Productivity and Growth
Large cross-country differences in per-capita output and income.
Large differences in per-capita growth rates.
|Per Capita||GDP||Average Growth Rate|
Data are from Summers and Heston's Penn World Tables.
Bank Data 1
World Bank Data 2
Productivity and its growth are the source of high living standards.
Slowdown in productivity growth after 1973. 1990s revival of productivity ?
Y: is output (real GDP)
K is the stock of physical capital (plant and equipment)
N: labor (the number and hours of people working).
A: measure of productivity (a higher value of A means that the same inputs lead to more output)
A: Total Factor Productivity
Y/N: Average Labor Productivity
Growth of Y depends on:
1. Increases in inputs (K, N)
2. Increases in A
Factors affecting A:
1. Technological progress
2. Skill level of the labor force (human capital)
3. Cost of imported inputs (oil)
4. Weather (droughts, bumper crops)
5. The economic, legal and institutional environment
Sources of Growth (Growth Accounting): decomposing output growth in components due to growth in A, K and N.
Y = A K1/3N2/3 (2)
Output = value-added = payments to factors of production:
Y = W N + R K = Total wages + Total profits
The exponents mean that one third of output is paid to capital in profits (and depreciation), and two-thirds to labor (wages).
Then growth in aggregate output Y is:
dY/Y = dA/A + 0.33 dK/K + 0.67 dN/N (3)
where dX/X represent the percentage rate of change of variable X over the period considered (for example one year): dX/X=(Xt-Xt-1)/Xt-1
Decomposition of output per worker (labor productivity Y/N):
Y/N = A (K/N)1/3 (4)
d(Y/N)/(Y/N)= dA/A + 0.33 d(K/N)/(K/N) (5)
d(Y/N)/(Y/N) = dY/Y - dN/N
d(K/N)/(K/N) = dK/K - dN/N
Output per worker (labor productivity) grows because of:
1. Growth in A (total factor productivity)
2. Growth in the amount of capital per worker (K/N).
Components of (3) for the US:
dY/Y = 3.1% per year (1960-85)
dK/K = 3.2% (0.33x3.2 = 1.1)
dN/N = 1.9% (0.67x1.9 = 1.3)
dA/A = dY/Y - 0.33 dK/K - 0.67 dN/N (3)
0.7 = 3.1 - 1.1 - 1.3
Figure 1 and figure 2 present the data in as the annual values of (the logarithm of) A and its rate of change.
Note the spikes: there are large short-term movements in A associated with business cycles.
Caveat. Advances in technology (A) appear in:
1. Capital (new and more productive machines)
2. Labor (better trained workers)
Components of (5) for the US:
d(Y/N)/(Y/N)= dA/A + 0.33 d(K/N)/(K/N) (5)
1.2 (=3.1 - 1.9) = 0.7 + 0.5 (= 0.33x(3.2-1.9))
US-Japan comparison (real 1980$ values)
|Real Output (Y)||2083||3103||2.66||620||1253||4.69|
|Contributions to Growth|
|Contributions to Labor Productivity Growth|
In the US the numbers underlying Figure 1 imply average
annual growth rates by decade of:
|Decade||Total Factor Productivity Growth Rate (dA/A)|
The corresponding number for labor factor productivity
are (from BLS
productivity data releases):
|Decade||Labor Productivity Growth Rate ( d(Y/N)/Y/N) )|
Historical Data from BLS
US productivity growth in 1990-2002: capital deepening or total factor productivity growth?
After over two decade of high productivity growth in the 1950s and the 1960s, we observe a significant slowdown of productivity growth in the 1970s and 1980s following the first oil shock in 1973.
We see much the same thing for other industrialized countries, so this is not purely a US phenomenon.
The debate on the causes of this productivity slowdown Several explanations of the slowdown have been suggested but none has been found to be fully satisfactory:
1. The energy crisis in the 1970s (1973 and 1979 oil shocks).
2. Exhaustion of the post-W.W.II technological boom.
3. Low investment and savings rate.
4. High taxation of savings.
5. Excessive government regulations.
6. Low rate of public investment in infrastructures.
7. Decline of R&D investment.
8. Sociological explanations.
9. Decline in quality of education.
Until the end of 1995 (when the fixed-weight system was being used to measure GDP and productivity) it appeared that there was a resurgence of productivity in the 1990s: total factor productivity grew at a 1.7% per year rate while labor productivity grew at a 2.2% yearly rate.
It appeared that a decade old process of corporate restructuring, reengineering, down-sizing had finally borne its fruits and led to a major resurgence of productivity in the 1990s, spurred by a boom of investment in computer and information technologies.
However, the switch in 1995 to the chain-weight method of measuring productivity changed drastically the picture: the new chain-weight data showed that in the 1990s total factor productivity grew at a dismal 0.9% per year rate while labor productivity grew at a 1.4% yearly rate, not much above the 1970s and 1980s rates. So the great resurgence of American productivity in the 1990s suddenly disappeared overnight by a statistical wand.
These numbers looked dismal because many economists believed that the process of corporate restructuring, reengineering, down-sizing of the last decade, together with the development and adoption of computers and information technologies in the corporate world, had led to a resurgence of productivity. The new chain-wighted numbers seem to imply that such productivity resurgence never occurred.
1996 debate on this puzzle. Two views:
1. Paul Krugman ("Stay on their Backs") and others argued that the new measures of output and productivity were substantially correct and that the productivity benefits of the Information Revolution had been overstated.
2. Those (see Stephen's Roach piece on "US: At Odds with the Productivity Revisionists") arguing that the new chain-weight method underestimated output and productivity because, among other reasons, of mismeasurement of the growth in productivity in the service sectors. See also similar views by Fed chairman Greenspan on this issue (and Roach's conversion to the other camp in his three revisionist pieces: MSEF May 8, MSEF May 9, MSEF May 14).
For more on this on-going controversy read the debate on Productivity Growth in the 1990s.
Latest Quarterly Labor Productivity Statistics from the Department of Labor home page:
1. High productivity growth in the manufacturing sector in the US in 1996.
2. Low measured productivity growth in the service sector.
Returns to Education
Education is an investment in people (or in "human capital).
Evidence that education is associated with productivity.
Education: formal schooling, job training, work experience.
Each year of school tends to raise one's wage by 5-7%.
Effect of education on productivity: educated workers are like an extra quantity of less educated workers. Example:
A worker with one year of college is worth 1.06 standard workers.
So, an increase in the average education level of the workforce by one year increases the effective labor force by 6%:
Y = A K1/3 N2/3 ,
This leads to a 4% increase in output and measured productivity [1.04 = 1.062/3].
Concern that the quality of US education has deteriorated.
Role of R&D
Basic Research versus Applied Research. US stronger in first.
Reduction in R&D expenditures in US compared to Japan.
Advanced technology might fail to translate in business success.
1. Savings and investment
2. Tax policy
5. Scientific research
6. Health care
7. Torts system
8. Anti-trust policies
9. Regulation policies