I have promised myself to do socially useful work today: working out the distributional effects of QE, trying to plot a more efficient and fair way forward for the policy. So I can, again, only put in a few words on this dreadful nef report (yesterday’s post was written in about 20 minutes. Which is about the length of time nef spent on thinking through their premises).
Chris Giles concedes that the method, in theory, is valid. Just that there are howlers in the execution. I agree. All activities have externalities. There are differences between private and social outcomes. This is very very basic stuff. But working out the true ‘social’ outcomes is not so easy that a nef researcher can do it. To be fair, no one can really do it. This is why progressive taxation makes sense, rather than a socialist all-knowing planner.
The normal, academically honest response to that difficulty would be to say “we don’t know” and make a few rhetorical points. The dishonest way would be to pretend you have found a precise method for gullible journalists (note: not Chris Giles), and then milk it for maximum headlines.
So what are the howlers? Starting with the “banker is worth £X” bit, which got the most publicity.
- Taking the average tax etc contribution of the City, but then subtracting the one off fall in wealth of the entire country following the crash. More accurate would be to take the averages of both over, say, 50 years.
- Using the ‘base level’ as the amount of GDP and wealth that existed before the crash, rather than doing a true counterfactual and asking where we might have been without it for many years. For the education of nef researchers, consider this graph comparing GDP over 18 years. The ‘choice’ that France, Germany made not to do finance well just meant that they climbed a lower peak. They fell, approximately, as far- and are still lower than the UK.
- Assigning the whole of the financial crisis to one entity in the complex systemic mess that created it. Not to overseas, or government action (see The Melangerie). Not to low rates, which had benefits before the crash that are unaccounted for by this knuckleheaded analysis. No: acting like it’s a video game, and the economy is a car, and the crash happened because one sector swerved.
- Ignoring the recessions that have happened in societies without large financial sectors.
- This is very big. Ignoring the way City activity enables other activity, which is not reflected in the turnover of the City. For some reason, this element, which is the core of their analysis of the work of nursery staff, is ignored for the City. But wholesale markets move funds from savers to borrowers. They enable other economic activity to take place. This is not entirely captured in what the City ‘makes’ – otherwise the activities would not take place. So, when a company sees a profitable use for funds, and uses the City to raise them, the returns it makes are mostly not captured by the City. Some are captured by the savers and investors. Some by the business being more profitable. The nef report cannot quantify this: you have to do difficult and openminded work to achieve this. Like this report. Or this.
- As a result, productivity in the UK without the City would be a lot less than just that explicitly assigned to it. It would be whatever an economy without wholesale finance could produce – which would be far far lower.
Moving on to the rest of the ‘workings’ (I will not dignify the simple minded recommendations with further publicity).
- The big mistake is to assign to a particular, replaceable activity all of the consequences that might arise from it. So if a nursery worker takes a kid away from a parent, this should be credited with all the value added by that parent working, according to nef. (What – even if the parent were a banker?) This involves double-counting – because presumeably, some (i.e. most) of their salary ought to be credited to the parents, otherwise you are giving their 9-5 no value. Not the nursery worker.
- No marginal analysis is done: no recognition that if it were not for that nursery worker some other second-best solution would be tried (another woker, or creches, or some other solution.) NO – the nursery worker explains and in some sense therefore deserves a further £78,000 of salary! Given that kids can be farmed out to 18 year olds – nef seem seriously to believe that such 18 year olds can instantly add £78,000 of value to the economy just by watching toddlers. Clearly, there is no interior censor in the mind of nef, going ‘huh?’
- In fact, the whole paper lacks marginal analysis. What if there were no City? Well, we’d use Frankfurt. So no crash? Um . . . . not exactly. What if nursery workers refused to work for less than £30k? Millions of others would step in, from around the world.
- They take the potential value of educating our young kids better – and select a report pricing the social benefits at £30bn annually, and again asign this entirely to the workers. Let us just pretend the £30bn is true for now. But there are all sorts of participants between the kid being better looked after and the social benefit. In fact, what nef are doing in a weird way is assigning ‘property rights’ to nursery workers’ relationship to kids. So, if you look after a child for 3 years pre school, you ought to somehow have reflected in your salary all the benefits that the kid’s education yields for the next XXX years. Sheer idiocy.
- Most of the remaining howlers are of the same nature.
I have no problem with some of their straw men myths that they attack, like the often overstated need to pay high salaries to attract talent. We have managed to have talent in this country for centuries with widely varying salaries. Some of the people threatening to leave because of an extra 10% of tax are going to be no great loss. We have to be utilitarian about this. If it fails to bring in revenue, scrap it.
But other elements are incredibly selective. The private sector is more efficient than the public sector, despite a mostly irrelevant story about the costs of paying hospital cleaners too little. Efficiency can be measured: outputs and inputs. It is not actually surprising: if the public sector destroyed those parts of itself that fell behind, all the time, it too would be more efficient. I am kinda glad it does not (see John Redwood discussion).
I repeat my earlier conclusion. The New Economics Foundation has taken a thin veneer of acceptable method – the theoretical distinction between private and social returns – and applied it to so weak and biased an analysis that it brings disrepute on (a) think tanks (b) economic research (c) the journalists that just repeat the press release without looking into it. In my view, it sets back the task of finding a truly just way out of this fiscal mess.
Let the next 20 press stories about this piece be those debunking it, please. This blog here is a good start.