A Thought (Requesting Feedback)

Economics Economics Questions
1

I just wanted to bounce some ideas off you guys and see what you think about a topic I’ve been thinking about for quite some time. The topic at hand deals with firm expansion, namely why economies of scale doesn’t yield a socialist economy where one firm has complete control over every single market. Now there are many well-known explanations, such as transaction costs, managerial diseconomies of scale, product differentiation, etc. Of course, there’s also the Austrian explanation, the calculation and coordination arguments, which explain why this final condition, namely the socialist economy, is entirely untenable. But the actual process, namely why the transition towards the socialist economy is inherently problematic, has always been somewhat vague to me. In other words, I understand why a socialist economy simply cannot exist for any extended period of time, but I don’t fully understand why the economy doesn’t naturally move towards this final state (at least, I can’t coherently articulate an argument with any degree of certainty) without referring to the explanations mentioned above (which are not characteristically Austrian).

My theory is basically a deduction from the works of Ludwig Lachmann who explains that each individual firm creates its own capital combination in an attempt to satisfy continuously changing consumer desires in an efficient manner, i.e., without squandering scarce resources. Additionally, the price mechanism attempts to coordinate all of the various capital combinations into a coherent capital structure, and profit/loss represents either failure and/or success in this endeavor. In other words, the price mechanism attempts to match expectations with reality, in a dynamic process of trial and error, by eliminating those firms that have failed to organize their capitals in inefficient ways. It, the firm, begins to create massive capital combinations that are disconnected from actual consumer preferences.

From this, it seems obvious that as firms (capital combinations at the micro level) expand, their capital combinations become larger and larger, therefore yielding relatively inelastic combinations that become increasingly difficult to adjust in the face of continuously changing consumer preferences and technological innovation (new technology, methods of production, etc). Since consumer preferences change at such a rapid rate, it would be seemingly impossible to calculate and forecast such changes for any extended period of time, making relatively smaller capital combinations (which are relatively elastic) relatively efficient (because they easier to adjust). In other words, new data continuously emerges, and entrepreneurs are expected to deal with such uncertainty and act accordingly. Economies of scale helps facilitate firm expansion in one direction, but the inelasticity of ever expanding capital combinations serves as a vital check, a counter-balancing force (along with the other explanations in the first paragraph) in the other direction.

Essentially, all firms are doomed in the very long-run. Small firms, in the short run, face a substantial disadvantage due the fact that larger firms are able to utilize economies of scale (they begin to absorb the smaller firms). But the larger firms, eventually, are forced to either sell-off entire blocks of capital (divisions) in order to remain in business, or will be unable to adjust at a fast enough pace, ultimately causing that firm to go out of business and lose access to their resources. And this is essentially what happens when firms go out of business (most obvious during recessions); car companies, for example, have to close down multiple factories and sell-off entire divisions in order to remain in business. I also believe that this theory fits in well with the Austrian theory of cycles.

Now it’s just a thought, which I haven’t really thought through (for example, I would have to include changes in time preferences as well. Presumably, a higher time preference would make firm expansion relatively easier compared to a lower time preference), but what are your thoughts? Does this argument make sense? Have you heard this argument before? Do you think that the other explanations are sufficient?

Thanks in advance.

2

This theory of yours makes sense to me. I’d say the new thing that you bring here is the advantage smaller firms may have vs. larger firms. They are more “agile”. When a firm splits up into parts, prices which were internal and “unknown” now face the market and become market prices. This could actually makes the parts better than the whole.

What I don’t get is that little blurb at the bottom about how a higher time preference would make it easier for a firm to expand. I thought a lower time preference would result in lowered interest and increased savings, which would make capital accumulation easier.

Aside:

I used to study fanatically the market action/corporate structure, techonlogy advances, patents, plans, etc… of Intel and AMD. Many people would attribute AMD’s ability to face Intel as a result of them being “smaller” or more agile. The company has since split into two parts (design and manufacturing). Since then, it’s become more obvious as to where their defficencies were, since the manufacturing anddesign parts now face market forces, each on their own. And also, they’ve managed to pull a profit for the first time in a long time. (Less capital rigidity? More efficiency brought in by adaptation to market forces?)

3

an interesting idea, but i am not sure that larger firms are inherently less flexible than smaller firms.

i’m thinking about it this way.

let’s say we’re talking about a car manufacturer. last year, muscle cars were the hot product. but this year suvs are in.

will the car manufacturer find it very costly to switch production from muscle cars to suvs? not if there are significant economies of scope between producing both products. then the average cost of producing and distributing suvs is lower simply because the firm is already producing muscle cars.

as a result, the muscle car manufacturer will find his production costs to be lower than a new, small firm that is just entering the automotive industry.

in that respect, it sounds like there can be cases where large firms are more “flexible” than small firms. so it doesn’t seem to me like there is any fundamental rule that the size of the firm inherently limits its ability to switch to producing new or modified products in response to changes in consumer preferences. in reality it really all depends on the extent of economies of scale, economies of scope, and how much consumer preferences actually do change.

as far as why one big firm doesn’t produce everything, i would say it has more to do with the fact that economies of scope and economies of scale don’t extend forever than with “flexibility” per se.

4

I think Peter Klein advances a similar thesis in his new book.

5

I can offer some anecdotal evidence to support this. I work for one of the largest companies in the technology sector, with roughly 95,000 employees. What you just said above completely describes what I see on a day-to-day basis. I work in the validation department and our job is to write, run and debug failures from test software that exercises a computer processor. This is an enormously complex task and deciding how to do it is very hard. The difficulty is multiplied many times by the fact that everybody has a different idea on how it should be done. Those who have the most say (management, principle engineers, technical leads) are most experienced but their experience is not always the most relevant to the decisions they make. That is, it is always the case that even though I’m less experienced than the decision-makers above me, I still know vastly more about what I’m working on than they possibly could, since I am working on one small piece of the problem in-depth where they are trying to make decisions that affect the entire problem. This means they have to do a great deal of interpolation about what I’m seeing, which often gives them a distorted picture of what’s really going on.

Where this expresses itself concretely is in the proliferation of test software and supporting software. Huge amounts of resources are poured into writing softwarede novo where industry-standard software could easily be used, instead, at a fraction of the cost. Too many resources are devoted to relatively unimportant support software that has a snazzy interface that can be sold to management and too few resources are devoted to important tasks - such as freestyle hacking of the processor. This is because the incentives are centered around impressing the chain of command. We talk about “internal customers” as if the support software we write is being sold on the market but this is a joke, when you think about it. There are not internal customers, there are just a bunch of internal bilateral monopoly relationships, arranged by management. We only get market feedback once every two years or so when a processor we just got done testing goes out into the market. And even then, any substantial feedback is silo’d and hushed up because bugs in the processor that are found by customers are treated as corporate top-secret - only the engineers and managers actually working on the team responding to the bug have any knowledge that it even exists. This further isolates the wider management and engineering departments from market feedback.

To put this all back into the words you said above, we are trying bazillions of combinations of test configurations for the processor but we are taking very little, if any, feedback from the market, leading to massive misallocations of resources. I’m not saying I have the solution, either, it’s always easy to sit back in an armchair and criticize but a lot harder to actually do something. However, I think the overall mistake is the attempt of management to insulate the processor testing department from the (real-world) market rather than trying to maximize its exposure to the market. Buy industry-standard tools if they are available, even if your engineers think they could write something slightly better in-house. Use industry-standard file formats for all your data, period. Try to involve as many engineers as possible with end-customers so they can have an intuitive grasp of the market-reality. Try to associate engineering decisions that are made with their consequences in the market, rather than on arbitrarily-chosen metrics that happen to sell to the chain-of-command.

Clayton -

6

I think you’re over-thinking this. To me each of the above are easily translatable into any of the others. The way I see it, the main problem with scale (of both states and firms) is the large separation between the human actor (as the main incentive-driven agent) and the market (pricing) forces that reward/penalize his actions. The more an entity is comprised of actors driven by command (rather than the profit incentive) the likelier it is to fail. The larger the entity, the more it is comprised of the former and less of the latter, by definition.

Z.

7

This can be observed in the economy. I think it is a decent explanation.

Obviously, information technology makes it possible for large firms and states to communicate and coordinate much more efficiently than in Mises time, but ultimately large firms end up needing constant subsidy and privilege from states in order to survive, just as large states support small states, until those large states themselves face collapse (where we are headed now).

It’s a shame there aren’t more business people interested in writing about AE, because talking about entrepeneurship and engaging in it are two different things. I had a miserable time trying to read Klein’s book, even though I am incredibly passionate about the subject matter.

8

I never bought this ‘economies of scale’ idea as a justification for large firms - since waaay back with microeconomics - and the more I come to think of it, the less apealing I find it as an argument of economics, and not engineering.

Any number of small firm could pool certain resources, financial or otherwise to produce anything, should that be required. History is full of such instances. Essentially the customer can be served by either large firms, or a myriad of small firms, cooperating where necessary, competing otherwise. But it a large firm is terribly inefficient form the managerial point of view, a market of small firms isn’t. Actually, I’m growing more and more convinced that a market of large firms only emerges with strict licensing requirements, or lese where entrepreneurs among the population are few.

9

Different firms can not internalize externalities, reduce transaction costs and guarantee supply as well as a centralized firm can. So such economies of scale may not pay unless there is central coordination and equity holding. A cartel can be thought of as half-way between several independent firms and an integrated corporation.

Questions of ownership of large investment capital are also harder to answer unless the capital is all owned by a single firm (which, itself, may be split into stocks among multiple capitalist-owners, obviously)

10

You are right. Economy of scale is 100% unrelated to ownership division. This also applies to the supposed increasing inflexibility idea. I still see no reason to believe any of this. Trade is trade. Hiring an “employee” and outsourcing are functionally identical, for example.

That is the same fallacy flipped. It makes no difference either way per se.

11

Yes, a lower time preference is an elevated supply of savings, which allows for more capital intensive methods of production, also known as expanding/lengthening the structure of production. But lengthening the structure of production refers to an expansion in the division of labor and capital (increased specialization), i.e., vertical disintegration. Hayek claims that the best way to think of this lengthening process is to “imagine one firm becoming two (Prices and Production).”

Of course. Asset specificity, or the degree of substitutability amongst its capital goods (if the capital goods are flexible themselves), is certainly a relevant variable that would influence the firm’s elasticity, i.e., its ability to adjust to ever changing consumer preferences, new production methods, etc. I don’t think that this contradicts my theory as much as it adds an additional layer, an extra dimension. I think that my theory is logically sound, all other things equal.

12

So then why do firms outsource at all? If outsourcing and vertical integration were in every way identical, then the size of the firm, the volume of its operations, would be absolutely immaterial. An economy comprised of (a) two major conglomerates would be “functionally identical” to one that is (b) comprised of millions of small and individual producers with no market power whatsoever (the emergence of either condition (a) or (b) would be entirely coincidental). But this is another question altogether (see Coase on Transaction Costs–there are costs associated with using the market).

We know that large firms have certain structural advantages; the volume of their operations allow them to make relatively favorable transactions where they can lock in lower average prices, and their capital intensity allows them to outpace their fixed costs by producing a large output (economies of scale), therefore lowering average costs and allowing them to charge lower prices. There are countless historical examples of this (U.S. Steel, Walmart, Microsoft, IBM, etc).

Now, the question is, why do these large firms eventually go out of business when they have such structural advantages? How can the smaller competitors (often referred to as the “fringe”) outcompete large firms and eventually eliminate them from competition when they face such significant structural disadvantages? This is the matter at hand; this is what I’m trying to answer.

(Again, I’m not as interested in explaining why firms expand as much as I’m interested in explaining why they can’t expand forever. But the two issues are, of course, closely related).

There’s a conflict of interest; there would be an extreme incentive to chisel (the problem with cartels in general). Additionally, a cartel of small producers may be unable to outcompete a single sngle large firm.

13

An interesting choice there: a large firm with an “internal equity market”, i.e. publicly traded, vs. many small firms. In theory capital is mobile in both cases, and can be assumed to allocate efficiently. The problem with the large firm is that it employs people, whereas many small firm outsource almost everything. Now, the moment you employ someone, you cut of the market for that service almost right away, freezing data the way it was at the time of employment. It is more difficult to adjust whatever service you’re employing someone to do, than to just switch an outsourcer.

Plus eventually it all comes down to relative advantages in production: which is the smallest unit that can display - and make use of - relative advantages? The individual. A market of many small firms does just that: it creates a market where specialized individuals sell their services, to each other and to the public, whereas large firms are internally socialistic, no matter how you cut it.

14

That is the same fallacy flipped. It makes no difference either way per se.

I see a difference when it comes to a single decision: employ or outsource? As I said above, relative advantages can be cut down to the individual. In theory, every individual specializes in one field, and it should be good practice to rely on others for every other service. Hence, employing someone for something would seem to me tantamount to cutting, for a time, the market for that service and taking upon yourself at the least the management of that service. Outsourcing is much more flexible and you have no management issues: you either like the product for the price being given, or not.

15

There’s a conflict of interest; there would be an extreme incentive to chisel (the problem with cartels in general). Additionally, a cartel of small producers may be unable to outcompete a single sngle large firm.

I’m not saying cartels of some kind. Let me take an example: syndicated loans. Say there are no large banks but only wealthy individual who loans out their own money. Now, some city hall might need a lot of money, and it can get that sum in two ways: form a large bank, who’ll be taking all of the risk, or from 20 individual, each bearing a small part of the risk. This is not a cartel, its just a way in which many small firms can combine monetarily to provide ‘economies of scale’.

16

That is because you are defining employment as inclusive of “management” and outsourcing as not inclusive, though you can send someone a bi-weekly paycheck for a year without looking at anything but the “product”:

You shouldn’t assume that everything is done for good reason or that it must be a particular reason if any.

You are using competition models and examples both affected by legislation specifically designed to create advantages for “employment” versus “outsourcing” and vice versa. I’m not disputing that the state can edict one particular action to cost more or less than another (down to the individual level). I’m disputing that there is any relevant tangible difference between the kinds of hire contracts in the nomenclature.

It has to be reducible to the individual level and still make sense. I’m thinking of a firm being a bundle of individual transactions and a cooperative of firms being a bundle of a bundle of individual transactions. Let’s say the cost of paper work does not scale with revenues. Not only can small firms share the structure of doing paper work to gain the benefit, the large firm can also. Using economy of scale logic every firm should always be consolidating with every other firm regardless of the nominal size of each firm. There are more or less efficient ways to do this, but the efficiency distinction is made not by the distinction of firms.

17

Not only can small firms share the structure of doing paper work to gain the benefit, the large firm can also. Using economy of scale logic every firm should always be consolidating with every other firm regardless of the nominal size of each firm.

Not necessarily. There is the question to which expansion is useful, whether or not you have the people available for handling a larger and more centralized structure, the nature of the product and production processes, among other things. Centralization and consolidation is not necessarily better, but it does have advantages.

18

That is because you are defining employment as inclusive of “management” and outsourcing as not inclusive, though you can send someone a bi-weekly paycheck for a year without looking at anything but the “product”

Sure, this would be a problemof definition, not substance. But still, I think you would agree that a large firm - as measured by staff, not equity - assumes upon itself the management of many services it could outsource.

19

That is almost my point. Operational technique is the competitive difference, not economy of scale. But the problem with leaving everyone to his own devices, though it may be cheaper, is conflicting interest. For example, the CEO selling short and dunking the company. I’m pretty sure the purpose of management bureaucracy is to keep people in line, not “increase productivity”.