[OPE-L] "Crowding-out effect" (coda)

From: Jurriaan Bendien (adsl675281@TISCALI.NL)
Date: Sun Oct 21 2007 - 12:01:33 EDT


Another thing I forgot to mention, maybe worth noting anyway. As regards the "crowding out effect": Laurence Harris (of Fine & Harris fame, on the left) does in fact discuss this in his classic book Monetary Theory (1981), an excellent introduction to the lengthy debate between monetarist and Keynesian views of money. However, this book is not a "Marxist" book, simply a very solid textbook on the origin and development of modern monetary theories. 

"The idea that fiscal policy is impotent because it crowds out private expenditure and merely diverts resources to the public sector (supposedly "distorting" the allocation of resources) has an ancient lineage. It was a direct corollary of the pre-Keynesian Quantity Theorists' emphasis on Say's Law as a long-run proposition. For, since this assumes that supply creates its own demand, the economy is assumed to be always at full employment. If resources are always fully employed, government expenditure on goods can be met only by diverting resources from goods produced for private demand; private sector expenditure is merely crowded out and government deficits leave real national income unchanged at the full-employment level. 

The modern Quantity Theorists' crowding-out theory is not, however, restricted to full employment equilibria. (...) One possibility is that government deficits themselves have a depressing effect on private sector expectations and thereby cause a reduction in private investment plans at all rates of interest. (...) Or it may be that a similar dislike of deficits reduces "confidence" and thereby increases the (precautionary) demand for money. (...) The argument in these cases, however is not very convincing, for, if crowding out is to occur in the sense of the multiplier being zero, one must make the arbitrary assumption that the expectation and confidence effects exactly offset the initial increase in deficit. 

An alternative... argument is based on the assumption that the private sector is "ultrarational", meaning that individuals view the government as merely an extension of themselves. Thus if the government increases its investment expenditure and finances it by issuing bonds, the future consumption yielded by the investment reduces by the same amount the demand for future consumption yielded by private investment. Private investment, accordingly, falls by exactly the amount that government investment has increased. The ultimate effect, therefore, is... that real income is unchanged by the government deficit. Similarly, an increase in government expenditure on consumption goods and services would, if individuals are ultrarational, be exactly offset by a decline in private consumption (even if the expenditure is financed by taxes rather than bond issues). The assumption of ultrarationality is a very restrictive one and its realism may be doubted. (...) 

The theoretical models of crowding out are...simple models whose conclusions follow simply from particular, restrictive assumptions. In general, however, the Keynesian reply has not concentrated on criticizing these assumptions. Instead, [the Keynesians]... argue that the ultimate effects considered so far are not in fact long-run equilibrium solutions... the concept [of crowding out] requires that asset stocks are unchanging. If asset stocks are changing and private expenditure and demand for money are subject to wealth effects, then the IS and LM curves will be shifting from period to period. In such a case, no period's short-run equilibrium would be a long-run equilibrium, for it would not be sustainable from period to period." (p. 438-440). 

As I mentioned before, and as Laurence Harris illustrates in a much longer story, a "crowding out effect" is in fact very difficult to prove empirically, and the equilibrium models are all contrived with counterfactual assumptions (in that case, you're much better off to study the empirical trends, if you want to understand what really happens) . Therefore the notion of a "crowding out effect" is really more in the nature of an ideology that swings elusively between the real and the ideal. 

In Marx's own theory, this ideology - rooted in the assumption of a zero-sum game between saving/investment and consumption, and a notion of equilibrium as a balance between them - does have a rational kernel however. It ultimately reflects a competition over the mass of newly produced surplus-value that can be distributed to owners of capital as income, where employers in the private sector argue that any government interference in markets will result in a misallocation of resources that does not benefit them, and consequently does not benefit economic growth. 

In particular cases, that could be true, or it could be false. Obviously, all kinds of arguments can be made in this regard, there are oodles of arguments about the efficacy of the public sector, and the "crowding out" hypothesis is only one of them. For a recent example, hard-core free marketeer Alan Greenspan argues against collaboration between the state and the private sector to create a $75billion-plus investment fund to buy the assets of troubled investment vehicles, because it would prevent market forces from establishing "true clearing prices for asset-backed securities". http://www.ft.com/cms/s/0/0f79b248-7e5c-11dc-8fac-0000779fd2ac.html?nclick_check=1 His critics might argue the exact opposite, i.e. the justification for such a collaboration is precisely that the true clearing prices are not established, for example through lack of "transparency". It is a debate that can never be resolved, because it is based on axioms which you either accept or reject.

In the real world, the controversy is obviously much more complex and qualified, given the reality that the state both "giveth and taketh away" - the private sector is also the beneficiary of government contracts, cheap loans, subsidies and hand-outs. This is no small matter, since, as Martin Wolf noted, extreme neoliberal shock-therapy to cut off all this government interventionism at the root had the result, in New Zealand (admittedly as he says "a small country", i.e. a peripheral player in the world economy), of a "net stock of liabilities [=debts] of more than 100 per cent of GDP" (he fails to mention also, that the country is transformed into a low-wage country with malfunctioning basic public services, trailing far behind Australia). This being the case, it is easy to see why an important section of the propertied class would not take "crowding out effects" very seriously, and recognise that without state interventionism, matters would be worse. That is, unregulated competition does not result in a situation approaching an equilibrium of any description. Consequently, the principle of state interventionism may be regarded as a "lesser evil". 

In my own view, as I discussed before, a basic problem is that in the "financialised, digitalised" economy of already industrialised countries, nowadays only weak incentives exist allround to expand material output on a scale that could recapture the cumulative economic growth pattern of the 1950s and 1960s, and in that case, it does not make all that much difference anymore what particular financial incentives are tried. Instead, much more significantly, there is an increasingly furious competition for already existing asset-wealth, and the access to that wealth. As I have illustrated before, there exists a lot of already existing asset-wealth (the results of previous rounds of accumulation), and should it not exist here, you can import it from there (from newly industrialising countries).

This is not immediately obvious from the GDP data, for reasons I have indicated at various times, but if we carefully unpack the kinds of activities included, we find that an increasing proportion of them essentially consist of a trade of "services" that involves the shifting of resources (or the property rights to those resources) from A to B, rather than making any net additions to those resources that could benefit those who simply don't have them. 

Robert Brenner pointed out once that a large portion of  the "economic growth" in the 1990s consisted of building more office space, and furnishing it. But as the competition for already existing asset-wealth has become more efficient, a lot of that office space is no longer even used. Here in Amsterdam, in 2006, about 18.5% of all office space was empty and unused (in some cases, squatters manage to occupy it, for a place to live), a matter of concern to the city fathers who would like to reinvent the city as a "dynamic metropolis" (they can hardly do that however, without much better incentives and supports for new business starts which are not just tax shelters, i.e. a multi-pronged economic strategy based on the realities of life here). 

The basic problem with the economics profession, I've often thought, is that its response to all this kind of thing is, to invent a never-ending stream of new "models", based at least in part on counterfactual assumptions, rather than to study the empirical trends in the real world comprehensively, to understand what is actually happening. Few people also study economic history anymore. 

When it turns out that the theoretical model does not apply to reality, unless you drop crucial assumptions of the model, all they do is build another model, which however turns out to suffer from the same deficiencies. That is just to say, there is something drastically wrong with the whole way in which the theories are formed in the first instance (both with regard to their internal logic, the methodology, and the social conditions within which they are formed). 

Tragically, Marxist economists often mimick this whole process, to bolster the theoretical coherence of their favourite axioms, wtih the same result. One is however left with a sense of futility, because the modelling explains little anymore, and it is often no longer clear what the problem really is. After considerable bickering and a long detour of theoretical ostentation, one arrives only at the conclusion that "there is much we do not know". Perversely, we have in this way created our own ignorance, in which case, people fall back on cherished dogmas. But that leaves little room anymore for rational discourse between opposed viewpoints, all you have is that one dogma is pitted against another, and the same things are endlessly reiterated.

In truth, if you really want to know at what "level of abstraction" a theorem might be valid, the only way to find this out, is to study the facts of experience. And so long as this is not done, the debate remains interminable, since for every hypothesis mooted, an alternative hypothesis can also be stated which, other things remaining equal, is just as reasonable. It is a rootless, meandering discourse of abstractions that leads nowhere, precisely because it is conducted in remote abstraction from the very real objects to which it is supposed to refer. Hence the frustrations which you have repeatedly voiced on OPE-L. (I have to get on with other things now)


This archive was generated by hypermail 2.1.5 : Wed Oct 31 2007 - 00:00:19 EDT