one assumes that K is essentially “buttery” (Joan Robinson’s term, I think)—the measure of the quantity of capital (let’s call it M) just is rK. M = rK. The standard equation then becomes X = PQ – wL – M. And, of course, it makes no sense to attach another r to M [rM = r(rK)].
Now, if I recall, the standard neoclassical definition of “labor exploitation” (as opposed to a Marxian one) is a case where labor is paid less than its mrp, and capital more. But—how does one calculate such a thing? Does one simply believe that such cannot be the case? Does one simply assume market conditions that would not allow it?
Can one simply determine that w = P(dQ/dL) [Have I got that right? Except for the partial-derivative symbol?] and, doing so, treat the mrp(k) as the residual? But what about industries subject to the complications I outlined above?***
[I recall a paper by Lester Thurow years ago (in the QJE?) wherein he attempted to determine whether there was any such labor exploitation in the economy generally. His conclusion was “Yes”, though I recall thinking that his attempt was pretty small beer (at the time: I couldn’t begin to make such a judgment now).]
Is this fundamental feature of neoclassical micro-theory really empirically untestable? Does one simply need to accept it, as the Austrians might claim, as a synthetic a priori principle?
______________________________________
Other considerations—
Even in a market where labor is relatively homogeneous, workers (except for day laborers) attempt to secure a job that will provide a return on their labor over time; and pay-rates are generally assigned to the job (except in cases of pay for piece-work). Also, low-skill does not necessarily mean no-skill; or that the (low) skills required are not learned on the job. There may be significant technology differences across even low-skill industries (e.g., fruit-picking versus meatpacking—although the latter, at least, does have some higher-skilled jobs) that effect the production function; whether or not such effects are significant (and, thus, whether or not it is reasonable to impose the same production function across industries, at least insofar as the technology factor is concerned) is an empirical question.
What happens when simplifying assumptions in theory yield robust predictions—but are the subject of anomalies in application? Do the social sciences, such as economics, treat such an affair differently from the physical sciences?
The “value” of labor in the neoclassical definition has nothing to do with justness or fairness—or any other normative judgments (the “scare quotes” simply indicate differing notions of the term “value”, nothing more). To say that economic (Paretian?) efficiency in any way “trumps” such normative considerations would be itself a normative judgment. [I cannot imagine you arguing that child-labor laws, to draw a more extreme example, ought to be abrogated as an unwonted intrusion into the labor market, whether or not the unavailability of children to work in industry drives up wages and product prices.]
Once more expansive labor immigration is allowed, how does this affect conventional assumptions of a full-employment economy in labor market models? The definition of the labor market is now expanded to include those (potential immigrants) who are unemployed vis-à-vis the “host” labor market(s). Given a relatively fixed capital arrangement, competition amongst that larger supply of labor will thereby drive down wages and (depending on price elasticity of demand in the product market) product prices. Although I never was a Marxian (being neoclassically trained), this does remind one of Marx’s “reserve army of the unemployed”.
The Austrians view all these (and other such) complexities as making empirical economics (and econometric models) simply untenable. [I am not convinced, however.] Their answer is just to let markets work without intrusion (other than , perhaps, basic governance structures—such as rights of property and contract—without which no market can really function). My answer might be that such complexities justify market intrusion based on non-economic normative considerations (such as the socio-political impact of any given income distribution, and redistribution).
In context of the original thread: does the income redistribution effect from lower-skilled/lower-income workers to higher-skilled/higher-paid workers (even if small) itself warrant any policy consideration? Is there any redistribution effect in terms of labor and capital (i.e., stockholders)? Does it matter? [My advanced macro prof was interested in the labor/capital distribution of income; Friedman thought that was silly.]
______________________________________
* Particularly in situations where OJT is the most cost-effective alternative (to, say, external schooling or “vestibule” training).
** I have substituted “X” for the usual the symbol for profit; I have dropped the natural resources component. (Note: I couldn’t begin to do the calculus anymore, maximizing the Lagrangian: have forgotten all of it. 🙁 )
*** “Do similar valuation problems arise for heterogeneous labor? The crucial difference with capital is that there is no theoretical presumption that competition will equalize wages across different types of labor, in the way that rates of return will equalize (adjusted for risk) across investments in different capital goods/industries. To the extent that heterogeneous labor reflects differences in human capital, the valuation problems for the neoclassical parables due to interest rate changes are only exacerbated.”
—Cohen & Harcourt; “Whatever Happened to the Cambridge Capital Controversies?”; Journal of Economic Perspectives, Winter, 2003; p. 204, footnote 3; http://www.econ.yorku.ca/~avicohen/Linked_Documents/JEP_Cohen_Harcourt.pdf
________________________________________
EDIT: continued for some other quotes...
“While neoclassical economics envisions the lifetime utility-maximizing consumption decisions of individuals as the driving force of economic activity, with the allocation of given, scarce resources as the fundamental economic problem, the “English” Cantabrigians argue for a return to a classical political economy vision. There, profit-making decisions of capitalist firms are the driving force, with the fundamental economic problem being the allocation of surplus output to ensure reproduction and growth (Walsh and Gram, 1980). Because individuals depend on the market for their livelihoods, social class (their position within the division of
labor) becomes the fundamental unit of analysis. The potential rate of profits on capital arises from differing power and social relationships in production, and the realization of profits is brought about by effective demand associated with saving and spending behaviors of the different classes and the “animal spirits” of capitalists. The rate of profits is thus an outcome of the accumulation process.7 Robinson argued—citing Veblen (1908) and raising the specter of Marx—that the meaning of capital lay in the property owned by the capitalist class, which confers on capitalists the legal right and economic authority to take a share of the surplus created by the production process.” (Cohen & Harcourt, ibid, p. 208)
“The Cambridge controversies were not a tempest in a teapot.” (ibid, p. 211)
Vistesd, I don't want to break your stride here, but typical neoclassical approaches to labour markets are nearly all dead and buried.
These days, search models are used where firms first decide to open vacancies and then search for candidates, while job-seekers (unemployed or even employed) search for vacancies.
It is assumed that it takes time to find the adequate candidate for the job for the firm and also it takes time to find the right job for the job-seeker. There is then a probability of matching, which differs for both sides of the market according to the mass of vacancies and job-seekers.
In that sense, it is much closer to what you seem to be trying to describe. Search costs serve the function of what you call 'transaction costs' and firms may even indulge in what is called "labour hoarding". For example, a temporary negative shock to productivity may not lead to the firing of workers because the firm knows it will have costs to find and recruit an adequate candidate. I wouldn't called it quasi-fixed factor but I think this is close to what you were trying to say.
Also, wages can be seen as the result of a bargaining process and so the typical neoclassical pricing of w=mrp will rarely hold. The exact wage will depend on the bargaining power of firm and worker. Also, the level of unemployment also affects wages via this bargaining process (affecting both parties' outside options).
This is a quick simplistic explanation, but I just wanted to illustrate that labour economics has changed dramatically in the last decade and a half.
Originally posted by PalynkaVistesd, I don't want to break your stride here, but typical neoclassical approaches to labour markets are nearly all dead and buried... This is a quick simplistic explanation, but I just wanted to illustrate that labour economics has changed dramatically in the last decade and a half.
Vistesd, I don't want to break your stride here, but typical neoclassical approaches to labour markets are nearly all dead and buried.
These days, search models are used where firms first decide to open vacancies and then search for candidates, while job-seekers (unemployed or even employed) search for vacancies.
It is assumed that it takes time to fi ...[text shortened]... illustrate that labour economics has changed dramatically in the last decade and a half.
You are absolutely welcome to break my stride! 🙂 I am happy to be corrected on anything I present here. My last work on this was 1982, when I defended my thesis. [Life changed; I went in other directions; I have a great talent for forgetting—absolutely forgetting—anything that I don’t use regularly for a moderate period: as I say, I can no longer maximize a Lagrangian or derive Slutsky equations—10 years ago I still knew what they were!]
Is it your sense that neoclassical approaches to labor economics are also nearly dead and buried in the U.S.? I had a phone conversation with my thesis advisor a couple of years ago, and my recollection is that he thought that not much had changed in the interim period.
I am also sure that econometric modeling has changed dramatically. I was never an econometrician (focus on micro-theory, especially in labor markets; for my own economic reasons, I had to settle for an M.A., which I tried to beef up by taking the advanced micro, and some of the advanced macro, coursework and opting for the thesis option). I really have to trust the econometricians, and can only look to their underlying theoretical framework.
I am taking a complete break from the other forums, and just thought it would be fun to see what happens here—including your comments! I’ve always been bothered (well, in the intervening years, whenever I thought about it!) by the Cambridge capital controversy. I originally tuned into it in a paper by L.L. Pasinetti. I have always thought that everything about micro theory ought to be subject to empirical testing.
I knew that you are an economist, Pal—what’s your particular focus? And please feel free to correct me on anything—who knows, maybe I can resurrect some of my forgotten knowledge in the process (I know longer have any of my resources, except for Samuelson’s 10th edition, which i cut my teeth on as an undergrad, and kept for sentimental reasons).