Long-term economic growth Total Factor Productivity and Technological Progress

Understanding the World Economy Master in Economics and Business Long-term economic growth Total Factor Productivity and Technological Progress Lecture 3 Nicolas Coeurdacier nicolas.coeurdacier@sciencespo.fr

Lecture 3: Long-term economic growth Total Factor Productivity and Technological Progress 1. Shifting the production frontier: TFP 2. Institutions and growth 3. Human capital, R&D and endogenous growth

Zambia - Just providing funds for investment doesn t work; need to consider intensive as well as extensive margin 20000 15000 GDP per capita if aid for investment for growth worked 10000 5000 Actual income 0 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994

Production function Output produced Buildings and machinery Labour input Technical knowledge and efficiency This lecture focuses on third input TFP

What is behind TFP? Two main aspects to TFP I) Efficiency A country may use its factors of production inefficiently and produce below the possibility frontier e.g, bureaucratic obstacles, bad policies, poor institutions, bad work practices etc. II) Technology A country may produce at the production possibility frontier but improvements in technology push the frontier out and enable more output to be produced for given factors of production

The double-dividend of TFP Higher TFP increases output for a given state capital stock Output per worker Higher = Capital per worker

The double-dividend of TFP Higher TFP also increases the steady-state capital stock Steadystate investment per worker = Higher = Capital per worker

The double-dividend of TFP Output per worker Higher = Capital per worker

TFP crucial in explaining cross country differences in GDP 1.6 1.4 TFP relative to United States 1.2 1.0 0.8 0.6 0.4 0.2 0.0 0.0 0.2 0.4 0.6 0.8 1.0 Output per worker relative to United States Source: Hall & Jones (1998), Why do some countries produce so much more output per worker than others?

Developmentaccounting Capital Education TFP Output per worker Canada 1.00 0.91 1.03 0.94 Italy 1.06 0.65 1.21 0.83 France 1.09 0.67 1.13 0.82 U.K. 0.89 0.81 1.01 0.73 Spain 1.02 0.61 1.11 0.68 Japan 1.12 0.80 0.66 0.59 Mexico 0.87 0.54 0.93 0.43 Korea, rep. 0.86 0.76 0.58 0.38 Netherlands 1.06 0.80 0.95 0.81 Iran 0.98 0.47 0.64 0.30 Chile 0.99 0.66 0.40 0.26 Peru 0.94 0.62 0.41 0.24 Egypt 0.45 0.58 0.72 0.19 Pakistan 0.58 0.39 0.57 0.13 Bangladesh 0.38 0.39 0.84 0.13 India 0.71 0.45 0.27 0.09 Lesotho 0.68 0.48 0.19 0.06 Kenya 0.75 0.46 0.17 0.06 Rwanda 0.44 0.34 0.29 0.04 Uganda 0.36 0.39 0.22 0.03 Human capital and capital stock important but explain only part of differences. Source: Hall & Jones (1998), Why do some countries produce so much more output per worker than others?

Why is TFP so different across countries? OneinterpretationofTFPistheleveloftechnology.Wewillgo through a model with research and development. A different interpretation of TFP is the level of effective technology, i.e. countries might have access to the same technology but the technology operates with different efficiencies in different countries. Why? Institutions?

Lecture 3: Long-term economic growth Total Factor Productivity and Technological Progress 1. Shifting the production frontier: TFP 2. Institutions, policies and growth 3. R&D, Human capital and endogenous growth

What is behind TFP? Two main aspects to TFP I) Efficiency A country may use its factors of production inefficiently and produce below the possibility frontier e.g, bureaucratic obstacles, bad policies, poor institutions, bad work practices etc. II) Technology A country may produce at the production possibility frontier but improvements in technology push the frontier out and enable more output to be produced for given factors of production

Digging deeper than the production function Endogenous Output Factor Endowments (Capital, Labor) TFP Partly Exogenous Trade/Policies Institutions Exogenous Geography Culture/History

What are institutions? Wide ranging concept rules of the game formal and informal constraints on political, economic and social interactions. Good institutions establish incentives to reduce uncertainty and encourage efficiency Particular organisational entities, procedural devices, regulatory frameworks Good institutions boost level of income and growth and help minimise impact of adverse shocks and reduce volatility

Measuring institutions Institutional Factors have many dimensions. Just as for firms the internal organisation of countries matter Legal system -Rule of law Protection of Property Rights Political Institutions Educational Institutions and Allocation of Talent Financial Institutions Regulatory Institutions Institutions for Macroeconomic Stabilization Institutions for Social Insurance

World Bank Governance Indicators Top 20 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 - NORWAY SWEDEN NETHERLANDS LUXEMBOURG DENMARK NEW ZEALAND FINLAND SWITZERLAND ICELAND CANADA IRELAND BELGIUM AUSTRALIA AUSTRIA ANDORRA GERMANY UNITED KINGDOM LIECHTENSTEIN PUERTO RICO ST. LUCIA Source: World Bank, 2009

World Bank Governance Indicators Bottom 20 CHAD Congo, Dem. Rep. IRAN ZIMBABWE BELARUS VIETNAM LAOS CHINA SAUDI ARABIA SYRIA SUDAN SOMALIA CUBA EQUATORIAL GUINEA LIBYA UZBEKISTAN TURKMENISTAN ERITREA KOREA, NORTH MYANMAR - -0.2-0.4-0.6-0.8-1.0-1.2-1.4-1.6-1.8-2.0-2.2-2.4 Source: World Bank, 2009

Institutions Better institutions lead to higher GDP per capita?

20

Institutions and growth Does better institutions increase output per capita or the other way around? Institutions are endogenous to the process of development. Correlation is not causality. How can we identify the sense of the causality? Find exogenous changes in institutions in a given country (or a set of countries) and then look at growth outcomes. Instrumental variables. Natural experiment. Acemoglu, Johnson, Robinson (2001).

Do institutions matter North and South Korea 14000 GDP per capita 12000 10000 8000 6000 South Korea North Korea 4000 2000 0 1950 1960 1970 1980 1990 1998

The colonial origins of comparative development Source: Acemoglu, Johnson and Robinson (2001)

The colonial origins of comparative development Source: Acemoglu, Johnson and Robinson (2001)

Institutions and rent seeking Economies flourish when institutions provide the right incentives Effort and investment in order to create value should be rewarded. Bad institutions encourage rent seeking, that is: do not reward value creation do reward value extraction Direct effect: waste of talent Indirect effect Rent seeking is like a tax Positive NPV projects may not be undertaken Misallocation of resources Dominated projects are selected Negative NPV projects are undertaken

Institutions and rent seeking Monarchy and Medieval Church Industrial Revolution happens after Renaissance and Reformation when individual rights get greater protection Kleptocracy Equatorial Guinea This is an extreme case, is there a systematic effect? Corruption Curseofrawmaterials Doing business hurdles

Source: Transparency International, 2014 Corruption Perception Index, 2014

Corruption Perceptions Index (score out of 100) Source: Transparency International, 2014

Bribe Payers Index 2011 Source: Transparency International, 2011

The Curse of Raw Materials Often find that countries rich in raw materials have disappointing growth Different channels Voracity effect raw materials subverts institutions and leads to rent seeking (including civil conflict) Dutch disease Due to high level of exports see sharp increase in real exchange rate which crowds out other sectors exports. Also revenues from raw materials boost government expenditure and raise price of non-tradable goods. Wastage and inefficiency low levels of TFP

The Curse of Raw Materials Countries with plentiful supply of raw materials growing at a slower pace? Sachs and Warner 2001

The Curse of Raw Materials Nigerian economy seen $350bn+ oil revenue but economic stagnation

Doing Business Starting a business can be an important area of growth (and often pro-poor orientated growth). The ease of doing business varies across countries. All countries requires firms to pass certain bureaucratic processes and often involves paying fees. Raises the possibility of using these bureaucratic processes to create a hold up problem with scope for corruption.

Economy World Bank Ease of Starting a Business Top 10 Worst 10 Ease of Doing Ease of Doing Business Rank Economy Business Rank Singapore 1 Niger 172 New Zealand 2 Eritrea 173 United States 3 Venezuela 174 Hong Kong, China 4 Chad 175 Sao Tome and Principe 176 Denmark 5 United Kingdom 6 Burundi 177 Ireland 7 Congo, Rep. 178 Canada 8 Australia 9 Norway 10 Guinea- Bissau 179 Central African Rep. 180 Congo, Dem. Rep. 181 In Australia it takes 2 days to start a business up and costs equivalent of 0.8% GDP per capita, in Brazil it takes 152 days and costs 10.4%. In Suriname it takes 694 days and in Sierra Leone it will cost you 11 times GDP per capita. Source: http://www.doingbusiness.org

Source: Djankov et al. (2002)

Regulation of entry and growth

Lecture 3: Long-term economic growth Total Factor Productivity and Technological Progress 1. Shifting the production frontier: TFP 2. Institutions and growth 3. R&D, human capital and endogenous growth

What is behind TFP? Two main aspects to TFP I) Efficiency A country may use its factors of production inefficiently and produce below the possibility frontier e.gbad work practises, bureaucratic obstacles, bad policies, poor institutions, etc. II) Technology A country may produce at the production possibility frontier but improvements in technology push the frontier out and enable more output to be produced for given factors of production.

Technology and growth Lecture 2: At the steady state, economy can no longer grow through capital accumulation alone. Technology is the growth miracle of capitalism. Output can always expand through technological progress. In Solow, technological progress comes from the sky. In reality, the outcome of research (R&D). Technological progress is not a free lunch. Need to think of what are the forces shaping incentives across time and countries to improve technology. 39

Research and Development (R&D) -if a country is far away from its steady-state, it can still grow through capital accumulation only. -ifacountryisclosetoitssteadystate,itcannotcontinuetorelyoncapital accumulation only. However, TFP increases can still sustain economic growth. Therefore as a country gets close to its steady state the incentives to encourage R&D rise strongly. WewouldexpectmoreR&Din (a) Rich countries (b) Economies that go through economic development. R&D less important for poor countries that can rely on capital accumulation and copy technologies of the developed countries. 40

R&D Expenditures

Expenditure on R&D as % of GDP 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 42 Sweden Korea Finland Japan Iceland United States Austria Denmark Germany OECD Total France Australia Belgium Canada EU25 United Netherlands Norway Czech Republic China Ireland Spain New Zealand Portugal Italy Hungary Turkey Greece Poland Slovak Mexico Source : OECD 2009 (MSTI)

Korean R&D as % of GDP % 4.0 3.5 3.0 2.98 3.23 3.47 R%D as % GDP 2.5 2.0 1.5 1.0 0.88 1.09 1.44 1.69 1.83 2.37 2.39 0.5 0.47 0.36 0.0 1967 1972 1977 1982 1984 1986 1990 1995 2000 2005 2006 2007 Source : OECD 2009 (MSTI) As Korea matures it invests more heavily in R&D 43

R&D in OECD R&D intensity varies across countries incentives to create new goods and ideas size of the market industrial structure competition education of workforce Public intervention Successful R&D increases TFP so differences in R&D may lead to technological differences 44

R&D and Convergence 4,0 3,0 R&D as % GDP 2,0 1,0 0,0 0 10000 20000 30000 40000 Rich countries more R&D intensive will create even further income differences - is a technological gap opening up? Source : OECD 2006 (MSTI) GDP per Capita, PPP

R&D and Endogenous growth Endogenous TFP growth( Solow) Abstract from capital for simplicity Production function for ideas. Production of new ideas proportional to the existing stock of ideas and to R&D (=number of researchers). No diminishing marginal product in the production of ideas. Based on Romer(1990). 46

Theory Production = Population = + = stock of knowledge (technology) =workers in manufacturing = =researchers = fraction of population working in R&D. Output per worker =/= 1 47

Theory Evolution of knowledge (technology) proportional to the number of researchers /A= /φ φ = cost of production of new ideas. Growth rate of output per worker /proportional to growth rate of knowledge: = = φ 48

Implications = = φ Endogenous TFP growth increasing with the resources in R&D (decreasing with R&D costs φ). Investment in R&D ( ) has ambiguous effects on output: - Decreases the number of workers producing output - Increases long-term TFP growth. Competition is not modelled but firms need to make profits to invest in R&D (pay the researchers). Need to protect innovating firms. 49

An increase in has ambiguous effects on output

Scale effects and international technology diffusion Scale effects: the growth rate depends on the size of the population, a larger L implies more workers engaged in R&D (for given A). Countries with larger populations should have higher growth rates, higher levels of technology and be richer. Very counterfactual! Another interpretation of the model is that a country's level of technology depends on R&D done around the world. Reasonable assumption if there is some international technology diffusion(keller(2002, 2004)). In this case, countries with large R&D sectors should be ahead but not grow permanently faster. PoorcountriesdonotneedtocarryoutR&Dandcan adopt the technology of the rich countries. 51

Technology spillovers in OECD (1970-1995) Averagedollar value of 1 USD of R&D spentin eachg-5 countries Keller (2002)

Summary Differences in TFP are a major explanation of differences in cross country income levels. Changes in TFP over time are due to changes in institutions & policies, and technological progress. Differences in institutions across countries can have very long-lasting effects on development. As countries get richer they shift their source of growth from capital accumulation towards technological progress. Endogenous growth models can explain persistent differences in income per capita across countries if technology cannot be easily transferred across countries.