Causation is MUCH Harder to Show Than Correlation: The 2021 Sveriges Riksbank Prize in Economic Sciences (in Memory of Alfred Nobel)
What is a “natural experiment?” How do economists effectively use observational data? What kinds of experiments occur "naturally" that economists can study?
The 2021 Sveriges Riksbank Prize in Economic Sciences (in Memory of Alfred Nobel) was recently awarded to three economists, David Card, Joshua Angrist and Guido Imbens. The Nobel Committee said of their work:
This year’s Laureates – David Card, Joshua Angrist and Guido Imbens – have shown that natural experiments can be used to answer central questions for society, such as how minimum wages and immigration affect the labour market. They have also clarified exactly which conclusions about cause and effect can be drawn using this research approach. Together, they have revolutionised empirical research in the economic sciences.
When setting up experiments, economists have to ask specific questions and avoid selection and observational biases. This challenge is often present when attempting to measure the effectiveness of something like a government program or policy change. How can economists confidently claim that certain effects are linked to a specific cause? This year’s Nobel Laureates’ work with natural experiments provide examples of best-practice research methods that work to answer important questions.
Watch the Presentation:
On November 4, 2021, Dr. Kevin Grier of Texas Tech University discussed the winners of the 2021 Sveriges Riksbank Prize in Economic Sciences and explored their backgrounds and contributions to the practice of economics. During this virtual session, educators and students learned about the winner’s work using natural experiments and novel analysis. Dr. Grier discussed several examples of natural experiments including:
Oregon Medicare Access (Baicker & Finkelstein) that explores a 2008 Medicaid program where participants were selected by lottery. These selected participants could then be compared to the individuals who’d signed up to participate in the lottery but were not selected.
Vietnam Draft Lottery (Angrist) where young men were drafted to military service based on birth dates. The research compares lifetime earnings of drafted to non-drafted individuals.
Mariel Boatlift (Card) analyzed immigration impacts on labor markets due to the arrival of Cuban refugees into Florida during April - September 1980.
Extractive Institutions (Dell) Between 1573 and 1812 in Peru and Bolivia, the Spanish demanded forced labor in the silver mines from villages inside a boundary drawn based on altitude. Researchers compare villages that were just outside and inside of this arbitrary boundary.
Internet and Violence (Grier) During the Libyan Revolution, eastern Libya experienced a state-implemented blackout shortly after widespread riots and protests began. This study compares violent protests and actions in rebellious areas with internet access to those in cut-off rebellious areas, and shows that access to the internet significantly reduced violence.
Northern and Southern Korea after the Korean war shows the impact of leadership and government policy on otherwise similar populations.
Connections to Arkansas Learning Standards:
Economics:
EM.2.E.1 Identify various allocation methods used in different circumstances, countries, and economies (e.g., price, auction, lottery, fiat, etc.)
EM.3.E.3 Evaluate intended and unintended consequences of government policies created to improve market outcomes
Quantitative Literacy:
SP.3.QL.2 Analyze, interpret, and compare charts, tables and graphs of real world data
SP.3.QL.3 Analyze statistical information from studies, surveys, and polls to make informed judgements as to the validity of claims or conclusions (e.g., bias, limitations, sampling, causation vs correlation, misuse of statistics)
Learn more:
Would you like to watch more presentations from our expert scholars? Make sure you are subscribed to the Arkansas Center for Research in Economics YouTube channel to receive alerts when new speakers and topics are added.
