Saturday, 23 October 2010

Multipliers, markets and why the New Economics Foundation make me cross

The frighteningly bright Jon Beech has challenged my frothing rant about the New Economics Foundation and indeed my contention that much of their economics work is “evidence-free”. In doing this Jon provides what I find a very helpful positioning of NEF:

NEF's primary agenda is to counter the consumerist agenda that's piped out through radio TV etc, and to help people to reflect on what their relative wealth actually buys them: neuroses and over-leveraged anxiety. They question the economies of scale which result in uninterested interest.

Now were NEF simply engaged on a moral crusade then that would be a fair position but they are clearly positioned as researchers engaged in a scientific (in so far as we can call economics science) process. Which – as I said in my rant - requires evidence. And science, being the troublemaker that it is, evidence means empirical study. It means the raising and subsequent testing of hypotheses. And this NEF does not appear to do. It isn’t sufficient (and this applies as much to those who prefer a libertarian analysis to a statist analysis) simply to select the statistics you like, find a few quotes and cite your own earlier work. Unless of course you're a simple blogger like me!

Let us take just one of NEF’s “products” – LM3:

LM3 was developed by NEF (the New Economics Foundation) as a simple and understandable way of measuring local economic impact. It is designed to help people to think about local money flows and how their organisation can practically improve its local economic impact, as well as influence the public sector to consider the impact of its procurement decisions. It was designed to be quick and relatively easy, and to highlight where an organisation can improve its impact.

This is presented as an econometric tool (indeed I used it in my MSc dissertation) for assessing local economic impact but it is very difficult to track back to the theory underlying the model or indeed to find whether NEF actually undertook robust, peer-reviewed tests of its validity before launching it (with the accompanying consultancy offer) onto a credulous public. And where it has been tested this is what we find:

Notwithstanding this conclusion, difficulties in data collection combined with inaccuracies inherent to the LM3 process created a large margin of error in the findings.

And this is before we have considered the theoretical critique of the Keynesian multiplier. So apologies for getting all academic for a minute:

The multiplier is a central concept in economics and especially regional studies where it is widely used to assess the long term impact on employment and output from different forms of investment. As such it represents a significant part of the Keynesian aggregate demand model of the economy and can be described as the impact of the marginal propensity to consume (mpc) on a given investment or expenditure where the higher the level of mpc the bigger the multiplier (Heertje & Robinson, 1979).

The problem is that measures of the multiplier do not take account of the input source (i.e. where the money is earned) and that it is very difficult to define what we mean by “local” or “local economy”. This is significant since by applying the LM3 model to public expenditure we do not take account of the opportunity costs associated with that public spending (i.e. what the taxpayer would have spent it on had it not gone in tax). With the result that:

It is quite misleading to leave public policymakers with the notion that their spending is not at the expense of the private sector because it may be autonomous or have multiplier effects

None of this intended to say that Keynes was wrong about the multiplier (although there is plenty of doubt about his arguments in this respect) but to point out that NEF appear not to have been rigorous in their appraisal of LM3. Or rather that I cannot find any peer-reviewed, critical appraisal of their model – there is nothing evident from their website and my (limited) searches find very little published work applying the LM3 model in an academic research setting. Even in NEF’s own published work in my area of specialist interest – street markets – there is only an assessment of inputs (how much consumers spend) and no evidence of how that money then flows (e.g. derived from sales figures, margins and staffing costs of market traders). I happen to agree with NEF’s conclusions about the negative job impact of supermarkets but then so does almost every piece of independent economic research which doesn’t excuse NEF from skimping on the evidence.

This is just one of NEF’s products – I cannot comment on other products or reports – but to me it suggests that Jon is right in suggesting that this think tank sees its role as challenging the moral basis for consumerism and perpetual growth rather than providing a substantial and robust evidence base for us to understand how an alternative might work.


No comments: