Productivity: the long global view
We’re often told that the key to raising wages is to solve the ‘productivity puzzle’. Output per worker has stagnated since the financial crisis, leading economists to worry that the outlook for the economy is bleak. But a range of evidence across many countries suggests that the fall in productivity can be dated back to long before the financial crisis. The outlook for the economy and wages might depend on better understanding these longer-term trends.
Productivity was the starting point for Margaret Thatcher’s diagnosis of the failures of the British economy:
In spite of what might seem the more immediate and pressing problems of strikes, price competitiveness and international recession, the root of Britain’s industrial problem was low productivity. British living standards were lower than those of our principal competitors and the number of well-paid and reasonably secure jobs was smaller because we produced less per person than they did. Some twenty five years earlier our productivity was the highest in Western Europe; by 1979 it was among the lowest. (The Downing Street Years, p. 93)
Go forward nearly forty years, and here’s Phillip Hammond in his speech for the 2016 Autumn Statement:
Mr Speaker, The productivity gap is well known, but shocking nonetheless: We lag the US and Germany by some 30 percentage points. But we also lag France by over 20 and Italy by 8. Which means in the real world, it takes a German worker 4 days to produce what we make in 5; which means, in turn, that too many British workers work longer hours for lower pay than their counterparts. That has to change if we are to build an economy that works for everyone. Raising productivity is essential for the high-wage, high-skill economy that will deliver higher living standards for working people.
So, what happened?
This blog reviews the growing body of research taking a longer-term view on productivity. I start with the OECD, going via the National Institute of Economic and Social Research (NIESR) to Andy Haldane’s valuable analysis from a couple of months ago (‘productivity puzzles’), adding a couple of charts of my own drawing on the Banque de France’s ‘Long-Term Productivity database’.
From the perspective of productivity growth a succession of Chancellors would find out that we never had it so good as before Margaret Thatcher took office. The same was and remains true for finance ministers across the world.
The OECD issued their annual productivity compendium this week (here). They describe the “post crisis period” as “a continuation of a long term slowdown in productivity growth that has gone hand-in-hand with weak levels of investment”. A veritable bombardment of charts illustrate how most countries have faced a progressive deterioration in productivity growth since the 1970s. For example:
A couple of weeks ago, NIESR published their Quarterly Review. In his thoughtful introduction, ’the economic landscape of the UK’, their Director, Jagjit S. Chadha took a longer view on the UK experience. His chart of labour productivity per head is shown below.
Discussing a contrast between the ‘more discretionary demand management policies’ of the early post war decades and the later ‘reliance on policy rules’ for inflation and fiscal policy, he observed:
The irony, to some great degree, is that the widespread adoption of rules for fiscal and monetary policy has not improved long-run performance and indeed may have done little to encourage addressing the more fundamental questions of productivity growth. … both total factor productivity [see below] and labour productivity seem to have descended to ever lower levels in the postwar period …
While Mrs Thatcher may have bemoaned the poor performance of British productivity, she was criticising a productivity performance that (from the point of view of growth) deteriorated substantially after she took office.
3. Andrew Haldane
Andrew Haldane’s March 2017 speech: ‘productivity puzzles’ set out a comprehensive assessment of the long-run evidence on productivity. Based on a sample of 116 counties from 1951 to the latest available years, he reported summary statistics for the global distribution of productivity. He shows that:
- robust productivity growth was the norm for the two decades after the war;
- these gains began to retreat from the 1970s, when lower but more erratic outcomes became the norm; and
- since the 1970s these more limited gains have become unevenly shared, with richer countries less disadvantaged.
Chart 7 in the published version of his speech shows the distribution of productivity growth across his country sample by year, showing the median and the upper and lower quartiles. He concludes:
First, the slowdown of productivity growth has clearly been a global phenomenon, not a UK-specific one. From 1950 to 1970, median global productivity growth averaged 1.9% per year. Since 1980, it has averaged 0.3% per year. Whatever is driving the productivity puzzle, it has global rather than local roots. Second, this global productivity slowdown is clearly not a recent phenomenon. It appears to have started in many advanced countries in the 1970s. [Haldane’s emphasis]
[Note his analysis is based on total factor productivity (TFP), which measures productivity per unit of labour and capital input; routinely released statistics tend to be based on productivity per worker or per hour.]
On chart 8 outcomes are examined separately for emerging and advanced countries. Both groups briefly enjoyed strong outcomes until the first half of the 1960s, when emerging economies began to fall behind. Both groups went into sharp decline at the start of the 1970s; afterwards advanced economies moved to greatly lower growth, and emerging economies into decline.
The trend was bucked from the new millennium when emerging economies picked up pace and for a brief period began to outperform advanced economies. However over recent years gains narrowed, with both groups of countries converging at near zero growth in the latest observation.
Chart 11 shows an alternative perspective to chart 8, comparing countries with the US. Advanced and emerging countries were catching up until the 1970s, and then fell behind again. Median TFP for emerging economies is now around 50 per cent of the US level, when at the peak it came close to 75 per cent (by eye). Advanced economies flat-lined in relative terms, then the gains towards the end of last decade more than evaporated at the beginning of the latest one.
4. Banque de France database
‘The Long-Term Productivity database’ was created as a project at the Bank of France in 2013 by the French academics Antonin Bergeaud, Gilbert Cette and Remy Lecat. The latest version now includes data for 17 countries. The chart below shows productivity growth by decade for selected major economies, and an average across all countries. This shows a progressive deterioration from a high point perhaps in the 1960s. In the most recent decade productivity growth has all but ground to a halt.
Productivity per employee, decade annual average percentage growth
In terms of relative performance, over the first four decades the UK conformed more closely to (though outperformed) the US and Canada than the other European countries. (My instinct is that the pace of growth in the latter group may have been related to the extent of the ground they had to make up after the Second World War.) Over the next three decades performance has been more uniform across all countries. The US has emerged from the rear of the pack to closer to the front of the pack in growth terms. But the dominant feature is the trend over time rather than the differences between countries.
5. A word on productivity levels
The discussion above has focussed on growth rates, but commonly politicians refer to comparisons of levels – blaming the average UK worker (or firm) for being x% less productive than the average French worker etc. Given relative country growth performance has been in line for decades, these level divergences in productivity must be very long-standing. The Banque de France dataset allows us to explore this – the chart below shows (for a subset of the countries) productivity levels relative to the US (the lines approach the US line when productivity catches up and vice versa).
The UK was the frontrunner at the start of the twentieth century, but fell decisively behind the US in two major steps coinciding with each of the World Wars.
Germany, France and Japan started the century far behind the UK and US, and then the World Wars set them even further back. But their post-1945 recoveries were relatively immense, with (on this basis) France overtaking the UK in the mid-1970s and Germany doing so in the mid-1980s. Japanese progress came to an abrupt halt when their financial crisis hit at the start of the 1990s. On this measure of productivity per employee, Germany and France then moved ahead of the US in the late 1980s early 1990s, but from the mid-1990s the relative gains came to an end and they converged on the UK and US positions.
Productivity per employee levels, US=100
In the parochial context, the period in which Mrs Thatcher took office saw little improvement in the relative position of the UK. In fact a convergence with the US that had been underway since 1965 ran out of steam in the early 1980s. France and Germany then ran out of steam from the mid-1990s, and for a brief period the UK did less badly, until the financial crisis set matters further back.
[It should be noted that level figures must involve significant statistical uncertainties, given national accounts data were only really prepared formally after the Second World War and there have been measurement challenges throughout.]
Policy makers in the 1970s made productivity central to the case for reform. But the policy choices made in the ‘80s and ‘90s: more restrictive monetary and fiscal policies, but set against looser capital controls, greater global monetary integration and a massive relaxation in financial restraint, coincided with (at the very least) a significant slowdown in productivity. Both in the UK and the global context, a historical perspective suggests that they were in fact looking back at a golden age in productivity terms.
It’s not just post-crisis policy that we need to examine if we want to solve the problem of how to deliver a higher productivity, higher wage economy. It’s not to want to live in the past to ask whether aspects of policy used to be better done.