Blog Post < Previous | Next >
Guy
Sacramento California Population Projection too High
Abstract
The California Department of Finance, Demographic Research Unit (DRU) recently released their projection for California's population out to 2070.
To benchmark the quality of the DRU's projections, I will benchmark them in two ways:
- First by comparing them to the UN Medium Scenario for the US. I will check that the relative population growth trends between California and the US are consistent when comparing the projection period (2024 - 2070) vs the recent historical period (2005 - 2024).
- Second by checking that the DRU annual natural growth rate and annual migration rate assumptions (2024 - 2070) in their projections are consistent with recent history (2009 - 2024).
The DRU projections do appear reasonable when compared to the UN Medium Scenario for the US. The DRU projections do show California growing much slower than the US. This is consistent with recent history.
However, the DRU uses overly optimistic migration rate assumptions that artificially boost their California's population projections. Using my independent model associated with much more realistic migration assumptions I calculate that by 2070, the DRU may have overestimated California's population by 1.5 to 3.3 million.
California's population growth history
California's population growth history can be divided into three periods:
1) The Boom period from 1950 to 1992. Over this period, California grew much faster than the US;
2) The Convergence preiod from 1992 to 2005. Over this period, California's population growth converged with the US.
3) The Slowdown period since 2005 to the present when California's population grew much slower than the US.
You can see these three different periods visual trends in the graphs below.
The table below quantifies the visual data shown above.
Since 2005, California's population growth has drastically slowed down. And, it has grown much slower than the US. This is due to several phenomena which had a more severe impact on California vs the remainder of the US. These included the Housing Crisis, COVID and Work From Home (WFH). The latter, WFH, had a profound effect on California vs close to none on the US as a whole.
The DRU California population projections vs the UN US Medium Scenario
In the right-hand graph below, comparing the DRU California population projection vs the UN projection for the US Medium Scenario; the DRU projections look reasonable. This is because it does show California growing a lot slower than the US. And, this trend is consistent with the current Slowdown Period (left-hand graph).
Below, we use a table and a graph to summarize the respective growth rates shown above.
The DRU underlying demographic assumptions
Natural Growth Rate
The DRU's natural growth rate assumptions are perfectly fine. They assume that California is aging more rapidly than the US (lower fertility rate and higher mortality rate).
As shown on the graph below, the DRU foresees that California's natural growth rate will continue declining rapidly and turning negative in the mid 2030s. By the 2060s, California's natural growth rate will be very negative at close to negative - 0.40% per year, much lower than the US that will be much closer to 0%.
The dashed vertical green line denotes when the forecast period starts in 2023.
Migration Rate
The DRU migration assumptions are way too high.
The migration rate into California was steadily a lot lower than for the US going back to 2009. And, it was steadily negative since 2017 to the present. Thus, in the midst of an immigration crisis, California has steadily experienced a negative migration rate. This tells you that the California outlook for migration in the State is not good. Yet, the DRU has California's migration rate bouncing right back into positive territory and near record pre-pandemic levels, and remaining near this record level till 2070. That is most unlikely.
The dashed vertical green line denotes when the forecast period starts in 2023.
Generating a realistic California population projection
The graph below compares the DRU projection with my model forecast output. I built this forecast by using the DRU's natural growth rate projections. And, I used a migration rate that was the 3d highest over the past 15 years (equivalent to 80th percentile). Thus, my model is still using a pretty optimistic migration rate assumption going forward all the way up to 2070. Yet, my using only a slightly more realistic migration assumption than DRU made a huge difference.
By 2070, the DRU population projections end up being 1.5 million higher than my slightly more realistic forecast.
The graphs above depicted my forecast using the 3d highest historical annual migration rate of 0.161% per year (out of 15 years in the data). And, my resulting forecast comes in close to 1.5 million fewer individuals by 2070. The table below shows my forecast if I used the 3d to the 7th highest annual migration rate.
As
indicated, the DRU forecast appears to overshoot by 1.5 million to 3.3
million depending on which migration rate assumption one use. Notice
that all the scenarios I have used a migration
assumption that is above the historical median. Thus, all these scenarios are still on the optimistic side.
The DRU most recent projection is just another one that will overshoot as they often have in the past.
THE END
Gaetan Lion is an independent researcher with expertise in economics, econometrics, investments, risk management, statistics, and quantitative analysis. His areas of interest include demographics, economics, investments, Climate Change, health care, sports, and politics.