Is the world closer to the free trade or no trade?

Factful Friday by Richard Baldwin, Professor of International Economics, 瑞士洛桑国际管理发展学院 (IMD) - 商学院 12 January 2024

so prices are the same for everyone everywhere), consumers will all have the same consumption pattern. As Keynes put it in a famous passage in his famous book on globalization: "the inhabitant of London could order by telephone, sipping his morning tea in bed, the various products of the whole Earth, in such quantity as he might see fit, and reasonably expect their early delivery upon his doorstep ...".

The identical-preferences assumption is a whopper but suspend your disbelief for a couple of pages and see if the result helps clarify your thinking.

The last logical point to stress is that the consumption pattern of each consumer has to match the global production pattern. The earth is, after all, a closed economy. At the planetary level, consumption and production are perfectly bundled.

The ideal free trade and no trade relationships.

So, what's the prediction? What would the data look like if trade were perfectly free? The answer is simplicity itself.

Countries would consume other countries’ production in proportion to their global income shares. Each country's income share would equal the value of its production as a share of world production. Each country's production pattern would be determined by national supply side constraints and the global balance of supply and demand.

To put it differently, a country's consumption of its own production would be in proportion to its world income share, so it would export a share of its production equal to the rest of the world's income share. The rest of the world's income share is exactly equal to one minus the country's income share, so in symbols:

That is, if the world really had perfectly free trade, the relationship between countries’ export and income shares would, graphically speaking, be a straight line with a negative slope of minus one. This is shown with the red line in the graphic. At the other extreme of autarky, all the national export shares would be zero regardless of the country’s world income share. This is shown with the blue line in the graphic. Just to be clear, country A in the diagram would have an export share shown by the blue dot if no trade were possible, but the red dot if trade were free. And the same for hypothetical countries B and C who have higher income shares

.What does the data look like?

In the real world, intermediate goods are super important, so a country’s production does not equal its GDP. This reality forces us to make choices that are not perfectly clean from the theory perspective. Here, I’ll use gross production and gross export data to work out the national export shares, and value added to work out the national income shares. The data comes from the OECD’s invaluable TiVA dataset.

The left panel of the chart below plots the export shares against the income shares as well as the ideal free trade relationship (red dots) and the ideal no-trade relationship (blue dots). Specifically, each red dot represents what the ideal export ratio would be for each country (given the country’s income share) if there were free trade. The blue dots do the same for the no-trade extreme.

When turning to the actual data for countries in the database, we start with the total economy export shares, that is, the ratio of national total exports of all goods and services to total national production. These ratios are represented by green dots. Observe that the fitted curve has a negative slope so bigger nations tend to be more closed. To which extreme does the data conform?

Plainly the real data (green dots) does not fit well with either extreme. Yet, to me, it appears that the green dots are closer to the blue dots than to the red dots. In other words, the data suggests the world is closer to autarky than it is to free trade.

What about non-traded goods, like services?

Now readers may object to using the whole economy when large swaths of most economies are non-traded. Many forms of services, for example, are not traded. The rebuttal is straightforward. If large parts of most economies are non-traded, then the world is not close to free trade.

Be that as it may, I can easily redo the exercise for manufacturing only. This may seem like a fairer comparison since manufactures are tradeable for the most part.

A moment’s reflection reveals that the same logic holds for the export ratio of the manufacturing sector as it does for the whole economy. These manufacturing export ratios are plotted as the yellow dots. The manufacturing export shares are generally higher than the whole-economy shares. The fitted line is closer to the red dots and has a steeper slope, but I would still say that the yellow dots are closer to the blue dots than to the red dots.

Most countries in the world, and in the database that I’m using, have very small shares of world income. That is why so many of the dots are crushed up against the vertical axis. To better see what’s going on, the right panel truncates the horizontal axis at 30% (which is close the US’s share of world income). The numbers are all identical, but the truncation helps us distinguish the dots from one another.

One other simple test for free trade

In passing, I mentioned the other obvious test for free trade – global equality of prices. As O’Rourke and Williamson (2002) taught us so long ago, using trade shares to measure openness is an endeavor fraught with pitfalls – not the least of which is the home market bias. The true test is the equality of prices. And indeed, in some highly integrated markets – say the foreign exchange market for dollars – arbitrage forces prices in different countries to be extremely close. The same is true but less so for widely trade commodities such as oil and wheat. When it comes to non-traded things, say housing services, prices can be radically different even within a single city.

In short, if the world were close to free trade, then prices in nations should be very similar. What does the data say about national price differences?

Given that there are a vast number of goods, I will go with a comparison of price indices that are based on typical consumption baskets. Using data from the World Bank’s World Development Indicator database, the next chart plots the series called ‘Price level ratio of PPP conversion factor (GDP) to market exchange rate’. This is a measure of the difference between the number of units of a country's currency needed to buy the same amount of goods and services in the domestic market and in the US market. A number above 1.0 means the country, say Switzerland, has prices that are on average higher than those in the US. A number below 1.0 signals the opposite.

The facts couldn’t be plainer. Of 194 countries and territories with data, only 30 are within one standard deviation of the US price level (the standard deviation is about 0.2). A couple are higher, but most are lower. The most common figure is four standard deviations below this level.

If there were free trade, those low prices would trigger arbitrage that would tend to equalize the prices. Since the price differences are large and persistent, this is evidence that the world is a long way from free trade

.Reality check and concluding remarks

Some readers may object. They may think I’ve stacked the deck. No one really thinks prices are anywhere near equalised globally or domestic consumption shares are close to national global income shares. So, what have we learned?

My answer is that there is a big disconnect between the widely held perception that the world is very globalized today, and the actual facts.

Naturally, there are many good explanations for why national export ratios are so low and prices are so different, with home-market bias being one of them. Cultural, linguistic, legal, and historical differences constitute another set of explains, along with natural and fabricated trade barriers. But these are reasons why the world is far from free trade, not rebuttals of the finding.

In short, the world is nowhere near free trade. I would argue that it is closer to autarky – and will be until digital technology lowers the barriers to trade in services. But that’s the subject of my 2019 book, The Globotics Upheaval, and the topic of future Factful Fridays.

References

Baldwin, R. (2006). “Globalization: the great unbundling(s),” Chapter 1, in Globalization challenges for Europe, Secretariat of the Economic Council, Finnish Prime Minister’s Office, Helsinki, 2006; ISBN 952-5631-15-X.

Baldwin, R. (2016). The Great Convergence: Information Technology and the New Globalization. Harvard University Press.

Baldwin, R. (2019). The Globotics Upheaval: Globalization, Robotics, and the Future of Work. Oxford University Press.

Keynes, J. M. (1919). The Economic Consequences of the Peace. Macmillan.

O'Rourke, Kevin H., and Williamson, Jeffrey G. (2002). "When did globalization begin?" European Review of Economic History, Cambridge University Press, vol. 6 (01), pages 23-50, April.