What has happened to NHS capital per worker?

 1. a New report "From Diagnosis to Delivery" by Allas et al has, on p.45 some information.

2. The figure is a figure relative to other countries. I don't put much store by that, what matters is how productive we are with the capital not how much we are spending.

3. The figure below says NHS capital per worker has fallen by 36% in real terms since 2010. i think this is a capital stock per worker figure.

4. The ONS data tell a different story.  That is capital services per worker.  If you go to 

https://www.ons.gov.uk/economy/economicoutputandproductivity/publicservicesproductivity/datasets/publicserviceproductivityestimateshealthcareengland

you can download the 1996-2023 data

from which you get that since 1996 capital has grown 7.6% faster than labour, but from 2010, -20%.  So more capital per worker over the whole period. 

UK productivity growth history since 1997

 A very useful graph from the Resolution Foundation that I hope they don't mind my posting 

How important are intangibles in accounting for productivity growth?

1. I was asked today: if intangible capital deepening slows, is that a big effect on productivity growth?  This is part of a broader question; what's the relative importance of intangible assets in accounting for productivity growth (value added per hour growth)?

2. Growth accounting allows a decompositition of labour productivity growth into the contributions of:  

a. reallocation = workers moving between industries of different productivity levels 

b. labour composition = increased skills, age and experience

c.  ICT capital deepening = increased ICT capital (computers, comms equip) per hour 

d.  NonICT capital deepening = increased NonICT capital (buildings, vehicles, non-ICT plant) per hour 

e.  Intangible capital deepening = increased intangible capital (R&D, software, artistic originals, design, marketing, business process, training) per hour 

f. TFP = increased total factor productivity (what's left over, which increased efficiency plus mismeasurement etc.)

Here are some results from our EUKLEMS-INTANProd database, in hopefully obvious notation.  Countries are US, UK, EU (France, Germany, Spain, Italy, Denmark, Holland, Austria, Sweden, Finland). Industries are all ommiting A (agriculture), OPQ (defence, education, health), B (mining), D-E (gas, electricity, water), F (construction).  All these are not well measured, and/or mostly public sector. 

As the results show, intangible capital deepening is much the most important contributor to labour productivity growth in the 2011-19 period (tangible capital deepening contributions is the sum of ICT and NonICT, which is still lower than the intangible contribution).

Table:  of contributions

 

And a picture

As we saw in a recent blog the employment rights bill will very likely lower intangible investment. That will lower intangible capital deepening, other things equal.  This then is a policy measure lowering the major source of UK productivity growth.  Not a good policy if you want growth to increase. 

Did persisently low interest rates post GFC lower productivity growth?

Many allege that the long run of low interest rates post global financial crisis lowered productivity growth via zombie firms.  These low productivity firms survived more than they should have done and hence productivity growth stalled.  

An alternative view is that low productivity growth, for other reasons, lowered r* and hence interest rates.  

A paper "Aggregate productivity decompositions using structural business surveys: Evidence from the UK by Russell Black, Rebecca Riley and Garry Young", available here sheds a bit of light on this for the UK.

It uses UK company data to decompose productivity growth into that 

a. within surviving companies

b. reallocation of market share between surviving companies

c. the net effect of exit and entry.

One might think that the zombie firms view would say that the net entry/exit effect would be less as fewer low productivity firms exit.  

Their chart shows this isn't the case.

1. Using various different methods the change in the net entry effect, see middle panel is very small, a slight fall.

2. instead, the fall in productivity growth is due more or less equally to falling within firm growth and falling between firm reallocation.  The latter might be a zombie phenominon, but it isn't clearly so.

Update, with some numbers
1. using the Foster. Halitwanger, Krizan decomp, the 5 year interval, 99-07, within effect is 1.95%, net entry 0.22% (total LPG is 1.4).  The same data, 2011-19 are 0.61%, 0.08% (-0.32%).  

2. so the net entry effect has slowed, but it's 8% of the slowdown.  The change in within effects are 80% of the slowdown.

Supply chains and the UK productivity puzzle: a framework

An interesting Productivity Institute meeting today on supply chains. 

One of the features discussed about the UK productivity puzzle is under the heading of “supply chain management”.  Maybe British supply chain productivity is very low. Maybe British supply chain productivity is not resilient enough, see COVID and issues around Chinese involvement in 3G technology for example.  so how should we think about supply chains? How should we answer questions about whether supply chain management is or is not adequate?

Let's start with an example.  Suppose we have three types of law firms.

1.     Law firm 1 employes a building, a receptionist, an operations manager, and a load of lawyers.  The process within the law firm consists of the following.  The client comes into the building and is greeted by the receptionist who then takes the client up to the lawyer.  The lawyer gives the client an opinion , and the client pays and walks away.  The operations manager designs the process by which the receptionist takes the client up to the lawyer (offering them tea, helping with directions to the next appointment etc.)  Notice, the client never sees the operations manager, and although the client sees the receptionist, neither the receptionist nor the ops manager is a trained lawyer.

2.     We then have some definitions as follows:

a.      Industry.  the law firm is in the law industry because its output is legal services.

b.     Process. The visible processes which the law firm is involved in consists of the process carried out by the receptionist and the process carried out by the lawyer.

c.      Activity.  The law firm is in fact involved in three different activities:

                                                                         i.      the provision of reception services,

                                                                       ii.      of operations services ,

                                                                    iii.      and legal advice. 

d.     Lots of the lawyers in law firms moan endlessly about the fact that out of their fees comes the expenses of the receptionists and the operations managers, none of whom know anything about the law. Likewise academics who complain about administrators and admission staff who know nothing about academia, footballers who complain about groundstaff who can't play football etc.

3.     now consider law firm 2. They have contracted out reception services to a separate firm, who has simply bolted a flat screen TV screen to the wall of the front office and provides receptionist services remotely.

4.     Now consider law firm 3. They have also contracted at reception services, but they as well they have contracted out operations management to a management consultant. the management consultant as supplied them with the book setting out a set of routines to which, let us say, the receptionist adheres, when taking the client up to the lawyer.

 

What can we say about productivity and the adequacy or otherwise of supply chain management in these examples?

1.     Let's start with the definition of a supply chain. Law firm one has an entirely internal supply chain, in this case the process by which the receptionist hands over the client to the lawyer. That supply chain is presided over by the operations manager.  Law firm three has a supply chain, but it is external. That is to say, the services provided by the receptionist and by the operations manager in simply bought in externally.

2.     Going back to the activities that are involved, in firm one the operations manager provides operations advice services, but this is done internally. If one had the management accounts for this firm, and one could figure out the wages and overhead costs of the operations manager, one could therefore figure out the costs involved in the provision of those advice services. In case 3 the matter is much easier; one just looks at how much the firm is paying to the management consultant.

3.     What about measuring productivity? there would seem to be two methods:

4.     method 1.  Process.

a.      In law firm one, break up the firm into the different processes that are involved. Try to measure the productivity of each process. So there will in practise be two output measures; first, the output of the reception process, and 2nd the output of offering legal advice after the client has been through the process of reception.

b.     In this case, one would have to try as well to allocate costs such as the operations manager into each process. This of course it's much easier to do in the case of law firm 3 since every part of the process is transacted for. Thus there is going to be an observable price and quantity for the entire provision of legal advice.

5.     Method 2.  Activity.

a.      For every law firm count the output as the provision of legal advice, so there is only one measure of output. However, account for the fact that different law firms undertake different activities, and these activities might potentially contribute to productivity. In each of the examples the activities are (a) reception activities, (b) giving-legal-advice-activities, and (c) operations activities.  

b.     Although measuring the activities is hard, by taking an activities rather than a process approach, one avoids the almost impossible problem of subdividing the output of legal advice into lots of different processes.  Indeed, in many law firms , who keep time sheets of chargeable hours, it might very well be feasible to gather data on the activity, because it is the different activities that are documented on the charge sheets (for example many charge sheets might categorise the activities of being with clients, marketing and customer acquisition, personnel and administration). 

c.  This activity approach would, under certain circumstances, count the knowledge of the ops manager as an intangible asset.  For more on intangibles, see here. 

6.     (Finally, an overall point on the economic modelling of all of this.  As Milgrom, Roberts and others Have pointed out, economists have a very stripped down model of the firm. When economists write down what is called a production function, they assert, on the face of it quite sensibly, the output is produced by capital and Labour. What many supply chain managers and the academics who study them will tell you is that production requires as well coordination activities. Economists tend not to bother modelling these coordination activities separately. They are either included in labour, or viewed as being small enough for one to ignore.)

 

 

 

Working from home

 

Notes on WFH

This has become super popular in the literature.  Let’s be careful not to get overexcited, and, for those who are in the professional and white collar sector, not to extrapolate too much from their own personal observation .

 

The ONS publishes data on working from home in their fantastic Business Impact of Coronavirus (COVID-19) Survey (BICS). Here are weighted results from Wave 16, link being:

https://www.ons.gov.uk/economy/economicoutputandproductivity/output/datasets/businessimpactofcovid19surveybicsresults

and specifically

https://www.ons.gov.uk/file?uri=%2feconomy%2feconomicoutputandproductivity%2foutput%2fdatasets%2fbusinessimpactofcovid19surveybicsresults%2fbicswave16/bicswave162supressed.xlsx

 

The table below shows some of the key figures. They start in column one and column 2 by asking firms whether more or less of their workers are working from home, referring to the week 5^th -1th October. Perhaps not surprisingly 26% on average in all industries of workers are working more from home, although 75% report no increase in working from home. Notice that the biggest increase in the number of workers working from home is in the water and sewerage industry , at 85%. Even in manufacturing, where you would imagine working from home would be rather complicated, 24% more workers are working from home.

 

 

 

 

A snapshot of working from home may not be very informative. Perhaps it's of more information therefore to ask firms whether they expect not working from home will be a permanent feature. This is set out in columns 8 and 9 (with 10 for not sure). In column 8 we can see that 17% of all industries 17% expect working from home to be a permanent feature of their business model. 65% do not expect it to be the case and 17% don't know. Returning to the water industry only 2% of firms expect this to be a permanent feature and 93% of firms expected not to be the case.

Notice that there are really only three industries which stand out as expecting working from home to be a permanent feature: information and communication and professional and scientific and education. There is fairly clear blue water between the numbers in those sectors, 37 percent, 34% at 28% and the other sectors.  This immediately suggests that perhaps the discussion of working from home might be dominated by the bias of those people who are typically writing about it.

 

Finally, we may look at the consequences of working from home for productivity. These are in columns 4,5 and 6. Perhaps the most striking result is how balanced these results are. Around 60% of firms think that productivity will be the same 19% of firms think that productivity will go down , and 14% of firms think the productivity will go up.  Once again the numbers who think productivity will rise are concentrated in a rather small number of industries most notably 46% in information and communications and 52% in other services.  Professional scientific and administrative have numbers at around 10% but otherwise none of the other industries have a particularly strong expectation of an increase in productivity.

 

Finally the scatter plot summarises some of these numbers. The vertical axis is the number expecting long run increase in WFH less those expecting long run decrease. The horizontal axis we have the net numbers in the industries who expect productivity to increase .

 

As we see in the scatter, the bulk of the industries are at the bottom left. There is some small expected increase in the numbers working from home, but productivity or net is anticipated to be lower. Rather few industries are in the top right where productivity is expected on net to increase and there was to be substantial more working from home. These are dominated essentially by ICT , other services and professional scientific and technical, where there is expected to be, on net a slight loss in productivity.  Notice finally the all industries: on 48% more firms expect less WFH then expect more WFH and 5% more of firms expect productivity to fall.

Update:

Finally, ONS also ask business who say they will have more (less) working from home why they intend to do so (not to do so) .  Of businesses not intending, 91% say that’s unsuitable for their business. Of those intending, we have

a.       64% say lower overheads

b.       61% improved staff wellbeing

c.       40% increased staff productivity

d.       22% ability to recruit from wider pool

e.       18% reduced sickness

f.        11% better able to match jobs with skills

 

 

Productivity in cars

 Numbers speak loudly. 

From Russ Robert's interview with Enrico Moretti: 

Russ: The statistic you quote that in the 1950s the average worker at General Motors (GM) could make 7 cars in a year, and now they make 28--that's an unbelievable transformation. 

Moretti: That's right. It means that for the same amount of cars sold, now GM needs 70% fewer workers.

Long run economic growth in the UK

 A fascinating new paper by Stephen Broadberry,The Industrial Revolution and the Great Divergence: Recent Findings from Historical National Accounting, looks at new long run data for the UK.  It takes up the point that much pre-1870 data e.g. in Maddison's database was based on guesstimation and we now have improved data.  Here is my reading of some of the key points.

1. Their figure 1 below shows some long run trends.

 

 

 

 As they say, some of the key findings already were that economic growth was slower in the Industrial Revolutoin than previously thought, which means the UK must have entered the Industrial Revolution richer than previously thought, to get to the same agreed final level of GDP per head.  

"Figure 1 shows the long run evolution of real GDP, population and real GDP per capita over the long period 1270-1870. GDP per capita stagnated during 1270-1348, before increasing sharply between 1348 and 1400, as population declined more sharply than GDP following the shock of the Black Death. GDP per capita then remained on a plateau between c.1400 and 1650 as population at first continued to fall and then began to recover from the late fifteenth century. A new GDP per capita growth phase started around 1650, as population stagnated and then declined slightly. Although GDP per capita growth slowed down after 1700 as population growth resumed, it remained positive and became increasingly stable, with fewer and milder years of negative GDP per capita growth. It seems, then, that the Industrial Revolution was less about growing faster and more about avoiding periods of negative growth or shrinking, than has previously been realised (Broadberry and Wallis, 2017)"

 

2. Table 2 

 "Table 2 presents the average annual growth rates for the same three series: GDP, population and GDP per capita. Notice how the growth rate of GDP per capita after 1800 was actually slightly slower than after the Black Death (1350s-1400s) and after the Civil War (1650s-1700), despite the fact that GDP growth was much faster. The reason for this was the very different paths of population in these three periods. Whereas population declined very sharply after the Black Death, and still declined slightly after the Civil War, it grew very rapidly during the first two-thirds of the nineteenth century. This points to a major difference between modern economic growth and pre-industrial growth, as highlighted by Kuznets (1966). Preindustrial growth required falling population, and this led to an increase in land per capita and capital per capita, which in turn led to higher output per capita. However, this was clearly not a route to sustained growth. For Kuznets, sustained or modern economic growth required rising output per capita together with a growing population."

 3. Structural change

"Another important aspect of modern economic growth is structural change. It has long been noted that economic development is associated with a shift in the structure of the economy away from dependence on agriculture. This has traditionally been seen as a process of industrialisation, although recent research suggests that this understates the role of services. Broadberry, Campbell, Klein, Overton and van Leeuwen (2015) note that the British economy diversified away from agriculture over a longer time span than was once believed by economic historians. Agriculture was less important and services more important earlier than widely perceived, with important consequences for sectoral productivity performance. Labour productivity growth was faster in industry than in agriculture during the Industrial Revolution rather than the reverse, as early quantification of the Industrial Revolution had appeared to suggest. The quantitative dimensions of the structural shift away from agriculture in the British economy are set out in Table 3. The first point to note is that agriculture’s share of output and employment declined in importance over time, while the shares of industry and services increased, as would be expected for a developing nation. Second, however, note that even as early as 1381, agriculture accounted for less than 60 per cent of employment and less than 50 per cent of nominal GDP, so that even in the fourteenth century, industry and services accounted for a substantial share of economic activity. Third, although agriculture accounted for a smaller share of output than employment for most of the period under consideration here, thus making agriculture a low productivity sector, this had ceased to be the case by 1801, a point first noted by Crafts (1985). Fourth, although industry increased its share of nominal GDP more rapidly than services until 1700, this ceased to be the case during the Industrial Revolution period. This may at first sight seem surprising, but can be explained by a decline in the relative price of industrial goods, as technological progress increased productivity and drove down prices. By contrast, the more modest productivity improvement in services led to an increase in their relative price, so that the share of services in nominal GDP increased more rapidly than the share of industry after 1700.

 

A fifth striking feature of Table 3 is that much of the shift of labour from agriculture to industry occurred before 1759, which has important implications for the pattern of labour productivity growth before and during the Industrial Revolution. (my italics) If, as was once believed, the shift of labour from agriculture to industry had taken place at the same time as the Industrial Revolution, then much of the growth of industrial output could be explained by increased labour input rather than by productivity growth. This counter-intuitive result was implicit in the work of Deane and Cole (1962), and also confronted more explicitly by Crafts and Harley (1992). With much of the shift of labour from agriculture to industry occurring between 1522 and 1759, there was a period of labour-intensive industrialisation (or proto-industrialisation) without dramatic industrial productivity growth, which can be tracked in Table 4. This was then followed by an Industrial Revolution, where capital deepening and technological progress raised industrial labour productivity rapidly after 1759.

 

4. Comparative growth

 

Finally, they have fascinating conclusions on comparative growth, addressing the argument about just when Europe got ahead of Asia.

 New estimates of GDP per capita during the period 1000-1870 have recently been produced in a number of European and Asian economies, making use of historical data collected at the time. These estimates show reversals of fortune within as well as between the two continents. First, they show a much clearer Little Divergence within Europe between the northwest and the rest of the continent than had been suggested by Maddison (2001), with Britain and the Netherlands overtaking Italy and Spain. Second, these estimates also show a much clearer Asian Little Divergence, with Japan overtaking China and India. And third, they show a later Great Divergence between Europe and Asia than suggested by Maddison, taking account of regional variation within the two continents. Although individual European nations or small regions were ahead of the whole of China as early as 1300, the leading Chinese region did not fall decisively behind the leading European nation until the eighteenth century. This is a lot later than suggested by earlier western economic historians such as Weber (1930), Landes (1969),or North and Thomas (1971), although not quite as late as suggested by Pomeranz (2000), who argued for parity until the early nineteenth century. However, Pomeranz (2011; 2017) has more recently accepted that his earlier claims were exaggerated and now sees the Great Divergence as dating from the eighteenth century.

 

 

Finally, some lessons for process.  

"One of the most interesting developments of the recent wave of research in historical national accounting has been the construction of annual estimates of GDP per capita reaching back to the thirteenth or fourteenth century for a number of countries. Using these data, a radically new picture of the Little and Great Divergences has appeared. Northwestern Europe forged ahead of the rest of Europe and also diverged from Asia not by growing faster during periods of positive growth, but rather by reducing the frequency and rate of shrinking during periods of negative growth (Broadberry and Wallis, 2017)....

 "Explaining the Industrial Revolution has more in common with solving the problem of development today than is usually acknowledged. Getting growth going in the first place, the traditional focus of analysis, is only part of the story. Just as important is ensuring that periods of positive growth are not followed by periods of negative growth, or shrinking. This has been highlighted in the case of developing economies today by Easterly, Kremer, Pritchett and Summers (1993) and Pritchett (2000). For the transition to modern economic growth in Britain during the Industrial Revolution, it means paying as much attention to the absence of negative trend growth after the gains of the post-Black Death growth episode as to the innovations that started episodes of positive growth during the eighteenth century."

". One way to think about Europe’s Little Divergence, and also the Great Divergence, is therefore not so much the beginning of growth, but rather the weakening and ending of periods of shrinking"