Feenstra, Chapter 5:

Movement of labour between countries: Immigration

Immigration is widely believed to reduce native workers’ wages but sometimes it doesn’t:

From May to September 1980, boatloads of refugees from Cuba arrived in Miami (125,000), resulting in an increase of 7% of the local population. But less-skilled natives’ wages were not affected

Similarly, in 1989, 670,000 high-skilled Russian Jews arrived to Israel (11% of the Israeli population), and during the 90s, the wages of the high-skilled increased

Large inflows of workers need not depress wages in the areas which they settle. In other large scale migration episodes (from Europe to America), natives’ wages indeed fell.

Effects of immigration in the short run: specific factors model

Labour is mobile among home industries but land and capital are fixed. The equilibrium wage is determined by the amount of labour in both of the industries, prices and marginal products. Wages are equal in the 2 sectors so there is no reason for labour to move between industries and Home labour market is in equilibrium. We assume that W* (foreign wage) < W (home wage) so there is an incentive for immigration. Home’s labour endowment increases by ∆L>0 so we expand the horizontal axis by ∆L (see diagram). Since PAMPLA is graphed relative to 0A, it also moves right by ∆L. There is no shift in PMMPLM because 0M has not changed. The new equilibrium home wage is lower at B – the number of workers in both industries has increase, diminishing marginal product of labour means that the wages in both industries declines.

Between 1870 and 1913, 30 million Europeans left the “Old World” to emigrate to the “New World,” because of higher wages, raising the US population by 17%. Real wages in both grew over time as capital↑ so MPL↑, but due to large scale immigration, wages in New World grew more slowly (compare blue dotted line to blue solid line). Large-scale migration contributed to a convergence of real

wages across the continents. Now US welcome foreign workers in agriculture and EU welcomes workers in high-tech industries. They do this even though those foreign workers compete with domestic workers in those industries, lowering wages.

There are 2 ways to compute rentals (wages and rents):

1. Earnings left over in the industry after paying labour 2. Marginal product of capital or land times the price of the good produced in each industry.

Under either method the owners of capital and land benefits from the reduction in wages due to immigration:

Therefore, the restriction on immigration should be seen as a compromise between:

Entrepreneurs and land-owners who might welcome the foreign labour Local unions and workers who view migrants as a potential source of competition leading to

lower wages The immigrant groups themselves who might also have the ability to influence the political

outcome of immigration policy if the group is large enough.

With more workers in both industries and the same amount of capital, the output of both industries rises. Immigration leads to an outward shift in the PPF. Given constant prices, output rises from point A to B – however this only holds in the SR when capital and land are fixed.

Effects of immigration in the long run:

In the long run, all factors are free to move between industries. We ignore land to simplify. This is like Heckscher-Ohlin except now allow labour to move between countries. R and W same in both industries. Assume that the more labour per machine in agriculture than in manufacturing (agriculture is labour intensive: LA/KA > LM/KM ie. labour capital ratio is higher in agriculture). At all w/r, K/L is lower in agriculture. Prices are determined internationally (free trade). We can use an Edgeworth Box diagram to analyze what happens in this model. The lines show the amount of labour and capital used in each industry

(K/L ratios = slopes). When the line is flatter, the capital-labour ratio is lower (there are fewer units of capital per worker in agriculture).

Factor prices (real wages and real rentals) are determined by MPL and MPK, which depend on K/L. If there is higher K/L then by law of diminishing returns, MPK and real rental must be lower, MPL and real wage must be higher because each worker is more productive. Each point along

the lines corresponds to a particular real wage and rental.

Because of immigration, total labour supply increases by ∆L so the labour axis expands. Since capital is able to move between industries, industry outputs adjust to keep K/L constant in each industry, so instead of allocating the extra labour to both industries, all of the extra labour will be allocated to

agriculture (the labour intensive industry). Some capital is also moved from manufacturing to agriculture to keep K/L constant in agriculture due to more labour, but also some labour will move from manufacturing to agriculture to keep K/L constant in manufacturing. The capital-labour ratio in each industry is unchanged (slope of lines unchanged) and the additional labour in the economy is fully employed. Since the capital-labour ratios K/L remain unchanged, so do the MPK and MPL. Hence: When capital can move freely between industries, immigration in the long run has no impact on the wage and rental rates. (In the SR, immigration depressed wages and raised rental on capital).

Rybczynski theorem: An increase in the amount of labour will expand the output in agriculture and contract the output in manufactures. The PPF shifts out more in the direction of agriculture – this asymmetry illustrates the new labour is employed in agriculture and it pulls labour and capital out of manufacturing in the long run to establish a new equilibrium. In SR, both industries expand but in LR one industry expands and the other contracts. Factor prices don’t need to change because the economy can absorb the extra labour by increasing the output of the industry using that factor intensively. Factor prices do not change – factor price insensitivity.

Movement of capital between countries: Foreign Direct Investment (FDI)

FDI: When a firm from one country owns a company in another country. ex. U.S. Department of commerce uses a 10% (or more) rule to determine FDI, for either a foreign country (FDI inflow) – foreign company acquires >10% of a US company - or a U.S. company (FDI outflow). Two types:

Greenfield FDI — when a company builds a plant in a foreign country Acquisition FDI (or brownfield FDI) — when a firm buys an existing foreign plant

Traditional view of FDI: capital moves from high-wage to low-wage countries to earn a higher rental.

Effect of FDI in the SR: specific factors model

Wages: As capital flows into the economy, it will be used in manufacturing in this model. Additional capital will ↑MPLM because workers have more machines to work with, so PMMPLM↑. As a result of this shift, W↑ and more workers are drawn into manufacturing (LM↑ and LA↓).

Output: Since LA↓ but amount of land the same, output↓ in agriculture. LM↑ and K↑ so output↑ in manufacturing. At constant goods prices (relative price lines have same slope) the equilibrium output shifts from A to B.

Rentals: Since LA↓ each acre of land can’t be cultivated as intensively as before so MPTA↓. Since RT=PA x MPTA and there is no change in the price of agricultural goods, the rental on land falls. In manufacturing, we know rentals can also be measured as revenue earned in

manufacturing minus the payments to labour. If wages are higher and everything else is the same then there must be a reduced amount of funds left over as earnings of capital so rental is lower.

In summary:

Effect of FDI in the LR: The H-O model

An inflow of FDI causes the amount of capital to increase which expands the edgeworth box upwards, and shifts the origin up. Now less labour and capital is used in agriculture, so output falls, and more is used in the other industry so output rises. Rybczynski explains why: an inflow of capital will expand the output of the capital-intensive industry (manufactures), and contract the output of the labour-intensive industry (agriculture).

This change in outputs is achieved with no change in the labour-capital ratios in either industry (the new lines have the same slope as before). Since labour can move freely between industries so as to keep the capital labour ratios constant, FPI (factor price insensitivity) implies no effect on either wages or rentals. Each person has the same amount of capital to work and each machine the same number of workers. MPL and MPK are unchanged, as are factor prices.

Outcome is same as with immigration in LR: in LR, an inflow of either factor of production will leave factor prices unchanged.

Gains from immigration:

Foreign investment and immigration are both controversial policy issues. Most countries have at some point controlled FDI but later became open to foreign investment. However, almost all countries impose limits on immigration: Why is immigration so controversial?

- Some groups oppose the spending of public funds on immigration- Other groups fear the competition for jobs created by an inflow of workers

We presume the immigrants are better off from higher wages in the country to which they move. In the host country, in SR, home workers face competition and receive lower wages whilst capital and land owners gain. On balance we find there are overall gains from immigration in the specific factors model (not inc immigrants wages). If we include their wages with Foreign income then we find emigration benefits the foreign country too.

We measure gains from immigration using specific factors model. In the diagram, home labour is measured from left to right. Home wages is downwards sloping because as immigrants enter, the wage falls. In the diagram: W>W’>W*. At start, since wages at home are higher than in foreign, workers migrate until wages are equalised. As foreign workers leave, it benefits those left behind by raising their wages. Point C is equilibrium with full migration (reached only in very long run).

Gains for home country: P.MPL=W. The first immigrant has MPL equal to W at point A. As more foreign workers migrate, MPL in both home industries falls due to diminishing returns. The immigrant’s MPL falls from W to W’ (from A to C). At C, all immigrants are paid W’ but all except the last immigrant had MPL>W’ (they started off with W) so therefore their contribution to the output of goods in the home economy exceeds the wage that they are paid (e.g. gain from first immigrant is W-W’ for home). Adding up these gains, we end up with the triangle ABC. The reason for these gains is diminishing returns: as immigrants↑ MPL↓ and since W=MPL of the last workers, it must be less than MPL of earlier immigrants.

Gains for foreign country: include remittances in our measure of foreign income. In absence of immigration, W=W*. As workers emigrate, MPL of remaining labour↑ so W*↑ to W’. Each of the higher MPLs in between equals the drop in foreign output from having workers leave. Under full migration all foreign immigrants earn W’ in home country which is higher than their foreign MPLs (between W* and W’). Gain to foreign = earnings of emigrants – foreign MPL (or drop in foreign output). Adding up these gains, we get the triangle A’BC.

World gains from immigration: A’AC (ie. the increase in world GDP due to immigration). E.g. first worker earns W* in foreign, and so when they leave, foreign output ↓ by W*, and home output↑ by W. The difference between home and foreign wages = net increase in world GDP.

The table shows workers’ remittances and net foreign aid in some developing countries in 2005. For all these countries, the income sent home by emigrants is a larger source of income than official aid.

Gains from immigration: (table below)

Borjas used a stock of immigrants = 10% of US workforce, assuming they compete for same jobs as skilled US workers, then ↑GDP = 0.1% but assuming they’re unskilled = 0.4%. Kremer and Watt looked at a specific type of immigrant: household workers (mainly female) whose presence allows the educated woman to seek employment in her home country.

Part B reports results from estimates of gains due to migration for several regions of the world.

- Walmsley and Winters: found triangle of gains (like A’AC) - Klein and Ventura: model differences in technology, so immigrants are more productive in

capital-rich home country. They look at increase in size of EU from 15 to 25 countries, and at NAFTA (where all initial countries are almost twice as productive before enlargement).

Gains from FDI: In the diagram, the rental earned on capital in each country is on vertical axis. Since A*>A, capital will flow from home to foreign. As it enters foreign, MPK↓ which bids down rentals, at home MPK↑ as capital leaves. Point B is equilibrium with full capital flows where rentals are equalised at R’.