China Research

One of the most durable relationships in international macroeconomics and trade is “Houthakker-Magee” (H-M) the observation that the elasticity of US imports with respect to US economic activity exceeds the elasticity of US exports with respect to foreign economic activity. The elasticity ratio of about 2:1 appeared in the post World War 2 data, when H-M first published their findings in 1968 using aggregate US trade data spanning 1951-1966.

The relatively higher income elasticity of imports has persisted into the 21^st century. Researchers have used different data disaggregations, different econometric methods, and different time periods to explore the source and stability of these findings. Mann and Pluck [1] found, in particular, that consumer demand for different kinds of products was an important source of the high elasticity of demand for imports. Overall, however essentially the asymmetry remains, and has had global ramifications.

First, to the extent that the US has persistently bought more imports than it sells exports, even when US and global growth are similar, the US trade deficit has tended to widen. Indeed the US trade deficit has been a feature of the global trading landscape for more than a generation, and over the 1990s, and 2000s in particular, it widened substantially to more than 6 percent of GDP. From the standpoint of global exporters, particularly China, the high US import elasticity yielded more robust export expansion than would have been the case if the H-M asymmetry did not exist. A complementary outcome of the H-M asymmetry has been international reserves build-up and the associated challenges of leaning against currency appreciation among our export partners, as well as the decision of how to invest those reserves.

It is therefore of some interest that, since the onset of the financial crisis and during the recovery in US GDP, H-M seems to have gone missing. The chart below shows the average annual growth rates for US real GDP and several categories of consumer goods imports. The green bars show the ratio of domestic growth to the growth of imports in the matched category for the historical period (1968-2010) and more recently 2011-2014q1. The decline in the ratio of import growth to domestic growth (the implied import elasticity) is apparent in each consumption category. One reason for the decline in implied import elasticity for consumer goods is the increasingly share of services in the consumptions basket, but that can’t be the whole story for the most recent period.

[1] “Understanding the US Trade Deficit: A Disaggregated Perspective,” (with Katharina Plück) in G-7 Current Account Sustainability and Adjustment, Richard Clarida ed.. MIT Press: Cambridge, 2007.Download the working paper version from www.CLMann.com.

What are the ramifications of this apparent change in the relationship between domestic growth and imports, particularly for consumer goods? Considering trade flows, the trade deficit may not widen very much as the US economy builds steam. Foreign exporters to the US that have been importantly dependent on the US market for growth in exports may find that sales to the US market will remain sluggish. Indeed, the table below shows that the implied elasticity of imports from China has been falling. In recent quarters (emphasized by using quarterly data at annual rate growth from previous period as in column 6) suggest that imports from China have collapsed.

If H-M has indeed gone missing, it has implications for other countries. Foreign exporters to the US that had expected to take over from China the production of relatively cheap and easy to produce consumer goods may not experience such a robust climate for their products as was the case over the 1990s and 2000s. In fact, projections for growth in global trade are muted, with the World Trade Organization projecting global trade growth for 2014 of 4.7%. Although the WTO projection for 2015 is at the 20-year average of 5.3%, global trade growth was substantially higher than that – around 6% — in the 1990s and 2000s before the bust. [1]

Considering the financial perspective, if foreign economies have less international reserves accumulation, because of smaller trade surpluses, this may imply less demand by foreign central banks for US Treasury securities (UST). During the past decades, the demand for UST by foreign investors, particularly China, allowed UST interest rates to be lower than they otherwise might have been. Now, with the Federal Reserve tapering their purchases of UST and potentially less demand from abroad, two sources of demand for UST will soften. UST interest rates could rise more quickly than expected.

Thus, it matters from both the real and well as financial perspectives and for the United States, China and the global economy, if Houthakker-Magee has gone missing.

[1]http://www.wto.org/english/news_e/pres14_e/pr721_e.htm

 

 

 

 

 

 

 

Catherine L. Mann
Rosenberg Professor of Global Finance, Brandeis University

 

Back to Start Page

Introduction and Testable Hypotheses
Globalization of information technology products (IT) relates to economic growth through three channels:

(1) Terms of trade. The fall in quality-adjusted prices of IT products favors use of IT in the domestic economy and yields TFP gains.

(2) Economies of scale. Total factor productivity is positively associated with the scale of production of IT products.

(3) Variety. Greater variety means that more domestic users find good matches between products and needs, which increases productivity and growth. Greater variety and uniqueness of exports is associated with increased prices and profits.
Figure 7.2 reproduced from the Chapter sets out these three factors in terms of an integrated hypothesis relating TFP and trade in IT products.

Empirial analysis using a sample of 36 countries for 2000-2007 finds:

(1) With regard to terms of trade, importers of IT gain relatively more than exporters, on average, from the declining prices of IT coming through international trade.

(2) With regard to economies of scale, despite falling IT prices, most exporters enjoy positive economy-wide benefits of trading in IT because of economies of scale in production.

(3) With regard to variety, countries with greater variety tend to experience greater TFP gains relative to those with less variety. Against this backdrop and empirical validation, where does China fit in?

 

China’s gains from IT trade
China’s participation in global trade in IT products exploded over the 1990s and 2000s. From 2% of global exports in 1990, China+ (China plus Hong Kong SAR) rose to 15% of global exports (ranked 1st) in 2004. On the import side, China+ moved from 9th ranked at 4% of global imports to top ranked with 20% of global imports. In terms of importance in global expenditure on IT, however, China+ both grew less and is less important. China+’s share of global expenditure, although it rose four times over and ranks sixth in global expenditure at 3%, remains quite small in comparison to the country’s importance in global trade.

Putting the data on trade prices, and production and expenditure into the framework reveals a pattern similar to what was hypothesized (Figure 7.4 from the Chapter replicated below). Increased imports of IT products (left panel) is associated with greater economic gains as measured by the social surplus concept, a concept that is isomorphic to TFP. This is consistent with other findings on the relationship between IT use and TFP in the domestic context.

For IT exporters (right panel), the relationship could be positive (if economies of scale gains outweigh terms of trade losses) or negative. Since the trend line cuts the y-axis at around 0.76%, it appears that a country does gain from being a producer and exporter; but the simple correlation coefficient of 0.016 suggests that there is not a strong relationship between the magnitude of exports and economic gain.

 

China, as an exporter, lies above the regression line, suggesting that it gains relatively more from its pattern of trade, production, and expenditure compared to other net exporters. The story for these relatively greater gains for China, despite being an exporter, pretains to the variety of China’s products in trade.

Variety can be measured in several ways; this paper uses the Herfindahl (H) index calculated across 178 different varieties of IT exports and imports allocated into broader product categories, including Computers, Components, and Other ICT (embedded ICT such as medical devices). Within a category, a Herfindahl close to 0 implies a high variety in product trade, and Herfindahl close to 1 implies highly concentrated product trade.
Figure 7.5 from the Chapter shows Herfindahl indexes for three countries and three product categories. Returning to Figure 7.4, Indonesia is an exporter with lower than average social surplus for exporters (lies below the trend line for exporters). It has a very high concentration of trade in Computer exports. With a highly concentrated export pattern, but lower than average social surplus, this suggests that economies of scale in production does not outweigh the terms of trade effect, and, with high concentration, there are no gains from variety.

 

China is also a exporter, but with higher than average social surplus (lies above the trend line in Figure 7.4) for exporters. Although China’s exports of Computers are somewhat concentrated, it has an even greater concentration in imports of Components. Therefore, China may achieve higher than average economic gain by importing and getting the benefits of the terms of trade on Components. And, with high variety (low H) in exports of Other ICT, China may gain from market power in trade, with attendent benefits of higher prices and profits.

 

This short paper draws on Chapter 7, “Information Technology, Globalization, and Growth: Role for Scale Economies, Terms of Trade, and Variety”, in Otaviano Canuto and Danny Leipziger, eds 2012. Ascent after Decline : Regrowing Global Economies after the Great Recession. © World Bank . https://openknowledge.worldbank.org/handle/10986/2233

 

 

 

 

 

 

Catherine L. Mann
Barbara ’54 and Richard M. Rosenberg Professor of Global Finance
International Business School, Brandeis University
Visiting Scholar, Federal Reserve Bank of Boston

 

Back to Start Page