The essence of trade is that both the seller and buyer benefit (win-win). Without that feature the trade would not take place. The expansion of trade locally and then globally increased the output and thus incomes of the average person dramatically.
In 1820, about 80% of the world’s population lived in extreme poverty (defined as living on less than $2.15 per day in today’s terms). By 2019, this figure had fallen to roughly 10%. This decline is especially notable given that the global population increased more than sevenfold during this period.
The pace of poverty reduction accelerated in recent decades. From 1990 to 2019, the global extreme poverty rate dropped from 43% to below 10%, with the fastest declines occurring since the 1990s. This progress was driven largely by rapid economic growth in Asia, particularly in China and India.
The increase in win-win gains in income from trade have been promoted by broad agreement on rules and norms for “fair trade” to maximize the increase in incomes that results. These have been developed over time through what is now called the World Trade Organization (WTO). Tragically, rather than further improving its rules, the U.S. has undermined the WTO by refusing to appoint new members to its dispute resolution body.
The benefits of such collaborative cooperation have been sought and gained in other areas as well. To take one, the climate benefits of nuclear energy also carries the risks of destruction from nuclear bombs. Agreements among the countries with such capacity to contain and minimize the associated risks are reflected in the Treaty on the Non-Proliferation of Nuclear Weapons (NPT) of 1968 (extended in 1995). The Comprehensive Test Ban Treaty (CTBT, 1996), several bilateral agreements with the USSR/Russia and others have further reduced the risks.
The dramatic development of Artificial Intelligence (AI) programs promises incredible increases in our incomes but also carries risks. As with nuclear energy, all would benefit from agreements that limit these risks. Cooperating in developing such guard rails is in everyone interest. The US is making a big mistake in attempting to stifle China’s AI development rather than a win-win cooperation with them to maximize its promise while minimizing its risk.
The case for such cooperation with China is powerfully made by Alvin Graylin in a recent presentation to the Committee for the Republic (on whose board I serve) the other day. https://www.youtube.com/watch?v=Jg6brPvFJGw.