Wednesday, 12 July 2017

Long Read, Short Research

Reading The Guardian back to front, as one often finds oneself travelling, I came across yesterday's long read. When I saw that it concerned Economics I immediately became sceptical - as I began reading, my fears were confirmed. One paragraph in particular was so dense with contradictions and misunderstandings - all too common - that I thought I’d write just about that.

At the risk of sounding polemic, the paragraph that talks about the Efficient Market Hypothesis (EMH) is, frankly, rubbish. The writer defines the EMH as the following:  ‘since a free market collates all available information to traders, the prices it yields can never be wrong’. I have no idea what the writer interprets a ‘correct’ price to be - within the field of Economics or otherwise - but either way the definition simply isn’t correct. The EMH certainly does not make a value judgement about what the price of an asset ‘should’ or ‘should not’ be. A fairer definition would be this:  ‘Given a financial market has many rational[1] agents, and given they all have access to the complete set of past information about relevant variables, then those who engage in arbitrage in the market should lose money as much as they earn money’. Even more simply put, on average, if the market is 'competitive' with 'full information', it will be impossible to earn profit simply through arbitrage in financial markets. Hopefully this reformulation makes clearer why the EMH result was an important milestone within the field of Economics - Fama was able to formalise, i.e. show analytically, that given a set of crucial assumptions, a result which is fairly intuitive (at least to me) follows.

However, the article then continues to detail Robert Shiller’s contribution to the literature on the issue, and seems to claim that the two economist’s conflicting papers is typical of the field's lack of ‘scientific’ grounding. It's true that Shiller found there to be strong evidence to suggest that the EMH doesn’t always hold in the real world. Famous examples would be Warren Buffet and George Soros, who have managed to earn profits from trade in assets for decades at a time. Shiller’s results bring in to question whether those agents who make up the market, do in fact have access to the complete set of past information, or indeed behave 'rationally'. His results actually demonstrate that, at least some of the time, the assumptions that the EMH rest on do not always hold in the real world.

Now, I’m not particularly well versed in the philosophy of science, but correct me if I’m wrong if the following is not the scientific method at work. First, a hypothesis is generated. Secondly, the hypothesis is tested against empirical data. Finally, the hypothesis about the real state of the world is revised in light of the data. The EMH is exactly that - a hypothesis. Fara and Shiller in conjunction carried out exactly this process, and were jointly awarded a prize for it. It seems obvious to me that Fara did not suggest that no one can ever make any money from financial arbitrage - he is not an idiot. In fact all he did was show analytically that such a market is a zero-sum game. To suggest that economists merely ‘believe what [they] want to believe’ is unfair, and certainly doesn’t follow from the above.

It seems that this misunderstanding has arisen from merely a cursory consideration of the importance of analytical results, grounded on assumptions. It's hard not to smile at the irony here, considering that the piece highlights the importance of economists' assumptions earlier on. Without much consideration, the writer criticises economists for not testing assumptions empirically, and then continues to detail exactly where Shiller won a Nobel Prize for doing just exactly that. Moreover, their misunderstanding about the EMH, it seems to me at least, has come about because they have ignored the assumptions the EMH rests upon. In the definition I gave earlier, the two clauses that follow the word ‘given’ are the EMH's assumptions (this trick is one to watch out for - it’s a common hint that an assumption follows). 

If I remember anything from my first tutorial in Economics, it was to always list, and then criticise, the assumptions of the model in question. Academic economists are obsessed with, to the point of being wearisome, being careful with assumptions (possibly due to the directly conflicting results of analytical models, if assumptions aren’t considered). It seems that rather than follow such an approach, the writer has committed the unforgivable sin of ignoring the assumptions Fama listed. 

For full disclosure, I stopped reading the article after this paragraph. But even then, this wasn't the only paragraph in the article where I took issue with something said. But this one did seem to be particularly funny, full of contradiction and peppered with superficial research. Another problem is the way the writer deals with 2008. In one sentence, they are happy to admit there are differing schools of thought about what the response of governments should have been, but then in a later sentence condemns the entire field as having failed - despite any internal contradictions, I think the issue is slightly more nuanced than that. I'm sure The Guardian means well when it publishes it's quarterly 'bash the economists' piece, and there are certainly issues within Economics that need to be addressed - but poorly researched articles do not a field reform. Criticising Economics should really be some of the lowest hanging fruit for a broadsheet in 2017, so it's surprising it can't be done in a more convincing manner. 

[1] Within Economics, a rational agent is usually considered to be one that consistently acts in accordance with their own preferences. Agents that make up financial markets can be fairly assumed to be maximising profit - so in this context, a rational agent is one that maximises profit.


  1. I did not read the article but from my memory I believe that the EMF comes in three varieties depending on what you mean by "information". Buffet and Soros seem to have the gift of acquiring and interpreting information differently from the mass of people. That is where their skill lies.

  2. Thanks Jeffrey. Yes, it does. The weak form states all traded assets reflect publicly available information. The semi-strong form builds on this, and says that prices will adjust immediately if new information becomes available. The strong form states that prices will reflect not just publicly available information, but also privately held information too.

    Does Soros' ability to interpret publicly available macro- information differently to other traders mean he is actually privy to *new* information than the rest of the market? I think it might! At least, that seems consistent with the EMH to me.


40 Favourite Albums of 2017 so far... (21-40)

This is part 2 of my favourite albums of the year so far. Go and read part one first if you haven't. Although saying that, they're ...