Big Data, Meet Big Government
Discussions of big data typically revolve around concerns about privacy, excessive data collection, and cyber security vulnerabilities, but there’s a bright side to big data. It can give lawmakers and the President insights into how to govern effectively.
Trade policy is well-suited for help from big data because it has a clear goal: to improve the conditions of the national economy. It also relies on statistical data sets that are already in existence.
An algorithmic system for regulating trade has appeal because it is less vulnerable to the subjective biases of politicians. It is much harder to manipulate objective statistics that feed into an open, transparent algorithm than it is to influence a president with campaign donations and speaking fees.
Just as the Federal Reserve sets monetary policy based on a vast collection of economic data, trade policy should also rely on economic data. Right now, many tariffs are set relatively arbitrarily with a strong degree of influence from corporations and special interest groups. In an ideal world, trade policy has one objective: to improve the welfare of its country’s citizens without harming others.
A statistical proxy for that goal can be defined as either U6 unemployment or median inflation-adjusted income. While median income serves as a strong indicator of the well-being of a nation, inflation is heavily prone to manipulation when the methodologies are shielded from the public view (Boring, 2014). U6 unemployment is harder to manipulate, but it doesn’t capture economic growth, job satisfaction, or well-being.
An algorithm designed to optimize for one of these outcomes relies on a collection of parameters categorized in two groups. One group of parameters can be adjusted through policy changes. These are the tariffs. The second group of parameters serve as intermediate variables that either move with or against the desired outcome.
In summary, the rules-based approach to trade policy maximizes the economic welfare of Americans by using tariffs to respond to several economic indicators and numerical scores:
- National security score
- Reciprocity score
- Income score
- EH&S score
- Degree of rawness
By responding to these indicators openly, lawmakers can make it clear to trade partners what it takes to reduce trade barriers with the U.S. In the process of doing so, they incentivize foreign countries to improve their own conditions for workers and raise their standards of living. In aggregate, a tariff that depends on these factors addresses the underlying issues that lead to American trade deficits.
Regardless of the method used to arrive at a tariff, the method should be transparent and the numbers should be easily accessible to all Americans.
The national security score is both country-specific and good-specific. It captures the amicability of the relationships between the U.S. and its trading partners. It also prioritizes onshoring the production of goods and services that are critical to the military capabilities of the United States.
The reciprocity score accounts for intellectual property infringement abroad, penalizing countries that don’t properly enforce intellectual property laws that protect innovators across the world. It also accounts for the openness of a trading partner to accepting American goods and services. Low tariffs abroad would benefit a country’s score, and the removal of bureaucratic barriers to our exports would have a similar effect. This score would also capture the tendency of a trading partner to adhere to American rules when exporting products to the States. A higher occurrence of violations would hurt a country’s score. For example, massive shipments of corn and soybeans have been falsely labeled “USDA Organic” before arriving at American ports (Washington Post, 2017). Trade policy should punish companies and countries that export goods to the U.S. at lower prices simply because they deceive American consumers and skirt our regulations.
The income score represents the median after-tax income among residents of the trading partner country. The EH&S score corresponds to environmental, health, and safety standards abroad. Countries with policies that take advantage of workers, harm the environment, and risk consumer health should be lesser trade partners than countries with progressive policies in place.
The degree of rawness captures how close a good or service is to being a raw material and natural resource. A good with a high degree of rawness, such as aluminum ore, should have zero tariffs because it is an important commodity in a variety of other products, such as aircraft components and beverage cans. A finite quantity of aluminum is available within our borders, so we should allow as much as possible to come in from overseas to enable the success of the many industries that rely on aluminum. On the other hand, a good such as a spacecraft has a low degree of rawness and high level of complexity. Manufacturing a spacecraft is considered an advanced form of manufacturing that provides high-wage jobs and is generally desirable to have within the U.S. For most goods, the degree of rawness will tend to rise over time as newer goods replace older ones.
Tariffs created according to this framework would have a tendency to be lower with developed countries than with developing countries, which would minimize impacts on the current global trade network. This data-driven approach would mitigate the trend among multinational conglomerates to offshore manufacturing and service centers to foreign cities with minimal labor protections.
After all, economic surpluses that might arise from these lower foreign business costs tend to collect in the hands of the economic elites that operate and own these companies, so moderate, targeted tariffs would be unlikely to result in widespread increases in prices of goods and services, even for Walmart and Target shoppers. Furthermore, tariffs would shift the burden of taxes off the backs of the hard-working middle class onto companies importing foreign goods, and in the process, these tariffs could make a dent in the national fiscal deficit.
Machine learning algorithms are at the core of companies raising massive amounts of venture capital in Silicon Valley. Why shouldn’t data analytics also drive government policy?
Browse graphs from the Observatory of Economic Complexity, the world’s leading visualization engine for international trade data: