Archive for November, 2012
Milton Friedman is quite a revered figure – among economists, conservatives, libertarians and some leftists – partly, of course, because he was a prolific economist, but also in large part due to his debating skills. He is generally perceived as able to shut down arguments from the left with simple, easy to understand, often amusing one liners. However, I have always found him unconvincing, and here I hope to show why.
There will be, of course, numerous conceptual disagreements which I will try not to discuss in this post: the phony market-government dichotomy, where government is some exogenous entity; the general idea that self interest will lead to the best of all outcomes; the invocation of the mythical ‘free market.’ Nor is the purpose of this post to draw attention Friedman’s major intellectual arguments themselves – though I have already done that with his stance on assumptions, corporate social responsibility, and I suppose by proxy I have commented on his interpretation of the Great Depression.
Instead what I want to do here is show Friedman’s general debating techniques are highly questionable. Many of his arguments rest on an abuse of the reductio ad absurdum. Sometimes Friedman was either ignorant about the evidence or just plain dishonest. Many of the ‘facts’ he cites don’t stand up to even a brief fact check. Here are some examples:
Here, Friedman references the late 19th century land deals as ‘minor.’ One billion acres of land is not minor. Much of this land was taken from citizens directly in the interests of privately owned corporations, and/or involved widespread fraud and corruption. The sheer volume of land seized suggests the deals played a massive part in establishing the railway lines, along with many other industries across the U.S. Would these engines of growth have been built if not for the coercive grabbing of masses of land?
The fact is that even cursory glance at social mobility in the United States puts lie to Friedman’s claims about it being high. Similar results hold for most developed countries, generally only changing as they become more Social Democratic.
In general I’ve noticed Friedman makes repeated vague references to ‘all of history,’* for which he never provides specifics, and which are actually completely at odds with the evidence. I can only conclude that what he says is based not on history but on armchair analysis of what must have happened, based on his own logic. But there is strong reason to doubt this logic – as I have discussed in my previous posts A Brief Anti-Economist History and How Natural is Capitalism, Exactly?, Western Capitalism did not just spring out of nowhere due to the magic of the market.
For example, was Friedman aware of hunting restrictions such as the Black Acts, which brutally enforced limits on peasant activity and so contributed to the initial rise in the industrial workforce? The rise in enclosure acts, which did something similar? Is he aware of the large amount of U.S. tariffs during the country’s rise to prominence, and similar trends in other Western countries, as well as more recently developed countries in Asia? Such historical debates are lost in a sea of sweeping assertions about the efficacy of the apparently omnipresent ‘free enterprise system,’ with narratives that would have Friedman fail a first year history essay.
To be sure, as with all historical analysis, there is always room for discussion, but Friedman’s arguments rest upon one interpretation of ‘facts’ that are usually incomplete or a blatant misrepresentation of what actually happened.
So it’s quite easy to find instances of Friedman presenting questionable evidence to support his arguments. But what about his famous purely logical put downs? Do they stand up to scrutiny? Here is quite a widely watched YouTube clip (always worth noting that the young man is not actually Michael Moore):
The issue raised is whether Ford acted immorally by failing to install some safeguarding blocks in their infamous Pinto cars, which resulted in a large amount of deaths. Friedman suggests the problem is amoral, and simply a matter of price: nobody can place an infinite value on a human life, so whether the car should have been released rests on a monetary trade off. Friedman suggests the young man who posed the question is not interested in principle, only price. In fact this is not true; Friedman simply asserts it and builds his argument from there.
A moment’s thought will suggest that the important principle is not price, as Friedman suggested, but human life, as the young man seemed to think. Consider: if the automobile will kill 10,000 people a year unless the defect is fixed, does it really matter how much it will cost to fix the problem? Surely, if fixing the defect in the automobile is affordable, then the defect should be fixed. On the other hand, if fixing the defect in the automobile is not affordable, then defective car should not be released. However, a profit driven firm necessarily insists that lowering costs by not fixing the defect outweighs saving a certain number of lives each year. Friedman abuses the reductio ad absurdum by taking the issue – where it was clear Ford simply should have installed the boxes – out of context, and focusing on price as the important variable.
Here is another example where Friedman abuses the reductio ad absurdum:
What the man posing the question to Friedman is actually alluding to – in a roundabout sort of way – is a simple concept called the ‘income effect:’ reducing somebody’s income via taxation may well increase the amount they work (and empirical studies suggest it does), hence increasing overall production. Of course, if you increase it to 98% you will impoverish people and destroy the economy, but nobody actually suggested that.
Perhaps some might interpret this as cherry picking. So, finally, here is a full interview with Milton Friedman. I will discuss Friedman’s remarks throughout the interview:
(1) At 2:06, the presenter asks if a place such as Central Park would exist in a ‘pure market’ situation. Friedman fails to answer the question directly, but during his response he blames Central Park’s problems during the 1980s on public management. But Central Park is in fact a public private partnership, where the private firm employs 4 out of 5 of people maintaining the park. Again, Friedman either has no idea what he is talking about or is lying.
(2) At 4:10, the presenter brings up Thalidomide. Friedman’s response contains two problems. First, when he says that the FDA stalls potentially beneficial drugs from being used, he fails to distinguish between a Type 1 error – falsely rejecting, say, a perfectly safe drug – and a Type 2 error – failing to reject an unsafe one. Type 1 is generally considered worse on the logic of ‘convicting an innocent person.’
Second, Friedman suggests that the company responsible for Thalidomide did not make a profit, therefore the market would ‘signal’ for it to go bankrupt. I am not sure whether they made a profit in that particular instance. What I am sure of, however, is that the company is still around today. Again, Friedman references facts that are questionable on even a cursory inspection.
Lastly, when Friedman suggests that the airlines will make sure all of its planes are safe, he neglects the ‘weighing up logic’ we saw in the Ford Pinto video. (I am playing Devil’s Advocate, as I don’t think any sane person would defend current U.S. airport security.)
(3) From the beginning the interviewer asks Friedman about the government setting information requirements on packaging. I don’t really understand how Friedman can make the ‘if it mattered a profit seeking firm would take advantage of it’ argument when the interviewer has explicitly stated that the government had to start to enforce information requirements because private firms were not doing it.
Note also that he doesn’t actually engage with the civil rights act question explicitly.
(4) During the final video the interviewer takes Friedman through every government program and Friedman advocates abolishing the majority of them. What confuses me about this part is that Friedman advocated some of these programs elsewhere – for example in his books. In Capitalism and Freedom, he advocated building infrastructure, a negative income tax, school vouchers, praised antitrust laws and more. This highlights Friedman’s dual roles as a propagandist and serious thinker, as a man who was willing to make sensationalist claims and advocate radical policies just to get attention, even if he didn’t truly believe in them. (Some may suggest Friedman was older and had matured here, but there are examples of him criticising these things he advocated elsewhere when he was younger, too. I also see no justification anywhere, ever, for his surreal last minute ‘abolish the federal reserve‘ position).
I don’t mean to suggest what I have to say is the final word; I merely hope to point out that Friedman was somewhat disingenuous and often used logical sleights of hand to get his point across. His interviewers and opponents rarely seemed to press him on it, and to be honest I never saw him go up against anyone particularly formidable. Furthermore, Friedman’s case highlights how little weight should be placed on verbal debates: one liners that seem persuasive at first can evaporate under close scrutiny; facts can be presented with few checks and balances; questions can be dodged and twisted. Friedman was prepared to argue the more ‘free market’ position merely for the sake of it, and was undoubtedly skilled at this role. But once you unpick some of the arguments and cross-check the evidence, his world view leaves a lot to be desired.
The final chapter Steve Keen’s Debunking Economics is a brief overview of the major competing alternative economic schools of thought. The question posed is whether these schools present a viable alternative to neoclassicism and marxism, both of which Keen has already dismissed. He now goes on to evaluate Austrian, Sraffian, post-Keynesian and evolutionary economics, as well as Econophysics. I will look at Keen’s evaluation of these schools of thought and discuss his conclusions.
Keen’s view on the Austrian school is similar to that of myself and other post-Keynesians: it shares many characteristics with neoclassicism. These include but are not limited to: an exogenous money supply (ex. Lachmann and Schumpeter), Say’s Law, a variant of marginal productivity theory, reductionism and a government versus markets perspective. Hence, many of his earlier critiques – Sraffa’s work on capital, the excessive focus microeconomics, and post-Keynesian views on banking and the money supply – could equally be applied to Austrians. Keen himself thinks that Austrians deal with uncertainty, but (again, ex. Schumpeter and Lachmann) I’m not even sure this is true – for example, Hayek completely misused the term. Hence, criticisms of neoclassical models based on irreducible uncertainty may also apply to some Austrian arguments.
While Keen applauds Austrians’ analysis of capitalism as more dynamic than that of neoclassical economics, he notes that they do seem to retain the belief that capitalism has a ‘natural’ state that should not be ‘interfered with,’ and they actually seem to take it much further than their neoclassical counterparts. This is particularly apparent – also something that I have noted – with Hayek’s ‘spontaneous order.’ Though it is an interesting concept, it has been misused as an ideological tool against government, without considering the ‘spontaneous order’ that may evolve inside government, or the possibility the dichotomy between governments and markets may be a false one.
All in all, it’s hard to deny Austrians are part of the marginalist tradition, something Mises explicitly said. Hence, I don’t consider the school a truly ‘alternative’ way of thinking about economics, even if it has something to offer.*
Keen praises Sraffa’s work as “the most detailed and careful analysis of the mechanics of production in the history of economics,” and notes the importance of the interesting conclusions that it brought to light. Nevertheless, Sraffa’s analysis is a static one that seems to be dependent on the existence of a long run equilibrium (here Keen quotes the Sraffian Ian Steedman as evidence). Due to the lack of dynamism in Sraffian models, Keen’s previous comments about dynamics and equilibria could be applied to Sraffa. Keen ends by noting the subtitle of Sraffa’s magnum opus: “Prelude to a critique of economic theory.” He suggests Sraffa’s main aim was to provide a basis with which to critique other theories, rather than present a positive alternative. I’ve no doubt Sraffian readers will disagree.
This school of thought is characterised by the application of modern chaotic modeling techniques to economics. Hence, the models produced are far better suited to generating the kind of instability we observe in capitalist economies than are those used in neoclassical economics. Keen comments that the school isn’t really a direct critique or challenge of neoclassical economics, instead dismissing it outright and presenting an alternative.
Bearing the lack of direct engagement with economists in mind, it’s not surprising that the physical scientists suffer somewhat from a curse of being a mirror image of economists. Keen says that they have been rediscovering old insights such as IS-LM, then using them with other, incompatible models such as rational expectations. They also seem to have an ‘everything looks like a nail when you have a hammer’ problem, and are applying inappropriate laws, such as conservation to the distribution of wealth, or electromagnetism to immigration.
Perhaps econophysicists should be more willing to read through the history of thought – as I noted in my post on mathematics, this type of imperialism/arrogance in physicists is no prettier than in economists (commenter Blue Aurora told me that some econophysicists have been more willing to engage with the discipline recently, which is a good development). Despite these flaws, the tools of modern chaotic modeling are surely a promising area for the future of economics.
Keen’s discussion of this field is the first time I have been properly introduced to it, so I’ll be brief. Keen seems to think that evolutionary science is an appropriate and promising field, but one that lacks maturity. Many evolutionary concepts, such as adaption and survival of the fittest, are surely applicable to capitalist firms and product evolution. Having said that, economics lacks the equivalent of the gene to ground the evolutionary approach, so many evolutionary models are often forced to rely on analogy. Perhaps – and hopefully – the evolutionary school will be able to establish a coherent grounding in the future, but for now it is not a strong enough alternative to neoclassicism.
As this is the school Keen and I both most closely align with, you’ll not be surprised to hear the many advantages we think it has to offer: dealing with uncertainty; the relative lack of ideological commitment to any particular system; paying sufficient attention to money, debt and banking; more reality based models of the firm; freedom from reductionist constraints, and much more.
The main problem with this school is the lack of coherency. It’s almost defined as ‘not neoclassical economics’ (and, Keen might add, not Marxism either). Post-Keynesiansism does not really have an agreed upon methodology, something that has worked against its status as a fully fleshed out alternative.
As a brief aside: personally I don’t see why class shouldn’t be adopted as the ‘official’ methodology of post-Keynesians. It is compatible with many of the core tenets of the school – for example, the idea that individual actions should be understood in their class context fits in with the post-Keynesian idea that microeconomics should have ‘macrofoundations.‘ Furthermore, there is also an element of ‘reclaiming classical economics’ to post-Keynesianism, and the classicals generally used class as a methodological starting point. Finally, many of the models – including Keen’s – already use classes as agents, so it seems like a natural progression.
Overall, it seems post-Keynesianism is simply less rigid and more reality-based than its neoclassical counterpart, and is more fleshed out than other alternatives, save a problem with a unified methodology.** Although I suggested that this methodology should be class, perhaps – and this something to which Keen alludes – the lack of a rigid methodology is a strength rather than a weakness. Viewed from this angle, post-Keynesian economics can accept and develop concepts from all of the alternative fields (as well as institutional economics, which Keen doesn’t mention). This also solves the ‘divide and conquer’ problem – part of the reason for neoclassicism’s dominance seems to be the splits between its rivals, which as you can see are many. Generally, I think cooperation between the alternative schools of thought may be the key to building a robust alternative to the curiously resilient school of neoclassicism.
*Obviously there are strong divides within the Austrian school. Rothbardianism is barely worth exploring, while Hayek and Mises have some insights but were fairly tainted by the government-market dichotomy. As I have noted above, Schumpeter and Lachmann seemed the most willing to abandon certain pretenses and come to interesting conclusions.
**Actually, judging from the constructive comments on my marxist economics post, I have more faith in marxist economics than does Keen. However, I will need to explore it more fully before I can come to a definitive conclusion.
Yes, yes, I know I’m far from the first person to use the pun in the title.
Chapter 17 of Steve Keen’s Debunking Economics is a rejection of the Marxist Labour Theory of Value (LTV), and with it the most generally accepted analytical form of Marxism. However, Keen does not reject Marx’s ideas outright, instead suggesting and praising an alternative interpretation: one shorn of the LTV, the tendency for the rate of profit to fall, and hence the inevitably of socialism.
Note that this is my first formal introduction to the LTV, so I can’t claim to know the subject in much depth.
The LTV suggests that labour is the only true source of value, as it is the only factor of production that can ‘add’ more than its cost. This can be demonstrated by the simple observation that workers produce more than workers receive in wages. Marx called what workers produced ‘labour power’ and what workers were paid ‘necessary labour time.’ The difference labour power and necessary labour time is the surplus, and the ratio of the surplus to the necessary labour time is the Surplus Value (SV). The rate of profit, on the other hand, was the surplus over the necessary labour plus other inputs (capital).
Because a similar distinction between ‘commodity power’ and ‘commodity’ could not be made for anything else, capital could not produce more than the value that went into it, but labour could. This meant that a higher ratio of machinery to labour would mean less SV for capitalists. Marx argued that over time, capitalists would replace labour with machinery (something they obviously like to do), so SV – and with it the rate of profit – would decline. This would lead to an attempt by capitalists to push down wages and eventually a socialist revolution.
Marx ran into some theoretical problems with this story. The most famous is the Transformation Problem. This arises because capitalists do not care about the rate of SV, but the rate of profit. Marx had already assumed that the SV was constant across industries. Following this logic, a more labour intensive industry would have a higher rate of profit than a more capital-intensive industry, and capitalists would continually move from more capital-intensive to more labour intensive industries in search of higher profits. This complicates the story behind the tendency for the rate of profit to fall.
Marx tried to solve this by arguing that capitalists do not secure only the SV accrued from their own industry, but that they are effectively stockholders in a joint enterprise that comprises the entire economy. Hence, SV and the rate of profit could both be constant between industries. He provided a numerical example to demonstrate that this was feasible: tables showing the various rates of profit, production and surplus, with the rates of profit and surplus uniform between industries. Marx’s example was mathematically correct – in that everything added up – but really it was nothing more than a snapshot of a particular point in time that may or may not have been reality.
At this point Keen channels Ian Steedman’s critique of Marx, which builds on Sraffa’s analysis in Commodities. Steedman starts with a Sraffian economy in which the various industries have to produce enough for the total inputs in the next period (i.e. enough to ‘reproduce’ the entire economy). He tries to convert the inputs and outputs into Marxian ‘values’ based on labour power and SV. From this, he derives output values and converts them into prices. However, he then runs into problems: what starts as an equilibrium destablises and rates of profit diverge, sometimes increasing.
So what happened? Steedman simply concluded that the entire idea of going values to prices was bunk – in his hypothetical economy, it was possible to calculate prices independently of any ‘theory of value,’ as did Sraffa. Sraffians believe that the ‘transformation problem’ is nonsensical and production should not be analysed from any perspective of utility or value, but from physical quantities and reproduction of industry. Note that this doesn’t necessarily imply that capital doesn’t exploit labour somehow; more so that Marx took a wrong turn in justifying this idea.
So it is hard to tell a consistent story that builds from labour value and ends up with a falling rate of profit and a uniform, economy-wide SV. Marx attempted to justify it with a special case snapshot, but Steedman showed there was no reason to expect the economy to be in or remain in this state, and no need to invoke ‘value’ in the analysis at all.
Furthermore, there is another significant problem with Marx’s theory of value in and of itself, one that he seemed to acknowledge elsewhere. The very premise that labour is the only source of value can be subjected to an incredibly simple, powerful critique.
Classical economists, including Marx, used to distinguish between two features of a commodity: the ‘exchange value‘ – what it sold for on the market – and the ‘use value‘ – how much it is worth to the buyer. Clearly, though, if this is true of commodities, then one can have a higher use value than exchange value, and hence can be a source of SV for a capitalist. This is a neat observation that can make Marxism a highly appealing analytical framework with which to analyse capitalism, one with the modification that socialism is not inevitable (even if it may be desirable on other grounds).
So, the LTV is quite hard to defend: Marx had to make some arbitrary assumptions that don’t seem to hold; his supposed equilibrium in which the rates of SV and profit would be constant turned out to be unstable; his premise contradicted his own distinction between use value and exchange value. Having said all this, Keen thinks that Marxism is stronger once it is rid of the LTV, and that Marx’s broader analysis of commodities and production is still a highly illuminating framework with which to analyse capitalism.
I have never thought of the macroeconomic production function as a rigorously justifiable concept. … It is either an illuminating parable, or else a mere device for handling data, to be used so long as it gives good empirical results, and to be abandoned as soon as it doesn’t, or as soon as something else better comes along.
– Robert Solow
When speaking about production and output, economists generally refer to ‘factors of production;’ things are inputted into the production process to produce something else. Most of the time, they use the two factors ‘capital’ and ‘labour.’ They are a firm’s presumed inputs in theories of the firm and supply curves, where a firm takes their values as inputs and, after some mathematical manipulation, produces a certain amount of output. They are also used in a macroeconomic model known as a ‘production function,’ which does something similar for the entire economy. There are various different production functions that use different maths, and include other variables such as technology or productivity – the most famous one is known as Cobb-Douglas.
The problem with this form of estimation is that it has long been known to be logically questionable. Anyone who has taken a science class past a basic level will know that checking your units – that they are consistent and balance out on both sides of the equation – is emphasised repeatedly. But this seems to be thrown out of the window in the basic analysis of production functions and firm behaviour.
The analysis of production takes two physical inputs – most likely capital and labour. Generally, the inputs are also assumed to be clay-like; available in infinitely small quantities. The inputs are combined (as far as I can see, this means flung together inside a black box) and produce a physical output of some other good, which is of course also infinitely divisible and clay-like. Labour is measured in terms of hours of work; capital in terms of money. This is where the problems start.
The Cambridge Capital Controversies revealed many problems with using a monetary value to measure capital equipment, certainly within a theory of distribution. However, there is another, far more simple and perhaps more fundamental objection: by definition, we are supposed to be measuring physical units of input. This means it is simply not coherent to measure in terms of cost. If we were to opt for measuring in terms of cost as a rule, then what would be the justification for not lumping labour in with capital, and just having a single input, perhaps labelled ‘stuff’? The answer is the justification for not doing the same with capital.
If we decide to use physical inputs, it seems there are ways around the problem. Instead of labelling one input ‘capital,’ we could consider a certain type of capital good – say, shovels with which to equip some ditch-digging labourers. It is fair to assume these are roughly the same and so we can add them up. However, this method lays bare problems that the blanket term ‘capital’ previously obscured.
First, we clearly need more than just people and shovels to dig a ditch. We might need wheelbarrows, land, a skip, sustenance for the labourers, transport for labourers, perhaps a supervisor – in fact, there is potentially an incredibly large amount of factors of production, something I’ve noted before. It becomes computationally difficult or even impossible to include everything that contributes to production, and some factors will simply be immeasurable.
Second, it is clear that these objects are not perfectly divisible. In the examples of ‘capital’ and ‘labour,’ we could divide both money and labour time into infinitely small units. But once we allow for production being ‘lumpy,’ functions are no longer smooth and differentiable, and as such marginal productivities simply do not make sense.* Furthermore, this belies the idea of an elasticity of substitution – the rate at which you can substitute one input for the other – since taking away a ‘lump’ will simply make output fall to zero (this is also something I’ve touched on before).
Economists will likely have various rebuttals to this style of thinking. The most used will be that Cobb-Douglas and various theories of the firm make good, testable predictions. But actually their predictions leave a lot to be desired – firms do not behave how economists predict, and the Cobb-Douglas production function has poor empirical results (economists generally refer to the initial estimations made by the creators of the model, but things have changed since then).
The other defense will be similar but not quite the same: it is just a simplification, used to illuminate a particular aspect of a problem. Well, the fact is that making counterfactual assumptions about the nature of a system does not illuminate anything; it simply tells us about a different universe. Furthermore, simplifications cannot be internally consistent. Even within the logic of ‘labour’ and ‘capital,’ it has been shown repeatedly that the conditions under which either of them can be aggregated are incredibly stringent. Similar arguments apply to other aggregate parameters used by economists, such as aggregate measures of technology or productivity.
Simple macroeconomic production functions smack of trying to turn macro into ‘applied microeconomics.‘ But it has repeatedly been shown that aggregation problems will always be present, and that it is best to study emergent phenomena rather than try extrapolate microeconomic parameters until they have no real meaning. At the other end, microeconomic production is just an attempt to reduce everything to ‘rigorously’ derived smoothly differentiable intersecting lines, rather than simply accepting empirical realities about firms and micro behaviour, and opening up the firm to see what happens inside instead of treating it as a black box.
Overall, it seems the whole idea of production functions and factors of production as anything other than vague, qualitative concepts is something of a dead end.
*I similarly expect that, once we allow that preferences may be lumpy, utility functions are no longer smooth. But lumpy preferences is something for another time.