Geography, History and Institutions October 2007 () Institutions October 2007 1 / 22
Background The evidence is roughly consistent with Solow model in that once countries achieve , ! high rates of technology adoption , ! high rates of investment in physical/human capital , ! contain population growth they can catch up (conditional convergence) BUT what determines these “proximate determinants?” () Institutions October 2007 2 / 22
Figure 2 Years for Per Capita Income to Grow From 2,000 to 4,000 (1990 U.S. $) 80 IRE N.Z. NETH AUS 60 BEL FRA U.K. Number of Years DEN CHIL JPN U.S. SWI S.A. 40 CZE AUSTRA FIN COL CAN NOR PERU TUR 20 PORT JOR BUL P.R. TAIW 0 1820 1860 1900 1940 1980 Year Income Reached 2000 (1990 U.S. $)
“Why do Some Countries Produce So Much More Output per Worker than Others ?” Hall and Jones (1999) Social infrastructure, institutions, geography + Investment, employment, technology adoption + Output per worker () Institutions October 2007 3 / 22
Levels Accounting Aggregate production function for country i : i ( A i H i ) 1 � α , Y i = K α Can be expressed as � K i � α Y i 1 � α = h i A i , L i Y i where h i = H i . L i () Institutions October 2007 4 / 22
Index of Human Capital Wage per unit of human capital is v i = ( 1 � α ) Y i H i , ! wage of individual j in country i is w ij = v i h ij In logs: log w ij = log v i + log h ij “Mincerian” wage regressions log w ij = a i + b i s ij + c i X ij + ε ij , () Institutions October 2007 5 / 22
Hall and Jones use index of average human capital: h i = e b i E i where E i = average schooling. b i = 0 . 13 for average schooling < 4 years = 0 . 10 4 – 8 years = 0 . 07 over 8 years () Institutions October 2007 6 / 22
Table 1: Productivity Calculations: Ratios to U.S. Values ——Contribution from—— ( K/Y ) α/ (1 − α ) Country Y/L H/L A United States 1.000 1.000 1.000 1.000 Canada 0.941 1.002 0.908 1.034 Italy 0.834 1.063 0.650 1.207 West Germany 0.818 1.118 0.802 0.912 France 0.818 1.091 0.666 1.126 United Kingdom 0.727 0.891 0.808 1.011 Hong Kong 0.608 0.741 0.735 1.115 Singapore 0.606 1.031 0.545 1.078 Japan 0.587 1.119 0.797 0.658 Mexico 0.433 0.868 0.538 0.926 Argentina 0.418 0.953 0.676 0.648 U.S.S.R. 0.417 1.231 0.724 0.468 India 0.086 0.709 0.454 0.267 China 0.060 0.891 0.632 0.106 Kenya 0.056 0.747 0.457 0.165 Zaire 0.033 0.499 0.408 0.160 Average, 127 Countries: 0.296 0.853 0.565 0.516 Standard Deviation: 0.268 0.234 0.168 0.325 Correlation w/ Y/L (logs) 1.000 0.624 0.798 0.889 Correlation w/ A (logs) 0.889 0.248 0.522 1.000 Note: The elements of this table are the empirical counterparts to the components of equation (3), all measured as ratios to the U.S. val- ues. That is, the first column of data is the product of the other three columns.
Main …ndings: Strong positive correlation between output per worker and TFP Strong positive correlation between average human capital and TFP Most of the di¤erence between developed and less developed countries is due to TFP di¤erences () Institutions October 2007 7 / 22
Figure 1: Productivity and Output per Worker PRI 1.50 ITA HKG FRA ESP LUX Harrod−Neutral Productivity, 1988, U.S.=1.00 SGP SYR SAU CAN YEM JOR GBR USA AUTBEL ISL NLD MEX DEU SWE CHE AUS BGD VEN ISR TTO MUS DZA OAN .75 BRA PRT BRB GTM MLT FIN EGY IRL DNK NOR TUN GRC OMN JPN SYC SUR ARG CYP IRN NZL COL SLE URY KOR PAK MAR REU CRI SLV TUR DOM SUN ZAF MYS SWZ FJI .40 PER CHL COG LKA ECU PRY THA SEN HND PAN YUG MOZ NAM BOL BEN CIV GAB HUN RWA NIC CMR GMB IND MDG HTI IDN CSK POL SDN JAM UGA PHL GHA PNG TCD .20 GIN MLI LSO ROM SOM BWA ZWE KEN ZAR CPV CAF BDI GUY AGO GNB NGA NER MRT Coeff = 0.600 TZA TGO BUR CHN MWI .10 BFA StdErr= 0.028 COM ZMB R2 = 0.79 1000 2000 4000 8000 16000 32000 Output per Worker, 1988 (in 1985 U.S. Dollars)
Determinants of Economic Performance Output per worker and inputs strongly correlated with: , ! social infrastructure — GADP Index , ! openness to trade — Sachs–Warner index Direction of causation ? , ! instrumental variables () Institutions October 2007 8 / 22
Figure 2: Social Infrastructure and Output Per Worker USA LUX CAN 32000 CHE AUS ITA BEL FRA DEU NLD SWE NOR GBR FIN ISL PRI NZL AUT ESP DNK ISR HKG SGP JPN IRL SAU TTO VEN GRC 16000 MLT OAN CYP SYR MEX ARG OMN SUN JOR BRB KOR Y/L (U.S. dollars, log scale) PRT URY DZA BRA HUN YUG IRN FJI COL CHL MYS CRI MUS SUR ZAF POL PER ECU 8000 REU PAN TUN TUR SYC YEM GTM CSK DOM NAM EGY MAR PRY SWZ GAB SLV LKA THA BOL PAK HND JAM BGD PHL NIC COG 4000 ROM IDN GUY CIV BWA IND PNG CPV CMR SEN SDN ZWE SLE LSO CHN BEN 2000 HTI KEN GHA MRT ZMB SOM NGA GMB RWA GIN MDG TGO GNB MOZ COM MLI CAF AGO ZAR UGA TZA TCD MWI BDI BFA 1000 BUR NER 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Observed Index of Social Infrastructure
Instrumental Variables (2SLS) Methodology Hypothesized structural model: log y i = α + β S i + ε i S i = γ + δ log y i + θ X i + η i , where = y i dependent variable (e.g. wages) S i = key explanatory variable (e.g. health) X i = vector of exogenous instrumental variables Reduced form for S i : S i = γ + δα + θ X i + δε i + η i 1 � δβ () Institutions October 2007 9 / 22
If X i is uncorrelated with ε i and η i then we can estimate the “…rst stage regression” S i = a + bX i + u i using OLS where a = γ + δα θ 1 � δβ and b = 1 � δβ Then run “second-stage regression” log y i = α + β ˆ S i + ε i using the …tted value ˆ a + ˆ S i = ˆ bX i Estimate of β should re‡ect impact of variations in S i that are due to exogenous variation in X 0 i s only () Institutions October 2007 10 / 22
Three key requirements of "good instruments": ! R 2 in …rst stage regression must be reasonably high , , ! must clearly be an exogenous determinant of S i , ! no other channels through which X i e¤ects y i (over identifying restriction) () Institutions October 2007 11 / 22
Hall and Jones use the following instruments: , ! distance from equator , ! language indices , ! predicted trade share ) results suggest that social infrastructure and openness are signi…cant factors BUT (1) aside form distance to the equator other instruments are not clearly exogenous (2) why does distance to the equator matter ? () Institutions October 2007 12 / 22
Table 2: Basic Results for Output per Worker log Y/L = α + β ˜ S + ˜ ǫ OverID Test Coeff Test Social p -value p -value Specification Infrastructure Test Result Test Result ˆ σ ˜ ǫ 1. Main Specification 5.142 .256 .812 .840 (.508) Accept Accept Alternative Specifications to Check Robustness 2. Instruments: 4.998 .208 .155 .821 Distance, Frankel-Romer (.567) Accept Accept 3. No Imputed Data 5.323 .243 .905 .889 79 Countries (.607) Accept Accept 4. OLS 3.289 — .002 .700 (.212) Reject The coefficient on Social Infrastructure reflects the change in log output per worker as- sociated with a one unit increase in measured social infrastructure. For example, the coefficient of 5.14 means than a difference of .01 in our measure of social infrastructure is associated with a 5.14 percent difference in output per worker. Standard errors are computed using a bootstrap method, as described in the text. The “Main Specification” uses distance from the equator, the Frankel-Romer instrument, the fraction of the popu- lation speaking English at birth, and the fraction of the population speaking a Western European language at birth as instruments. The “OverID Test” column reports the result of testing the overidentifying restrictions and the “Coeff Test” reports the result of testing for the equality of the coefficients on the GADP policy index variable and the openness variable. The standard deviation of log Y/L is 1.078.
Table 4: Results for log K/Y , log H/L , and log A Component = α + β ˜ S + ˜ ǫ —— Dependent Variable —— α 1 − α log K/Y log H/L log A Social 1.052 1.343 2.746 Infrastructure (.164) (.171) (.336) OverID Test ( p ) .784 .034 .151 Test Result Accept Reject Accept ˆ σ ˜ .310 .243 .596 ǫ ˆ σ DepV ar .320 .290 .727 Note: Estimation is carried out as in the main specifi- cation in Table 2. Standard errors are computed using a bootstrap method, as described in the text.
“The Colonial Origins of Comparative Development: An Empirical Investigation” Acemoglu, Johnson and Robinson (2001). Basic Idea: To capture the true impact of institutional di¤erences on economic performance, we need an exogenous source of variation in institutions Early determinant of current institutions in many countries was the nature of early colonization after 1500 Two broad kinds: (1) extractive (e.g. the Belgian Congo) (2) “neo–Europes” (e.g. Australia) () Institutions October 2007 13 / 22
BUT what determined nature of colonization ? , ! settler mortality rates ? Sierra Leone (1793), Niger expedition (1805) Pilgrim fathers: US vs. Guyana convicts: Australia vs. Gambia () Institutions October 2007 14 / 22
Recommend
More recommend
