Friday, 4 July 2008

The gains from trade for the poor

Christian Broda writes at on China and Wal-Mart: Champions of equality. He opens his article by noting that the public debate, especially in the USA, has taken for granted that inequality has risen as a result of globalisation. And then asks, "But has it really?" He argues that in fact it hasn't.

Broda points out that
How rich you are depends on two things: how much money you have and how much the goods you buy cost. If your income doubles but the prices of the goods you consume also double, then you are no better off. Unfortunately, the conventional wisdom on US inequality is based on official measures that only look at the first half, the income differential. National statistics ignore the fact that inflation affects people in different income groups unevenly because the rich and poor consume different baskets of goods.
Inflation affects the rich and poor to different degrees, and these inflation differentials change our view of the evolution of inequality in the US. Broda writes
Inflation of the richest 10 percent of American households has been 6 percentage points higher than that of the poorest 10 percent over the period 1994 – 2005. This means that real inequality in America, if you measure it correctly, has been roughly unchanged. And the reason is just as dramatic as the result. Why has inflation for the poor been lower than that for the rich? In large part it is because of China and Wal-Mart!
The important point here is that relative to wealthier families, the poorer families in the USA, and most other countries, spend a larger share of their income on goods whose prices are directly affected by trade – obvious examples being clothing and food. What we see is that the wealthier you are the more you spend on services. Such services are less subject to competition from abroad. From the data we see that since 1994 the price of goods in the US has risen much less than the price of services. Broda writes,
This trend can partly be explained by China. In U.S. stores, prices of consumer goods have fallen the most in sectors where Chinese presence has increased the most. Take canned seafood or cotton shirts, for instance. Exports of China to the rest of the world in these categories have increased dramatically over this decade. Inflation in these sectors has been negative over the last decade, while in other sectors with no Chinese presence inflation has been over 20 percent. Moreover, as China produces goods of relatively low quality, sectors with strong Chinese presence are disproportionately consumed by the poor.
Broda goes on to note that the effects of the expansion of superstores like Wal-Mart has been good for the poor. He explains,
The expansion of superstores – like Wal-Mart and Target – has also played an important role in accounting for the inflation differentials between rich and poor. Superstores sell the same products as traditional shops at much lower prices. Today the poor do roughly twice as much of their buying of non-durable goods in these stores than the rich. So poor consumers have been the biggest beneficiaries of Wal-Mart coming to town.
Broda ends his column by making the important point that,
We need to remind politicians and the public that the gains from trade are broadly shared. Every time the discussion over trade is diverted towards the problems facing specific producers, be they farmers in France or textile workers in the U.S., we miss the central point. Trading allows everyone, and especially the poor, to buy things that they could not otherwise afford. Without better public understanding of these facts, governments will not only keep supporting policies aimed against China and Wal-Mart but may receive the uninformed support of many consumers who are benefitting from trade. (Emphasis added)


Crampton said...

See DeLong's discussion here. He agrees with the big-picture conclusion but not the method.

Paul Walker said...

I can't help thinking, either the Broda and Romalis paper is wrong, in which case let DeLong show why; or its right in which case the problem is with DeLong.