Abstract or "absentee" ownership, especially abstract ownership of capital, is an externality to markets: it contradicts a fundamental assumption of "free" market economics. Capital ownership as an externality is largely ignored not just by Randian (Chicago/"freshwater") economists, but also Keynesian ("saltwater") economists. Economists assume that if there's an economic case for the allocation of capital to an endeavor, capital will be allocated to that endeavor. Because the capitalist class is internally competitive, and because they are largely rational (either by intelligence or selection), this assumption is often at least partially met; the externality of ownership is often mostly irrelevant. However, under certain circumstances, the externality dominates economic behavior; in all circumstances, the externality biases economic behavior to some degree.
To understand and justify this assertion, let me first define my terms.
Abstract ownership is ownership is ownership (in the common, intuitive sense) without physical possession and use, in contrast to concrete ownership, ownership with physical possession and use. I physically possess my car (it's in my driveway; I alone have the keys), and I am its sole user. I therefore hold concrete ownership of my car. My landlord, however, does not physically possess my apartment — indeed he cannot usefully physically possess it — therefore his ownership is abstract. Additionally, he borrowed money from the bank; until the loan is paid back, the bank still abstractly owns the money even though my landlord "physically" possesses and uses it; this abstract ownership entitles the bank (and its shareholders) to collect interest. Furthermore, I have deposited money in the bank, which the bank has loaned to my landlord; my abstract ownership of that money obligates the bank to pay interest to me.
Capital is labor time expended in the present instrumentally for a benefit in the future. When Acme construction built my apartment building, labor time had to be expended at that time to collect and fabricate the materials and assemble them into a house. All that labor time had to be paid for at the time the house was built: the workers who built the house needed to eat then, not now (when I myself am deriving benefit from the house), and all the people who worked directly or indirectly to feed those workers needed to be paid to sacrifice their immediate consumption then, not now.
Abstract ownership entitles the owner to collect economic rent for the use of the object owned. In a truly free market (or a market that strongly resembled a truly free market) the rent (price) would fall to the cost of creating, maintaining and reproducing the object owned, including the owner's administrative time. In a truly free market, then, abstract ownership should be just another job. But we know empirically that abstract ownership is not just another job: abstract owners of capital consume or control orders of magnitude more labor hours than they contribute. Indeed it is often the case that abstract owners of capital consume orders of magnitude more labor hours than an ordinary worker while doing no work whatsoever. Therefore we must conclude either that the derivation of our microeconomic conclusion that prices always fall to cost is fallacious, or that one of the assumptions of that derivation is not applicable to the real world. The derivation is not fallacious, so there must be an assumption that does not hold.
It doesn't matter what "natural" (not intentionally imposed by human beings) barriers exist to the abstract ownership of capital. Microeconomic theorems prove only that it will take more or less time to overcome those natural barriers and achieve sufficiently wide distribution of abstract ownership of capital to reduce its price to its cost. Even if that time was on the order of a millennium, we should see over just decades a trend towards wider distribution of abstract ownership of capital and a decrease in the rent collected by its owners. However, the observable trend is (on the whole) towards the concentration of capital into fewer hands, and an increase in the rent collected by its owners.
There is a small short-term, local, and individual selection pressure against saving for capital formation and therefore indirect selection for immediate consumption. This selection pressure is large enough that non-human animals — whose evolution is driven by extremely short-term selection pressures — do not engage is large-scale capital formation; their capital formation is limited to very small-scale endeavors such as beavers' dams or birds' nests, or endeavors that have accumulated over tens or hundreds of millions of years (and very narrow specialization), such as beehives, anthills, and termite mounds.
Human beings, however, by virtue of our big brains, can overcome short-term, local and individual selection in favor of longer-term, global and social selection. We can "think our way past" some selection pressures and away from a local "fitness" maximum to a higher global maximum. The vast and (at an evolutionary time-scale) instantaneous social and economic progress we have made as a species is overwhelming evidence that our ability to overcome local selection pressure is not merely present but enormously powerful. Achieving an optimally efficient distribution of capital ownership — defined as a distribution that causes the price of capital to be at its cost — should be not just possible but inevitable.
Point out any natural (non-human) obstacle to the efficient distribution of capital, and I'll show you a strategy to overcome it. I'm not that smart: if I can do it, anyone else can. Anyone who successfully implements an idea should be able to transmit that idea to everyone: the idea, by virtue of its very success, should become ubiquitous. Take, for instance, economy of scale: owning a small amount of capital isn't very beneficial, but owning a large amount of capital is disproportionally beneficial. This natural barrier is easily overcome by creating a social group that pools its individually small savings into a large enough size to be mutually beneficial to the members of the group. Indeed this is precisely the idea behind the mutual fund.
The only explanation for capital ownership to be distributed not just sub-optimally but the complete opposite of an optimally efficient distribution must be the presence of some externality. Moreover, this externality must not be natural (nature has never shown an ability to out-think human beings) but rather artificial, imposed by human beings.
Abstract ownership requires social coercion. I can physically protect my car by myself, by locking it and holding the key, but my landlord cannot physically protect my apartment: the whole point of renting it to me is to let me use it. (Likewise, Avis cannot physically prevent me from stealing a rental car: their whole business consists of giving me the key and physical possession of the car.) My landlord must necessarily depend on the social construction of law and custom that enables him to demand that the sheriff, who has a gun, to coercively evict me from my apartment if I fail to pay the rent.
The primary means that the capitalist class uses the socially constructed basis of enforcement of abstract ownership is indirect: law and custom in a capitalist society strongly protects the capitalists ownership of capital, but does not protect a worker's ownership of his or her actual labor; it protects only his ownership of his labor power. Without access to socially accumulated capital, an individual worker cannot use his own labor to survive, therefore the pressure of immediate necessity forces him to relinquish physical ownership of his actual labor in return for physical ownership of his labor power. Therefore, possession of actual labor must be guaranteed abstractly, or it doesn't exist. But capitalist governments do not guarantee abstract ownership of labor. The differential enforcement of abstract property rights constitutes a coercive externality on the market for capital ownership, making it un-free.
Does differential enforcement constitute coercion? It seems unobjectionable that some form of coercion is justifiable and socially acceptable in a "free" society: few people — even few anarchists or libertarians — would object to the socially constructed coercive prohibition against one person hitting another on the head and taking his stuff. In other words, physical ownership unobjectionably warrants a degree of supporting abstract ownership. But what if our social construction protects only blue-eyed people against forcible theft? If you have blue eyes and anyone (blue- or brown-eyed) hits you on the head and takes your stuff, the police will, with substantial efficacy, track down the perpetrator, recover your stuff, and discourage the perpetrator from repeating his objectionable behavior. If, however, you have brown eyes and someone hits you on the head and takes your stuff, the police will do nothing. These conditions seem obviously objectionable, but the objectionable behavior of the police here is itself not coercive; we are not arguing that they should not be coercive at all (we do want police protection against theft), we're arguing they do not act coercively when coercion is warranted: it's clearly the differential in differential coercion that's objectionable. This counter-argument renders invalid the argument that non-coercive behavior is by definition never objectionable. There might be other good arguments that the government should not protect a worker's ownership of his own actual labor, but it's invalid to argue that it is never objectionable for a government — even a minarchist government — to refuse to act coercively.
The effect of differential enforcements of abstract property rights is that under laissez faire capitalism, the ordinary worker has zero surplus to accumulate. Zero multiplied by many years and/or many bodies is zero: ordinary workers simply cannot become capitalists. The "selection pressure" against accumulation is not just short-term and local, it's long-term and global.
But of course the Keynesian "revolution" of the 1940s-1970s overcame this objection: some of workers' ownership of their surplus labor (the difference between their actual labor and the cost of their labor power) was protected by the government. And indeed millions of workers used this surplus to propel themselves and their children into the professional-managerial middle class and in general caused a much wider distribution of capital, actually lowering its price nearer to its cost.
It took the big capitalists almost thirty years to reverse this distribution. Their primary mechanisms were first to "pick off" the smaller accumulations of capital in cyclical downturns. For example, when housing prices fall, the homeowner loses her entire down payment before the bank loses any of its debt: anyone who is forced by circumstances to sell her house in a cyclical low loses all her capital, while the bank loses none.
The second method was outright theft — enabled again by differential enforcement of abstract ownership — of workers' accumulated capital, accumulated mostly in pension plans.
The third method was the creation of consumer debt. Increased consumption was necessary (and desirable) to drive increased production and increased productive efficiency, but there were two primary methods to do so. First, we could have socially created "front loaded" demand: just give individual workers more money, and make that money deflationary to encourage its use in consumption. But instead the capitalist class loaned the workers the money, cheaply at first, until a person's standard of living depended on massive amounts of debt. Once the standard of living is established, the capitalist class (who owns the debt) can raise interest rates to expropriate upper-working and lower-middle class accumulated capital.
Accumulating debt rather than capital is not economically irrational. It is economically and physically reasonable to expect almost endlessly rising income as we continue to make economic production more physically efficient (more goods for fewer hours). Accumulating debt is not even politically irrational: the government was for thirty years willing and able to prevent the capitalist class from manipulating the finance system to use debt to transfer capital; more than a generation of success is reasonable warrant for continued confidence.
Remember: while accumulating debt (to drive the economy in the short term) the working and middle classes were also accumulating capital (to make capital allocation more efficient in the long run). The working and middle classes were acting rationally, both economically and politically, balancing short-term and long-term benefits. While some people (sadly myself included) acted shortsightedly, on the whole we did not fail to accumulate capital, our capital was stolen from us. And it was stolen from us because the capitalist class regained control of the government; they were successful because we underestimated the determination and will of the capitalist class to regain the vast power they controlled before the Great Depression.