Skip to content

US: Period Effects Matter More than Cohort Effects With Fertility Rates

Study: Age-period-cohort analysis of U.S. fertility: a realistic approach

Demographers have reignited the long-standing debate on what drives changes in fertility rates: period or cohort effects.

  1. Period Effects: These are influences that affect an entire population at a specific time, regardless of age or generation. For example, the 2008 Great Depression or pandemics like COVID-19 are events that have a period effect. These events impact everyone living through that time, influencing behaviours, decisions, and outcomes like fertility rates.
  2. Cohort Effects: On the other hand, are influences that affect a specific group of people born around the same time. Cohorts share similar experiences, attitudes, and values because they grew up during the same historical and social period. For instance, people born during the Great Depression may have different spending habits than those born during a time of economic prosperity. Similarly, a generation that grew up with the internet will have other behaviors and skills compared to a generation that did not.

This discussion, with roots going back to seminal reviews in 1982 by Hobcraft, Menken, and Preston, remains a hot topic in demographic circles. The challenge? There’s still no consensus on separating period and cohort effects from age-specific fertility data, which is crucial for understanding broader fertility trends.

The U.S. Baby Boom: A Case Study in Demographic Analysis

The focus of this new research is the U.S. baby boom, a pivotal period due to its scale and societal impact. Utilizing an Age-Period-Cohort (APC) analysis, researchers dissected this demographic phenomenon using U.S. fertility rates from 1933 to 2015. The goal was to parse out the distinct effects of age, period, and cohort – each a key factor in understanding historical fertility trends and forecasting future ones.

A New Perspective on APC Analysis

Researchers opted for the APC method previously applied to mortality data. This approach sets itself apart by focusing on identifiable and measurable parameters. It’s a strategy grounded in solid statistical reasoning, as shown by earlier research.

Methodological Innovations

Eschewing unproven assumptions, the team built upon prior work by Pullum in 1980, employing a more grounded approach. Their findings suggest that changes in both period and cohort influenced fertility, with period effects playing a more dominant role. This insight contributes to both methodological literature in demography and our grasp of the U.S. baby boom.

Yet, APC analysis isn’t without its challenges. The standard model often grapples with identifiability issues due to the intertwined nature of age, period, and cohort components. While solutions like the Constrained Generalized Linear Model (CGLM) have been proposed, the quest for a definitive resolution continues.

Key Findings

  1. Significance of All Three Components: The analysis demonstrates that age, period, and cohort are all essential for understanding long-term trends in U.S. fertility rates. The complete APC model, including all three components, is preferred due to its lower AIC value.
  2. Period vs. Cohort Analysis: The age-period and age-cohort models generally show a better fit when comparing two-factor models (age-period and age-cohort). This suggests that period effects influence U.S. fertility trends more than cohort effects.


  1. Consistency with Earlier Research: These findings align with previous analyses like Pullum (1980), which emphasized the relative importance of period identification over cohort.
  2. Identification of Non-Linear Trends: The study observes non-linear trends, particularly around the post-World War II era, indicating significant period effects like the baby boom and baby bust.
  3. Cohort Effects: Although less pronounced than period effects, cohort influences, especially for those born around the 1940s, are observed as continuous over time.

Policy Implications

  1. Informed Family Planning and Social Policies: Understanding the relative importance of period effects over cohort can guide government policies in family planning and social welfare, particularly in response to historical and cultural shifts.
  2. Healthcare and Education Planning: Recognizing the demographic trends and their underlying causes can assist in planning for healthcare and education needs, and adapting to changing fertility patterns.
  3. Economic Forecasting and Labor Market Strategies: Insights from this study can be valuable for economic forecasting and developing labor market strategies that align with anticipated demographic changes.
  4. Demographic Research and Methodology: The methodological approach and findings of this study can be extended to other demographic analyses, providing a template for examining age-specific vital rates in various contexts.
  5. Broader Application of Findings: While focused on the U.S., the method and conclusions could be relevant to other countries facing similar demographic shifts, aiding in global comparative demographic studies.