Why 2000 Census data is the most reliable measure of New Orleans’ pre-Katrina population
Published: Jul 25, 2025
In August 2005, Hurricane Katrina and the failure of the federal levees caused more than 1 million people to be displaced from the New Orleans metro area and across the Gulf Coast.1 In the two decades since, hundreds of reporters and researchers have examined the demographic changes in New Orleans and the surrounding area. While the Census Bureau produces annual population estimates for New Orleans and the metro as recent as 2024, the challenge is identifying the most accurate pre-Katrina population estimates.
The Census Bureau publishes several population estimates for 2005, but unfortunately, each of these are unreliable representations of New Orleans’ pre-Katrina. This blog briefly describes each of the Census Bureau’s 2005 demographic data products and why they do not accurately reflect New Orleans’ pre-Katrina total population or demographics.
Annual Population Estimates
Each year, after the U.S. Census Bureau completes its once-every-ten-year head count, the Bureau produces an estimate of each counties’ total population and demographics which they call a “postcensal” estimate. Then, after the next end-of-decade head count is completed, the Bureau generates “intercensal” estimates that “smooth” the numbers between each decade headcount.
The two sets of annual estimates—postcensal and intercensal—that the Census Bureau produced for New Orleans for the years 2001-2005 differ greatly. In their postcensal estimates, they show the population of New Orleans declining from 2000 to 2005. In their intercensal estimates, they show the population of New Orleans increasing from 2000 to 2005.
It is rare for the Census’ postcensal and intercensal estimates to vary so much for the same geography reflecting the same years, but sometimes it happens. When the two vary as widely as they do for New Orleans, then that casts doubt on the certainty of either set of estimates.
American Community Survey
Some reporters have relied on the Census Bureau’s 2005 American Community Survey (ACS) data for their pre-Katrina benchmark. But they don’t realize that the ACS uses the Census’ postcensal estimates—which as described above is quite uncertain. Moreover, in 2005, the American Community Survey was in its infancy and only measured demographic characteristics of “household population” rather than “total population” at that time. In other words, it excluded college dorm residents, jail populations, etc. Thus, it should not be compared with current “total population” estimates as a measure of population change.
Furthermore, the American Community Survey collects demographic characteristics on a sample of households every single month of the year and averages those observations all together to derive their detailed demographic profile for the year. In the later months of 2005, the Census was not able to even *locate* their survey takers in New Orleans! Needless to say, they weren’t collecting any data in those months. So, the American Community Survey data for New Orleans has a host of problems and should definitely not be used as a measure of pre-Katrina New Orleans’ population.
2000 Census
At the end of the day, the most reliable pre-Katrina measure of New Orleans’ population comes from the 2000 Census—which is the gold standard. This data provides a reliable and understandable benchmark for New Orleans’ pre-Katrina population.
For example, in our wildly popular analysis of “Who Lives in New Orleans and Metro Parishes Now?” we find that: “Compared to 2000, pre-Katrina, there are now 121,142 fewer Black residents living in New Orleans (Orleans Parish), and there are 23,344 fewer White residents, as well as 15,284 more Hispanic residents.” And “In Orleans Parish, the share of the 2024 population that is Black—while lower than in 2000 when it was 67 percent—continues to represent the majority of city residents at 56 percent.”
Moreover, in our heavily trafficked web pages covering New Orleans Neighborhood Data you can access dozens of demographic indicators about each of New Orleans’ 72 neighborhoods, reflecting both 2000 and 2019–2023 (the most recent neighborhood level ACS data) for easy analysis of changes from pre-Katrina to now.
Conclusion
While it may be tempting to rely on the more recent 2005 estimates or early ACS data for a snapshot of New Orleans just before Hurricane Katrina, these sources simply do not offer the consistency or accuracy required for meaningful analysis. The 2000 Census remains the most reliable and comprehensive benchmark for understanding the city’s pre-Katrina population and demographic landscape. Whether tracking neighborhood-level change, analyzing racial and ethnic shifts, or assessing long-term trends, researchers, journalists, and policymakers should ground their comparisons in the 2000 Census to ensure clarity, credibility, and historical integrity.
Endnotes
1
https://www.columbia.edu/itc/journalism/cases/katrina/Federal%20Government/Congress/U.S.%20Senate/Senate%20Hearings/Senate%20Hearing%202005-10-06/Paulison%20Remarks%20Oct%206%202005.pdf