Introduction to economic growth (1) EKN 325 Manoel Bittencourt University of Pretoria M Bittencourt (University of Pretoria) EKN 325 1 / 32
Introduction In the last century the USA has experienced a tenfold increase in income. In addition, income in the USA and western Europe is fifty times higher than in sub-saharan Africa why are economic growth rates important? Do they make any difference in the long run in terms of welfare? If so, how? during the 1989 meetings of the American Economic Association, the Harvard economic-historian David Landes posed the question: Why are we so rich and they so poor?, http://www.jstor.org/stable/pdfplus/2006534.pdf M Bittencourt (University of Pretoria) EKN 325 2 / 32
Introduction Moving a little back in time, Adam Smith called his 1776 magnum opus An inquiry into the nature and causes of the Wealth of Nations and Thomas Malthus was so concerned about the future prospects for growth in the 19 th century (recall the Malthusian prediction) that Economics became known as the dismal science back to the 20 th century, modern economic growth theory begins with MIT s Robert Solow in the 1950s with the publication of his two papers on the subject (which earned him the Nobel Prize in 1987) M Bittencourt (University of Pretoria) EKN 325 3 / 32
Introduction The papers by Solow highlight the role of physical capital accumulation, and more importantly, the role of technological progress as the engine of sustained economic growth. However, in these papers technology is exogenous, or a residual in the 1960s a group of economists, including (among others) Nobel Prize winners Kenneth Arrow and Edmund Phelps, tried to extend and improve the Solow model with their AK models in the 1980s economists at Chicago, Paul Romer and Robert Lucas (a Nobel prize winner himself), taking advantage of the new mathematical and economic techniques then available, proposed models which were based on the 1960s AK models and these models of endogenous growth incorporated the role of ideas and human capital (technology was/is at the forefront of economic growth) M Bittencourt (University of Pretoria) EKN 325 4 / 32
Introduction In the 1990s, with the availability of new data sets, economists started testing these models (Robert Barro from Harvard started the cross-country studies which led to an explosion of empirical studies about the determinants of growth) up to this point in time (2017), economic growth is probably the hottest area in macroeconomics, with better data sets and improvements in econometrics, not to mention better theoretical models, more and more researchers are investigating the causes of economic growth and prosperity worldwide moreover, most of those involved in this research project have already been either awarded with the Nobel Prize (Solow, Arrow, Phelps, Lucas), or will be in the near future (Romer and Barro), which illustrates the importance of the topic in current macroeconomics M Bittencourt (University of Pretoria) EKN 325 5 / 32
Introduction A word of caution is required though, growth and development are associated with policy implementation, however before implementing policies which are conducive to growth, we have to first understand what are the causes of growth and development so that better policies can be implemented M Bittencourt (University of Pretoria) EKN 325 6 / 32
Introduction Figure: Lights and development, or is it the other way round? M Bittencourt (University of Pretoria) EKN 325 7 / 32
The world is a heterogeneous place, there are countries which are rich, especially when compared to some less fortunate ones, others which are poor and others in the middle of the distribution (or in transition) more fundamentally, some countries are growing fast (China and India), others not growing so fast (South Africa) and some not growing at all (Somalia) so, we are going to initially check some data, or facts, on growth and development, or how the rich fare in terms of GDP per capita, or per worker, and how the poor and the not so poor fare as well M Bittencourt (University of Pretoria) EKN 325 8 / 32
Bear in mind that most of the data used here are freely available from the web, the Penn World Table is available from I http://cid.econ.ucdavis.edu/pwt.html, and Maddison s historical datasets are available from I http://www.ggdc.net/databases/index.htm, and the Gapminder, which provides a more dynamic view of growth, I http://www.gapminder.org/ and there are links to these pages from my website M Bittencourt (University of Pretoria) EKN 325 9 / 32
M Bittencourt (University of Pretoria) EKN 325 10 / 32
Fact 1, there is significant variation in per capita income across countries, or heterogeneity amongst countries. For instance, some of the poorest countries have incomes per capita which are only 5% of the income in the richest countries. The differences in living standards are staggering! table 1.1 reports some statistics, the first column reports the real GDP per capita (GDP/population) in 2008 US$ in different groups of countries; rich, poor, economic miracles and disasters M Bittencourt (University of Pretoria) EKN 325 11 / 32
An aside is the purchasing power parity (PPP) theory, or the adjusted exchange rate. The PPP is used to measure the ability of a particular currency to buy similar products elsewhere, so it takes into account the cost of living in particular countries. For instance, the Economist magazine makes available the Big-Mac index (if the Big Mac costs US$ 3.00 in the USA and 300 yen in Japan, then the PPP is 100 yen per dollar bear in mind that the Big Mac is a homogeneous product world wide, so such comparisons are possible). Extending the sample to a number of products (or a basket of products) we can construct a PPP exchange rate that can be used for international comparisons, which is much less volatile than the current exchange rate M Bittencourt (University of Pretoria) EKN 325 12 / 32
Table 1.1 is interesting for a number of reasons, the second column provides information on GDP per worker, or (GDP/labourforce). The third column provides information on labour force participation, the ratio of the labour force to population (labourforce/population), or effort in producing output but then the question is, which measure should we use for welfare comparisons amongst different countries? It is now widely accepted that the GDP per capita is a reasonable indicator of economic welfare. However, also bear in mind that the GDP per worker can be a reasonable indicator of productivity or of how productive the labour force is in a particular country M Bittencourt (University of Pretoria) EKN 325 13 / 32
Now let s look at what the numbers are telling us; India and Nigeria have GDPs which are less than 10% of the one in the USA. Additionally, the USA GDP is more than 60 times higher than the GDP in Ethiopia in other words, a worker in Malawi has to work for two months to earn what a similar worker earns in a day in the USA. Moreover, approximately 40% of the GDP is spent on food in Ethiopia, in the US only 7% of the GDP is spent on food on the top of that, figure 1.1 shows that in 2008 two-thirds of the world s population lived in regions with less than 20% of the USA GDP per worker. China and India are pushing this up a bit, given the size of their populations, however this is still significant M Bittencourt (University of Pretoria) EKN 325 14 / 32
M Bittencourt (University of Pretoria) EKN 325 15 / 32
However, figure 1.2 also shows how this distribution has(is) changing since the 1960s, and the distribution has, in fact, equalised. The share of the world population living in countries in which the GDP per worker is less than 10% of that in the USA has decreased over time. Again, China and India are playing an important role, since they are the countries which have seen substantial increases in their GDPs per worker (even relative to the USA) M Bittencourt (University of Pretoria) EKN 325 16 / 32
M Bittencourt (University of Pretoria) EKN 325 17 / 32
Another important information contained in table 1.1 relates to growth miracles, or the newly industrialising countries (NICs) such as Hong Kong, Singapore, Taiwan and South Korea. These countries have been growing consistently fast and that has had an effect on GDP per capita and per worker, or on the amount of output available in those economies (needless to mention China and India as well, with their recent high growth rates) Fact 2, the rates of economic growth display enormous differences amongst different countries. Table 1.1 provides information about growth rates across countries (the average of annual changes in the natural log of GDP per capita) from 1960 to 2008 the NICs (growth miracles) display rates of growth which are higher than the ones in more mature countries, except Japan. However, some countries also display negative growth rates (growth disasters), in particular sub-sahara African countries (but not only) M Bittencourt (University of Pretoria) EKN 325 18 / 32
An aside relates to how to calculate growth rates. An easy way is simply ġ = y 2005 y 2004 y 2004, or alternatively y t+1 y t y t = ġ. An useful manipulation from the above delivers y t+1 = y t (1 + ġ), this can determine the value of per capita income in y t+1 if we know y t and ġ moreover, an useful interpretation and application of the growth rates was proposed by Lucas, a country growing at ġ percent per year doubles its per capita income every 70/g years. For instance, the US doubles its GDP per worker approximately every 43 years [70/1.6]! That means that in a matter of less than 2 generations a typical American will be 2 or 3 times as rich as his grandparents for example, if y t grows at 2% per year, then y doubles every 70/2 = 35 years M Bittencourt (University of Pretoria) EKN 325 19 / 32
All in all, growth rates matter, in a short period of time different growth rates seen in different countries make significant differences in per capita income and therefore in the living standards of the population Fact 3, growth rates are not constant over time. In fact, growth rates were virtually zero for most of documented human history (the Malthusian stagnation), growth is indeed a new development in the history of humankind, a matter of 200 years or so the 20 th century has seen incredible growth rates, even when we take into consideration modest growth rates of 2% per year. Historically, humankind has lived in a stationary (Malthusian stagnation) period until approximately 1790 when modern sustained growth begins M Bittencourt (University of Pretoria) EKN 325 20 / 32
M Bittencourt (University of Pretoria) EKN 325 21 / 32
More specifically, between 1950 and 2008 world per capita GDP grew at 2.6% per year. Between 1870 and 1950 world per capita GDP grew at 1.10% per year. However, before 1850 world per capita GDP grew a mere 0.2% per year! I an aside is, how do we calculate the growth rates in the USA between 1870 and 2004? By simply using ġ = yt 1/t y 0 1, and substituting 37,500 2,500 1/135 1 = 0.02, or 2% over the period Angus Maddison suggests that between 500 CE and 1500, growth rates averaged zero! Again, growth is a modern phenomenon, just like electricity M Bittencourt (University of Pretoria) EKN 325 22 / 32
All in all, because of this recent growth, the world per capita GDP today is 15 times higher than in 1500. Again, the differences in living standards are staggering! growth within countries is also of interest. For instance, Mauritius presented negative growth rates between 1950 and 1970. However, between 1970 and 2008 Mauritius grew at 3.1% per year. China is another example, during 1960 to 1978 the country grew at 2.1% per year. However, since 1979 China s GDP per worker has grown, on average, at 7.7% per year the above leads to another fact of economic growth, M Bittencourt (University of Pretoria) EKN 325 23 / 32
Fact 4, the position of particular countries in the distribution of per capita income is not immutable, countries can move from a low position in the distribution to a high one, and vice versa think of Argentina, one of the richest countries just a matter of 100 years ago, and now it has only about one-third of the USA per capita income, or Finland, a poor country 100 years ago... so, mobility is possible M Bittencourt (University of Pretoria) EKN 325 24 / 32
Other Stylised Facts The 4 facts seen above apply to the countries of the world in general. The next fact applies mostly to the USA, however it is important to understand what happens in the USA since they are a country which went through different stages of development, therefore useful. Moreover, some of the USA features are important for most economies in the long run Fact 5, in 20 th -century USA; firstly, the real rate of return to capital r has presented no trend whatsoever; secondly, the shares of income devoted to capital and labour have presented no trend either; thirdly, the average growth rate of output per capita has been positive and constant over the period (these are the Kaldor facts) the first statement above is simply telling us that the real interest rate in the USA has no trends (upward or downward) I recall that to get the real interest rate you have to discount the effect of inflation from it (r = i π) M Bittencourt (University of Pretoria) EKN 325 25 / 32
Other Stylised Facts The second statement is about the factors of production, capital K and labour L. With information on salaries, wages and self-employment remuneration we can get the ratio (labourshare/gdp). This labour share has been constant over time, around 0.7 needless to say that if the labour share is constant, the capital share must be constant over time too, around 0.3, recall the simple Cobb-Douglas production function (Y = K, L) the third statement is better illustrated by figure 1.4, which shows the growth of GDP per capita in the USA between 1870 and 2008. The long-run trend is positive all over the period, so the question is, what is causing growth then? M Bittencourt (University of Pretoria) EKN 325 26 / 32
Other Stylised Facts M Bittencourt (University of Pretoria) EKN 325 27 / 32
Other Stylised Facts Fact 6, growth in GDP and growth in the volume of international trade, or openness for short, are positively correlated in a cross-section of countries. The volume of trade is defined as the sum of exports and imports over GDP (x + m/gdp) M Bittencourt (University of Pretoria) EKN 325 28 / 32
Other Stylised Facts M Bittencourt (University of Pretoria) EKN 325 29 / 32
Other Stylised Facts Fact 7, skilled and unskilled workers migrate from poor to rich countries. That implies that wages are higher in rich countries, so the migration. However, the question is does it make any economic sense? If skilled labour is scarce in developing and poor countries (which also implies higher wages in those countries), why then highly-skilled Africans keep migrating to the UK, Australia, Holland and USA in large numbers? M Bittencourt (University of Pretoria) EKN 325 30 / 32
The road ahead Why are we so rich and they so poor? what is the engine of economic growth? If technological progress is the missing link, then we should also ask what determines technological progress? how can we account for growth miracles? is modern growth sustainable? How about the role of natural resources? M Bittencourt (University of Pretoria) EKN 325 31 / 32
The road ahead Understanding what causes economic growth is vital, the NICs are examples that (upward) mobility is possible. Moreover, Argentina, one of the richest countries in the world a mere 100 years ago, is an example of (downward) mobility. Economic growth has the potential of increasing living standards in a matter of a generation. Lucas, in his Marshall Lecture at Cambridge in 1988, stated: I do not see how one can look at figures like these without seeing them as representing possibilities.... The consequences for human welfare involved in questions like these are simply staggering: Once one starts to think about them, it is hard to think about anything else. M Bittencourt (University of Pretoria) EKN 325 32 / 32