Having emphasized that each of the three measures of

Having emphasized that each of the three measures of economic activity at the sectoral level
is distinct, we also want to note that each of them has its limitations as a singular measure. For
the case of sectoral employment shares, a key issue is that employment may not reflect changes
in “true” labor input, for example, because there are systematic di ff erences in hours worked or
in human capital per worker across sectors that vary with the level of development. For the
case of value added and consumption expenditure shares, a key issue arises from the need to
distinguish between changes in quantities and prices. This is often di ffi cult empirically because
reliable data on relative price comparisons across countries are hard to come by. In addition,
consumption and production need not coincide because of the presence of investment and of
imports and exports, so that neither measure alone is su ffi cient.
2.2 Production Measures of Structural Transformation
In this subsection we document the patterns of structural transformation based on examining
production measures in several di ff erent data sets. We first review the available historical time
7

Figure 1: Sectoral Shares of Employment and Value Added –
Selected Developed Countries 1800–2000
Employment
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
0 . 6
0 . 7
0 . 8
6 . 0 6 . 5 7 . 0 7 . 5 8 . 0 8 . 5 9 . 0 9 . 5 10 . 0 10 . 5 11 . 0
Agriculture
Share in total employment
Log of GDP per capita (1990 international $)
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
0 . 6
0 . 7
0 . 8
6 . 0 6 . 5 7 . 0 7 . 5 8 . 0 8 . 5 9 . 0 9 . 5 10 . 0 10 . 5 11 . 0
Manufacturing
Share in total employment
Log of GDP per capita (1990 international $)
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
0 . 6
0 . 7
0 . 8
6 . 0 6 . 5 7 . 0 7 . 5 8 . 0 8 . 5 9 . 0 9 . 5 10 . 0 10 . 5 11 . 0
Services
Share in total employment
Log of GDP per capita (1990 international $)
Value added
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
0 . 6
0 . 7
0 . 8
6 . 0 6 . 5 7 . 0 7 . 5 8 . 0 8 . 5 9 . 0 9 . 5 10 . 0 10 . 5 11 . 0
Agriculture
Share in value added (current prices)
Log of GDP per capita (1990 international $)
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
0 . 6
0 . 7
0 . 8
6 . 0 6 . 5 7 . 0 7 . 5 8 . 0 8 . 5 9 . 0 9 . 5 10 . 0 10 . 5 11 . 0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5
Manufacturing
Share in value added (current prices)
Log of GDP per capita (1990 international $)
0 . 0
0 . 1
0 . 2
0 . 3
0 . 4
0 . 5
0 . 6
0 . 7
0 . 8
6 . 0 6 . 5 7 . 0 7 . 5 8 . 0 8 . 5 9 . 0 9 . 5 10 . 0 10 . 5 11 . 0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5
Services
Share in value added (current prices)
Log of GDP per capita (1990 international $)
Belgium Spain
Finland France
Japan
Korea Netherland Sweden
United Kingdom
United States
Source: Various historical statistics, see Appendix A.
8

series evidence for currently rich economies. We then turn to the evidence for currently rich
and poor countries.
2.2.1 Evidence from Long Time Series for Currently Rich Countries
We construct individual time series of sectoral employment shares and value added shares over
the 19
th
and 20
th
century for the following ten countries: Belgium, Finland, France, Japan,
Korea, Netherlands, Spain, Sweden, United Kingdom and United States. 7 Since the early data is
sketchy and we want to highlight trends over long periods of time, we report the latest available
observation for each decade, if any. We note that for these historical time series we only have
measures based on production.