The extent to which they accumulate productive assets.

Transcription

1 Technology Transfer Our analysis of the neoclassical growth model illustrated that growth theory predicts significant differences in per capita income across countries due to : The extent to which they accumulate productive assets. The level of technology that they operate. The Romer model provides with a theoretical framework for understanding how the level of technology is determined at a point in time. But, it is not very helpful for understanding differences across countries. 193

2 Two ways to view the Romer model: 1. As a model of the world as a whole, ignoring any barriers to factor flows across countries. In this case it is a story of where labour augmenting technical change comes from. 2. As a model of a country which is a technological leader. Recall that for the U.S., Canada, Western Europe, Japan, Australia, New Zealand, etc. there isn t much to explain with regard to differences in technology levels. For most of the world, however, total factor productivity growth has little to do with domestic innovation. 194

3 For these countries, technological progress consists mainly of the application of technologies developed in a few technologically advanced countries. We can address this issue of technology transfer by combining the Romer model with the neoclassical growth model of Mankiw, Romer, and Weil: All countries produce a homogeneous final output good, Y : Y = L 1 α h 0 x α j dj (169) L: total labour input (no sectoral division) 0 j h: the range of intermediate goods produced is determined by the country s skill level, not the world-wide technology level, A. 195

4 The country has no research sector and does not produce ideas. All labour is employed in production: L Y = L Firms in this country can convert capital into any intermediate good that they have the skills to create: h(t) 0 x j d j = K. (170) As in the Romer model, the technology for producing the final good makes it optimal to use all intermediate goods in equal quantities: x j = x for all j. (171) 196

5 So, we have: Y = L 1 α h(t) 0 = L 1 α hx α x α j d j = L 1 α h 1 α (hx) α = K α (hl) 1 α (172) This is the usual form of technology. The only difference from before is that the level of technology is given by the skill level, h(t), rather than by the stock of knowledge, A(t). 197

6 The accumulation of physical capital is the same as in all the models we have studied so far: K = s K Y dk The accumulation of human capital is a hybrid of the Romer and Mankiw-Romer-Weil models: ḣ = µe ψu A γ h 1 γ 0 < γ < 1. (173) i. As before, u is time devoted to schooling and ψ is the effectiveness of schooling. ii. A is the world-wide stock of knowledge. iii. h is the current skill level in this country. 198

7 The equation above decomposes the change in the skill level into two parts: µe ψu The effort devoted to acquiring skills through training and the effectiveness of this effort (µ, ψ). A γ h 1 γ A weighted average of the world technology level and the country s current skill level. We can re-write this equation: ḣ h = µeψu [ A h ] γ where h A. (174) 199

8 Note that for given µ, ψ, and u as the skill level, h, approaches the technology frontier, A, the growth rate of the skill level falls. This feature of the model is intended to capture the idea that it is relatively hard to acquire the skills to use technologies that are close to the frontier. We assume that the technology frontier grows exogenously at rate A A = g We can interpret g as the rate of growth in a country (or small group of countries) which behave as in the Romer model. These are the technological leaders. 200

9 We assume (for now) that u is constant and exogenously given. (This is just as in our analysis of the M-R-W model) Later, when we study endogenous growth, we will make u a choice variable. Also assume (as always) that the population grows at rate n. Along a balanced growth path, the growth rate of skills, ḣ, must be constant. h (By now, this should be apparent to you. Given the production function, h is labour-augmenting technological change for this economy.) 201

10 Since, ḣ h = µeψu [ ] γ A(t), (175) h(t) ḣ h constant A(t) h(t) constant (176) This implies, ḣ h = A A = g. (177) So, along a balanced growth path this country acquires skills at a rate equal the advance of the world technology frontier. 202

11 The problem here, then, is to determine this country s level of skills relative to the world technology frontier: h/a since ḣ h = g, we have ] γ g = µe ψu [ A h [ ] γ h = µ g A eψu h A = [ ] 1 µ γ g eψu (178) 203

12 1. The higher µ or ψ, the faster the rate of skill acquisition given the level of effort, u. This leads to a higher level of skills relative to the world technology frontier along the balanced growth path. 2. The higher u, ceteris paribus, the higher the rate of skill acquisition (due to increased effort). 3. The higher the growth rate of the world technology frontier, g, the further the country will be behind the boundary technology even though skills are accumulated rapidly. 4. The higher the parameter γ, the smaller the contribution of current skills to ḣ and so the lower will be h/a. 204

13 To derive the balanced growth path, note first that output per unit of effective labour, ỹ is: ỹ = Y hl = y h. As usual, the production function may be written in per capita or per unit of effective labour units as follows: Y = K α (hl) 1 α (179) y = Y L = ( K L ) α h 1 α = k α h 1 α (180) ỹ = y h = k α (181) 205

14 The capital accumulation equation is: K = s K Y dk k k = s K y k (d + n + g) Combining and solving, we get the usual: k = ỹ = [ [ ] 1 1 α s K n + g + d ] α 1 α s K n + g + d 206

15 Now solving for per capita income, y = ỹh: y(t) = [ s K n + g + d ] α 1 α h(t) (182) = [ ] α [ ] 1 1 α s K µ γ n + g + d g eψu A(t) }{{}}{{}}{{} I II III (183) The level of per capita income at a point in time is the product of three terms: I. Physical capital accumulation: The same as in the neoclassical growth model 207

16 II. Skill acquisitions III. World-wide technological change (determined by R&D of the technological leader(s)) This model accounts for differences in the state of technology (which we considered in our study of the M-R-W model) by attributing them to differences in the levels of acquired skills: Countries that: spend little time accumulating skills started from a low level of skills will be well below the world technology frontier. 208

17 Technology transfer. h/h A/h

18 The process of technology transfer. We have assumed that anyone can learn a technology by devoting time (u). In fact, however, there are various barriers to technology transfer that complicate matters: 1. Explicit protections: Patents. Some technologies are embodied in goods (medicines, etc.) Limits on sensitive technologies having to do with military applications, etc. 210

19 2. Trade regulations: These may have multiple effects... i. These can keep out new goods and services. The absence of these products from the economy may impede the development of the skills needed to produce them. ii. By limiting export possibilities, they can lessen the incentive to acquire certain skills. 211

20 These institutions can constitute barriers to all sorts of investment decisions, broadly defined: savings schooling innovation technology adoption and transfer Some of these institutions are also, however, somewhat paradoxically advanced as aiding growth and development. On the one hand, they protect domestic industries from foreign competition. On the other (and this is where we focus) they negatively affect agents incentive to undertake investment, broadly defined. 212

21 The role of incentives and barriers to accumulation. There are two main categories of factors which can potentially account for differences in per capita income across countries: 1. The return to capital or investment more generally. 2. People s preferences. In economics we typically focus on the first, because we don t think of preferences as directly observable. As such, the hypothesis that some countries are just more industrious or patient than others is difficult to assess empirically. Moreover, this category of explanation stresses social or cultural factors, rather than economic ones. 213

22 The first category of explanation, in contrast, stresses the incentives for people to accumulate assets, whether they be capital or skills. In the Canadian economy, there are strong forces working to equalize the rates of return on various types of investment activity. Financial markets price assets so as to equate returns wages adjust so as to provide the incentives for people to acquire necessary skills. These forces are also at work internationally: international capital flows migration Sustained differences in the return to investment activities require distortions. 214