Does Smoking Cigarettes Actually Lead to Cannabis Use? Testing Gateway Theory with Economics | Research

The takeaway

Using cigarette prices as a natural experiment in Israel, researchers found evidence that cigarette smoking led to cannabis use, but the evidence that cannabis led to harder drugs was much weaker.

Read this if you have heard that cannabis is a gateway drug and want to see what the research actually shows.

Smoking to cannabis link supported; cannabis to hard drugs link weak

What the researchers found

The researchers used variation in cigarette prices across birth cohorts in Israel as a natural experiment to test gateway theory. By treating cigarette prices as an external factor that randomly influenced smoking rates, they could examine whether changes in smoking rates caused corresponding changes in cannabis use, and whether cannabis use in turn caused hard drug use.

The data supported the first link: cigarette smoking appeared to cause cannabis use. However, the evidence for the second link, that cannabis use caused hard drug use, was much weaker. These results held across multiple statistical methods including two-stage logit, bivariate probit, and frailty analysis for survival data.

Why it matters

Gateway theory, the idea that softer drugs lead to harder ones, has been one of the most influential and contested concepts in drug policy. This study's use of econometric methods and natural experiments brought a different analytical approach to a question usually studied through correlational survey data. The finding that the cigarettes-to-cannabis link was stronger than the cannabis-to-hard-drugs link challenged the simplistic version of gateway theory often used to justify cannabis prohibition.

The numbers in context

Specific effect sizes and price elasticities were not detailed in the abstract. The study used Israeli data across multiple birth cohorts.

How the study worked

This was an observational study using econometric methods applied to Israeli population data. The researchers used cigarette prices as an instrumental variable, exploiting the fact that price variation across birth cohorts created a natural experiment that randomly influenced smoking rates. Multiple statistical approaches (two-stage logit, bivariate probit, frailty analysis) were used to test causal relationships in the drug use sequence.

What this study cannot tell us

The study relied on Israeli data and may not generalize to other countries with different drug markets and cultural contexts. Using cigarette prices as an instrumental variable assumes that price variation affects only drug use through smoking behavior, which may not hold perfectly. The abstract did not provide detailed effect sizes or confidence intervals.

How to read the evidence

This is an observational study using sophisticated econometric methods with instrumental variables, providing moderate evidence through causal inference techniques.

When this study was published

Published in 2002. Gateway theory continues to be debated, with subsequent research generally supporting common factor models over simple causal chain explanations.

The bigger picture

Gateway theory remains debated. While sequential patterns of drug use are well-documented (most hard drug users did try cannabis first), the causal interpretation is contested. Common factor models suggest that shared genetic and environmental risk factors, rather than cannabis use itself, explain the correlation. This study's finding that the cannabis-to-hard-drugs link was weak when tested causally has been echoed by subsequent research.

Questions still open

Does the gateway pattern reflect a causal chain, shared risk factors, or simply drug market dynamics where the same dealers sell multiple substances? Would similar econometric approaches in other countries produce the same results?

Common questions

Is cannabis really a gateway drug?

This study found that while cigarette smoking appeared to lead to cannabis use, the evidence that cannabis use itself caused progression to harder drugs was weak. Most research now suggests that shared risk factors and drug market dynamics explain the sequential pattern better than a direct causal chain.

How did the researchers use cigarette prices to study drug use?

Because cigarette prices varied across birth cohorts, they created natural variation in smoking rates that was unrelated to other factors that influence drug use. This allowed the researchers to isolate the causal effect of smoking on subsequent cannabis use, rather than just observing a correlation.

Read the original research

Testing Gateway Theory: do cigarette prices affect illicit drug use?

Journal of health economics, 21(4), 679-98

View on PubMed ↗

Citation

Beenstock, Michael; Rahav, Giora. (2002). Testing Gateway Theory: do cigarette prices affect illicit drug use?. Journal of health economics, 21(4), 679-98.