U.S. flag

An official website of the United States government

Skip Header

New Population Estimates Rely on Monthly Flight Data to Capture Puerto Rico’s Migration Flows

Written by:

Net outmigration from Puerto Rico, which increased after Hurricane Maria in 2017 and a 6.4-magnitude earthquake in 2020, declined during the COVID-19 pandemic, according to U.S. Census Bureau population estimates released today.

How did the Census Bureau calculate migration patterns and improve migration estimates in the wake of these major events?

For the latest estimates, the Census Bureau relied solely on annual flight passenger data to estimate net migration between Puerto Rico and the rest of the world.

By developing new methods of measuring migration that incorporates flight data with Census Bureau survey-based estimates.

For the latest estimates, the Census Bureau relied solely on annual flight passenger data to estimate net migration between Puerto Rico and the rest of the world.

Moving to a flight data-based method improves the accuracy and timeliness of net migration estimates for Puerto Rico. As a result, flight data can better capture the impact of COVID-19 and reduce the need for future adjustments to account for major events that affect migration. 

Puerto Rico Net Migration Through the Years

Net outmigration is the number of people who moved away from Puerto Rico minus the number who moved to its islands.

Throughout the decade, net outmigration from Puerto Rico remained at consistent levels until Hurricane Maria resulted in unprecedented high outmigration. However, this was followed by return migration the succeeding year.

From July 1, 2019 to June 30, 2020, population estimates showed Puerto Rico returning to lower levels of net outmigration. This coincided with the start of the COVID-19 pandemic in March 2020, which slightly reduced net outmigration for the period.

Slowdown in net outmigration continued into this year. Population estimates from July 1, 2020 to June 30, 2021 released today show net migration was only slightly negative.

Using Flight Data to Measure the Impact of Migration Disruptions

For estimates year 2021, using monthly flight data helped reflect the impact of COVID-19 on annual migration patterns for Puerto Rico. The result: Levels of net outmigration from the island were much lower than typically seen in past years.

Puerto Rico normally experiences an influx of passengers during the winter and summer holiday season, followed by a reverse outbound flow.  

Prior to Hurricane Maria, these seasonal patterns were predictable and recurring. After Hurricane Maria, seasonal patterns eventually returned to normal until the earthquake in January 2020, and the start of the COVID–19 pandemic in March 2020.

The new estimates cover a period that falls fully within the COVID-19 pandemic.

Monthly patterns show higher out-passenger movement for Puerto Rico in August 2020, and atypical in-passenger movement in September 2020, and in March and June 2021.

Overall, net outmigration from Puerto Rico declined, most likely because fewer people left the island during the pandemic.

Other than return migration immediately after Hurricane Maria, Puerto Rico has not experienced such a dramatic drop in total net outmigration within the past decade. This making the impact of COVID-19 on migration particularly significant.

A New Way to Measure Net International Migration for Puerto Rico

Starting this year, the Census Bureau began using Airline Passenger Traffic (APT) data from the Bureau of Transportation Statistics (BTS) to directly measure net migration to and from Puerto Rico and the rest of the world.

The previous methodology estimated migration flows from Puerto Rico to the mainland United States based on responses to the American Community Survey (ACS) residence one year ago (ROYA) question, which asks where respondents lived one year prior to the survey.

Conversely, migration flows from the United States to Puerto Rico were obtained from the Puerto Rico Community Survey (PRCS) ROYA question. The net migration estimate was derived from subtracting the in- and out-migration flows.

ACS/PRCS data are collected on a continuous basis throughout the calendar year, though movement could have occurred at any time over a two-year period, depending on when the respondent was included in the sample and when they actually moved.

Airline Passenger Traffic Data

In the United States, flight data are compiled from monthly reports filed by over 200 commercial U.S. and foreign air carriers with BTS, including both domestic and international flights.

Data are reported for all flights following federal reporting guidelines that went into effect in October 2002. APT data include information on all travelers, without differentiation of passenger type. It includes tourists and visitors who make up the majority of passengers.

Nonmigrants are counted on both inbound and outbound flights. Migrants are only counted in one direction unless leaving temporarily, in which case they would be counted again upon their return.

However, there are limitations to this method.

For example, it can only provide a number for “net” migration, with no information on total in- or out-flows. Migrants cannot be distinguished from the total number of passengers entering or leaving Puerto Rico.

Another limitation is that no demographic characteristics of net migrants are included in the data. Additionally, this method only applies to a country or territory without any land borders, such as islands like Puerto Rico, where airplane travel is the predominant method of arriving or leaving. Ship passenger movement to and from Puerto Rico is assumed to be minimal.

Monthly tallies of net airline passenger flow movement reflect seasonal variation related to tourism, with presumably greater movement into and out of Puerto Rico in the summer and winter vacation months.

Depending on the measurement period, such as a calendar year, this could lead to year-to-year fluctuations related to annual tourism trends.

For example, a high number of tourists could be counted in December, while the return of these same tourists might occur in January of the following year. Over time, these fluctuations are thought to balance out.

How the New Methodology Works

Annual net migration totals are calculated using in/out passenger flow data for the 2021 estimates year based on Airline Passenger Traffic data: Inbound total passengers to Puerto Rico are subtracted from outbound total passengers from Puerto Rico.

This method includes international passengers. Typically, domestic flight data are received with a three-month lag, while international data are available six months later.

However, the Census Bureau has an internal agreement with the Bureau of Transportation Statistics to access international data with only a three-month lag, enabling us to include current international flights in our estimates. 

There are several advantages to using a flight-based methodology over the former survey-based methodology:

  • Flight data can provide up-to-date estimates for the period we want to estimate.
  • Flight data include movement between Puerto Rico and the rest of the world, while the former methodology only included movement from Puerto Rico to/from the United States.
  • The period covered by flight data aligns with the July 1-June 30 estimates year while the ACS and PRCS cover the January-December calendar year. 


Jason Schachter is chief of the International Migration Branch in the U.S. Census Bureau’s Population Division.

Angelica Menchaca is a demographer in the International Migration Branch.


This story was filed under:


DATA GEMS: How We Estimate the Population in the United States


Our email newsletter is sent out on the day we publish a story. Get an alert directly in your inbox to read, share and blog about our newest stories.

Contact our Public Information Office for media inquiries or interviews.


Back to Header