Economic Growth by 2xuH469


									Economic Growth V:

  Christine Greenhalgh
     Hilary Term 2008
Why productivity matters:
• “Productivity isn’t everything, but in the long run it
  is almost everything” (Paul Krugman)

• Productivity growth drives growth of real wages
• Permits government tax revenues to rise without added
  burden on taxpayers
• Welfare effects from improved allocative efficiency are
  swamped by benefits of rise in productivity
• Productivity growth is one factor underpinning
  competitiveness – see tomorrow’s lecture
Lecture outline
• How do we define and measure productivity?
• What factors determine productivity growth?
• Why did the UK under-perform during the
  ‘Golden Age’ of the 1950s and 1960s?
• To what extent did the UK catch-up in the 1980s
  ‘Thatcher era’?
• Has the New Economy (ICT) made a difference in the
• How is the UK performing today relative to its major
• What are the five drivers of productivity according to New
Measuring productivity growth
• Growth of output = weighted growth of all inputs plus
   growth of efficiency or total factor productivity (TFP)
• Growth of inputs includes increases in Capital, Labour,
   (Materials and Energy)
Simple measure of productivity -
• Labour productivity growth or growth of output per
• This measure will vary if other inputs like capital vary
More sophisticated measure of productivity -
• Growth of total factor productivity = growth of output
   minus weighted growth of inputs per worker
• TFP growth is a macroeconomic measure of the impact of
   technology factors driving efficiency in production
Partial productivity measures
Basic idea of productivity measuring ‘output’ per unit of ‘input’
Output = GDP = value added
Labour productivity measures
   ‘Value added per hour worked’ is perhaps the best.
   ‘GDP per worker’ is another measure of labour productivity

     Y  AK  L1              where 0    1
            Y    AK 
            Ak  k= capital to labour ratio
            L     L
Assuming A fixed, workers with more capital will produce more
e.g. k in 1992: UK 22k, Ge 41k, US 36k, Fr 37k
          (Penn World Table 5.6, const 1985 $)
Total or Multi-factor productivity
• Unease with partial measure gave rise to TFP (or MFP)
• Calculating the growth rate of TFP is most common
• The growth of TFP attempts to identify the increase in value
  added not associated with increases in labour or capital.
• As can be seen below, this can be linked to a simple
  production function where A is ‘technology’
Y  AK  L1          where 0    1, taking ln and diff. w.r.t. time
ˆ ˆ        ˆ           ˆ
Y  A   K  (1   ) L where  means growth rate, or
                ˆ ˆ       ˆ            ˆ
growth of TFP  A  Y   K  (1   ) L

Hence, TFP growth can be thought of as the rate at which the
production frontier shifts out over time
Growth accounting
The derivation of TFP is directly linked to a form of analysis
   called growth accounting
Growth accounting analyses how factor inputs and value added
   have grown, and attributes the remainder, or residual, to
For example, assume GDP depends on physical capital (KPHY),
   information-comm-tech capital (KICT), labour (L) and TFP
One can try to assess how much of GDP growth is due to each
   factor, with the residual being allocated to TFP
(To do this one must know functional form of production
   function, e.g the α weights)
        ˆ      ˆ           ˆ           ˆ   ˆ
       Y   P K PHY   I K ICT   L L  A
                  P    I
      from Y  AK PHY K ICT L L
Sources of new technology for TFP
TFP rises due to innovation resulting from both
    • Product innovation - Higher quality or new products
    • Process innovation - Better ways to use existing inputs
Domestic invention:
    • Good public science base
    • Private R&D in firms
Technology transfer:
• Learning by imitation of best practice of leading countries
• Role of FDI in bringing new techniques
• Role of trade in helping firms to imitate foreign products
(Refer to lectures on Convergence in Growth Theory and see
   tomorrow’s lecture)
Complementary and inhibiting factors for
raising TFP
Complementary factors:
• Human capital – investment in education and training
   (Refer to lectures on Education and Training in Micro
• Rate of capital investment
   New capital can embody technical progress
Inhibiting factors:
• Trade Unions?
• Lack of an entrepreneurial culture?
• Uncompetitive market structures?
UK under-performance in Golden Age
Notes on the Maddison table:
• GDP per person = GDP per worker times the
  employment/population ratio
• Thus, although related to labour productivity, it is also
  affected by demographic factors
• Of the 16 countries, only four grew at below 3%, of which
  the UK is the lowest at 2.2%
• Convergence? Evidence supports, as Col.1 (income in
  1950) is inversely related to Col.3 (subsequent growth rate)
• But also see overtaking – 7 countries which were below
  UK in 1950 were above UK by 1979 in GDP per person.
Broadberry’s and Crafts-Toniolo’s views
• The UK could not have grown as fast as Germany and
  France in the Golden Age since it had fewer technology
  catching-up opportunities and less scope to move labour
  out of farming
• Nevertheless, growth was lower than it could have been by
  about 1 per cent a year
• This was due to poor supply-side policies, such as the
  failure of industrial relations, lack of competition in
• There were only modest increases in the supply of highly
  trained and educated workers
• Investment from 1950 to 1973 was lower in UK than in
  France or Germany
1970s, 1980s and Thatcher ‘miracle’
•    Like most OECD economies, manufacturing TFP growth in the UK
     slowed in 1970s
•    Common factor in 1970s - all countries adjusting to oil price shocks,
     struggling to contain imported inflation and adjusting to new patterns of
     international comparative advantage
•    UK manufacturing experienced high labour productivity growth in 1980s,
     attaining a growth rate of almost 4% p.a., better than performance in the
     Golden Age
                                                   Source: Oulton (1995)
    UK output per person employed
    Average annual growth rates
                                     Manufacturing       Whole economy
    1960-68                                3.43                 2.80
    1968-73                                3.88                 3.10
    1973-79                                0.62                 1.21
    1979-89                                3.98                 2.00

Manufacturing: in 1990, 24% of GDP; by 2006 16% (but 55% of exports)
Reasons for the ‘Thatcher miracle’
Two possible explanations for the slowdown and speedup:
  • Structural change: institutional rigidities and strong
     unions in 1970s held back growth
  • Exaggeration due to measurement biases: capital
     scrapping; labour hoarding; single deflation bias
  Changes in 1980s:
  • weakening of trade union power
  • withdrawal of state-subsidies
  • increased subcontracting
  • catch-up to international best practice
  • rise in foreign direct investment
 Manufacturing productivity – Cameron (2003)
                                                 average % growth per annum
Growth accounting decomposition of value added p.w.
per worker (Y/L)                      1960-73       1973-79       1979-90     1990-95
value added / worker                    4.20          1.50          4.62        3.46
capital stock /worker                   1.62          1.35          1.59        1.26
Standard calculation of TFP (MFP)       2.58          0.15          3.03        2.20

Decomposition of TFP
Bias in calculation                    0.46          1.73            -0.28     0.36
Trend                                  3.04          1.88             2.75     2.56

Decomposition of trend (sources of %)
Skills                                 0.52            0.34          0.29      0.22
Union                                 -0.11           -0.06          0.25      0.06
R&D                                    0.92           -0.11          0.50      0.55
Unexplained component of TFP           1.72            1.72          1.72      1.72
 Skills = ratio of admin/technical/clerical staff to total workers
 Union = full time manual males covered by collective agreement / total workers
 R&D = business R&D to capital stock
Cameron’s analysis of ‘Thatcher miracle’
• 1st panel decomposes ΔY/L into ΔTFP and ΔK/L
• Note much greater variation in ΔTFP than in ΔK/L

• 2nd panel estimates trend ΔTFP after eliminating
  measurement errors, business cycles and seasonal effects
• Trend ΔTFP is now greater in 1960s than in 1980s

• 3rd panel decomposes trend ΔTFP
• Rising skills and increasing R&D contributed to 1960s
• Unions were negative influence in 1960s and 1970s, but
  fall in unionisation was positive in 1980s and 1990s
Structural Change in the Thatcher era
Greenhalgh and Gregory (Oxford Bulletin, 2001)
• Data 1979-90 for 6 sectors: distinguish high and low
  technology sectors in both manufacturing and services;
  also distinguish traded and non-traded services
• Big story was the rise in financial and business services:
  doubled its share of value added (GDP);
  grew even faster as share in gross output (economic activity
  before netting out inter-industry purchases)
• Shrinking sectors were primary (mining) and non-high-tech
• Rapid productivity growth in financial/business services so
  differential sector growth was contributing to rise in TFP
   Shift-share analysis
Shares of total growth                     Between               Within                Total

TFP                  Whole economy         17.1                  82.9                  100.0

                     Manufacturing         10.2                  89.8                  100.0

Labour               Whole economy         4.4                   95.6                  100.0
                     Manufacturing         3.0                   97.0                  100.0
Source: Gavin Cameron, James Proudman, and Stephen Redding, ‘Deconstructing Growth in UK
Manufacturing,’ (London: Bank of England Working Paper 73, 1997).

   ‘Within’ component shows how much is due to the growth in productivity within
       individual sectors of the economy
    ‘Between’ component shows how much is due to movements of labour and capital
       between sectors of the economy
   While most productivity growth occurs within sectors, the reallocation of resources
       contributes one sixth of whole economy TFP growth
1990s onwards: GDP per worker
 Overall, UK gains on competitors in GDP per worker (except
 US post 2002)
 UK now level with Germany, but 15% more hours per worker
150                                                                         France             Germany
                                                                            Japan              UK





      1991   1992   1993   1994   1995   1996   1997   1998   1999   2000   2001     2002   2003    2004   2005

 Source: Office of National Statistics, International Comparison of
 Productivity, Jan 2007
 Comparing GDP per hour
UK has gained on competitors
But, still 15-20% lower that FR, GE, US
                                                  GDP per hour, UK indexed to 100

150                                                                                 France             Germany
                                                                                    Japan              UK





       1991   1992   1993   1994   1995   1996   1997    1998    1999     2000      2001     2002   2003    2004   2005
      Source: Office of National Statistics, International Comparison of
      Productivity, Jan 2007
Broadberry and O’Mahoney (see The UK’s Productivity
Gap, ESRC, 2004)
• Attempt to decompose year 2000 differences into:
(a) Physical capital
(b) human capital (skills)
(c) total factor productivity (TFP)
UK vs US: TFP accounts for 2/3rds, capital 1/3, skills about
  the same
UK vs FR & GE: Physical capital differences explain 75%,
  skills 20%
Can also look at sector differences:
Wholesale and retail, finance, and machinery show largest gaps
 with US
UK manufacturing TFP cp. to the USA
Source: Cameron et al (2005)
Industry        RTFP   RTFP   RTFP 70 RTFP 92
                1970   1992    skills   skills
                              adjusted adjusted
Food & drink    71.5   55.8     68.4     55.3
Chemicals       49.5   66.7     49.4     66.6
Primary metal   49.2   62.2     49.7     62.5
Machinery       82.0   73.0     79.5     72.4
Elec. Eng.      59.8   53.4     58.9     52.7
Transport       46.3   59.8     44.8     59.9
Instruments     62.8   78.8     62.1     81.2
Mean            57.1   65.3     56.2     65.2
Industry-level TFP comparisons UK/US
• The above results show the relative level of TFP in the UK
  compared with the US for selected industries
• Cameron et al (2005) show 14 sectors which are included in
  the Mean value quoted here
• Two sets of estimates reflect two difference measures of
  labour inputs being netted off to get TFP (second corrects
  for the skill composition)
• Good performance – UK gained ground in 11 sectors
• Biggest climbers were Chemicals, Metals, Transport and
  Instruments ( includes high technology sectors)
• Biggest falls by Food, Machinery, Electrical Engineering
  (also includes some high tech sectors)
Evidence of ICT capital deficit relative to US
From DTI March 2004 Raising UK Productivity Data for 1999
                  Capital per employee   Capital per employee hour
ICT Capital
UK                          34                      37
US                          100                     100
France                      37                      43
Germany                     28                      34
Non-ICT Capital
UK                          64                      69
US                          100                     100
France                      92                      108
Germany                     78                      97
EU/US comparison of ICT productivity

  Change in annual growth in output per hour from 1990–95 to 1995–2001
                                         Increase in annual
                                         growth rate – from
                                         1.2% in 1990–95 to
                                  U.S.                                 EU
                                           4.7% from 1995                    Static growth – at
                                                                            around 2% a year –
  ICT-using sectors                                                         during the early and
                                                   3.5          -0.1             late 1990s

  ICT-producing sectors                   1.9                                    1.6

  Non-ICT sectors        -0.5                            -1.1

  Source: O’Mahoney and Van Ark, 2003                                                                        3
 Investment since 1990
Business investment per worker ($) – recent trends

G5 comparison, 1990-2002 in US dollars at 1995 prices and
purchasing power parities. Source. DTI (2003), origin OECD
NOTE: UK’s high participation ratio affects ratios to Ge, Fr
Why does UK have low capital/worker?
Could be ‘supply’ constraints (i.e. availability of funds)
    City of London ‘too short term’
    Banks lend too little & equity too expensive
    Active merger and acquisition so management short term
    Macro-instability (interest and exchange rates)
    Low levels of public capital (transport, ports)
Or low demand outcome (i.e. consequence of few
    Unskilled workers
    Low and ineffective R&D, weak innovation
    Weak university/science-industry links
     Conclude low investment is symptom not cause?
UK policy framework since 2000
– ‘Five Drivers’ of productivity
• Investment - physical capital gives direct impact of rise in
  capital per worker + indirect impact of implementing new
• Skills - skilled workforce will be flexible and able to adapt
  to new technology
• Innovation - increase R&D spending to generate higher
  stock of knowledge
• Competition - affects incentives to innovate or reorganise
• Enterprise - improve attitudes and capabilities of
HM Treasury (2000) Productivity in the UK: The Evidence and the
  Government’s Approach
Fawcett and Cameron’s review (2005) of
the evidence for the ‘Five Drivers’
• Investment – clearly positive on output per worker but
  neoclassical growth shows impact on level of productivity
  not growth.
• Not clear how strong is indirect effect on TFP from being
  complementary with innovation.
• Do UK firms under-invest? Some evidence that UK firms
  face more financial constraints from capital markets than
  EU firms
• Skills – positive on productivity, but again not clear
  whether this is a levels or a growth rate effect – depends on
  choice of growth model: Mankiw (levels) v. Lucas (growth)
Fawcett and Cameron’s review (ctd)
• Innovation – focus on evidence of returns to R&D Key
  issue is whether social returns are higher than private returns in
  which case firms will be under-investing – Griliches says
  there is evidence of this difference in returns; he also found
  returns to ‘basic’ (more scientific) research were highest
• Competition – direct benefits of competition on
  productivity growth (Nickell)
• Indirect effect of competition on innovation - evidence
  suggests some degree of market concentration better than
  perfect competition (Schumpeter’s thesis)
• Enterprise – data/evidence rather scarce, but some
  evidence of need for good entrepreneurs to exploit the
  potential of results from R&D
Summing up
• Long-run growth in living standards is driven by improving
  technology – UK lagged others in Golden Age of growth
• Level of GDP per capita is driven by capital/labour ratio
• About half of the UK ‘productivity miracle’ in 1980s was
  due to mis-measurement and about half was due to an
  improvement in the supply-side of the economy
• Despite recent improvements in productivity there is not
  much sign of a big ICT effect on productivity in the UK
• Europe has large deficit in ICT capital relative to the US
• UK has deficit in non-ICT capital relative to US and EU
• UK’s GDP per capita is roughly the same as that of France
  and Germany, despite productivity being lower
• UK is able to do this by working longer hours and having a
  higher employment rate of the population (so less leisure)
Policy pointers and problems
• Government believes important factors driving
  productivity are investment, skills, innovation,
  competition, entrepreneurship
• To improve output per worker UK needs a higher
  investment rate – this has not been helped by higher
  relative interest rates and overvalued sterling in last few
• Introduction of UK R&D tax credit by Brown in 2000
  (small firms) and 2002 (all firm sizes) should have helped
  firms with their costs of R&D
• Higher education and skills will also help, but only if skills
  relevant to jobs being created in a changing economic
  environment, so correctly predicting future structure of
  occupations and key skills is crucial to success

To top