Chris Giles does us a favour by highlighting a very interesting speech by external MPC member Ian McCafferty on the failure of British productivity to recover as it has in  previous expansions.  But I am puzzled at what a strong conclusion Chris takes from this.

The speech is mostly good.  It breaks down this productivity failure to a sectoral level, and uses masses of survey evidence from the Bank’s Agents to work out how much of this is really cyclical – people complaining about “tough trading environment” which alongside the need to maintain a certain minimum level of staffing level will manifest itself as weak productivity.  McCafferty concludes that around 40% of the weakness in these sectors has been cyclical, and the other 60% comes from real factors: the run down of the oil fields being one, the permanent hit to financial sector revenues another.

(You can take issue with some – housebuilders have long had to deal with increased environmental regulation, for example.)

By my reckoning, McCafferty isolates a 5 percent shortfall in GDP, and is saying that 3 percent is definitely never coming back owing to identified structural factor.  Fine: I think this has been in the calculations of the productivity-puzzlers for a while.  As far as I can see, this is what McCafferty has done:


That is useful. But it is a small amount.  3 percentage points of an estimated 16-17 percentage point fall off in cumulative productivity has been thrown into the “permanent” box.  No doubt a considerable amount of the rest is also going to end up permanent;  whether or not that was avoidable is another question.

What I don’t understand is the lesson Chris takes from this, for which I quote (please subscribe and read the whole article):

This should not be a surprise because there was never a natural underlying productivity growth rate in individual industries before the crisis. Instead, Britain had a series of fizzing sectors – oil, manufacturing, and then finance – which maintained the appearance of stable and rapid economywide growth. Then our luck ran out.  Cyclical forces should drag up the performance of productivity, but no one should count on pre-crisis growth rates to resume, let alone a return to the previous trend.

My emphasis.

Perhaps there is an implicit assumption that, absent the structural and cyclical accidents of the last 6 years, these sectors might have merely kept the same productivity for seven years.  I would find that a rather strong assumption, to say the least.  But in a sense, the analysis appears to move from a set of explanations for one-off negative shocks to various sectors to a very strong conclusion about the sustainability of the trend before, and its ability to continue going forward.  I don’t think McCafferty goes this far at all. And to remind us quite how long this trend has been, here it is, in logs:


It is the standard propensity of capitalist economies to produce fizz.

Even during the dire 1970s, Britain usually found ways of getting about 1% more out of each worker each year.  From 1980 to 2005 it looks more like 2%.  But one should not expect this to be anything but very lumpy at a sectoral level.  In a good economy, the financial sector switches resources between sectors all the time, and labour moves around too.  So you should expect the story underneath this graph to show sectors moving in fits and starts – including in the last few years.  This doesn’t in anyway mean that the growth was illusory.  We will no doubt lean more heavily on some sectors than others to “fizz”.

In fact, let’s go further. Here is the USA going back to 1840, per capita GDP (Maddison data):

US Log Per Capita Maddison

There are plenty of useful insights to be gained by examining sectoral data. I wish I had the time to go further.  But the overwhelmingly striking fact about the collapse in GDP per unit of input* is how it coincided with the biggest ever fall in cash spending in the economy relative to trend – the biggest cyclical event since the war.  Since that event, our mostly service driven economy has failed to get more than around 0% more per worker per year.

I think while we are in this funk it is impossible to make big statements about future trend rates.  The economists Chris mocks for always forecasting an imminent return deserve the mockery – forecasting deserves that.  Clearly something has changed significantly in our economy since the last time we had a recovery.  Some of it must be a humbled financial sector and drier oil fields.  But much more of it feels like it is in the labour force – as Van Reenen et al analyse here: “Much of the explanation for this seems to lie in the increased real wage flexibility and lower investment seen in the UK in recent years.”

Chris has done an extraordinary service by continuing to bang on about this important topic.  I bet he longs for the debate to be solved. But I just don’t think this McCafferty speech does that.

PS> Samuel Brittan, writing about the 1966 election campaign, reminds us what a real productivity problem looks like.  The Conservatives wanted to campaign on “9:5:1”.  That is the ratio between the increase in wages, prices and production.  GDP up 1%, wages up 4% in real terms?  There’s your productivity problem …

*it winds me up too that we focus so much on GDP/Labourer.  When a collapse in nominal demand is expressed in the number of workers employed, people cry “demand problem”.  When it comes through their unit cost -the wage – they immediately say “productivity problem”. We should instead focus on GDP/Labour*Capital …



2 thoughts on “No, Ian McCafferty has not just “solved” the productivity puzzle

  1. Nice post. I agree like in the 30’s we will see a sharp increase in productivity once nominal spending catches up. Probably we will also see articles explaining the cause is due to robotics or something.

    1. Yes-when it takes little technological breakthrough to fill shops and restaurants more, pay health care staff better, all that. But following sight of Chris Giles’ more granular data I need more thinking here too….

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s