January 14, 2017

Is California's Population Forecast Greatly Overestimated?

Is California's Population Forecast Greatly Overestimated? I became interested in this while trying to imagine possible futures for my job working on buildings and their energy demand. Here is a chart of California (CA) population so far:
Figure 1. Chart of California Population 1900-2016
Next, take a look below at the annualized growth rate taken from the population data. The growth rate has been on a neat trajectory toward zero for the last 100 years, with a 0.76 correlation coefficient:
Figure 2. California Population Rate of Growth, 1900-2016
What will happen in the future? Will population growth cease in 2032, where the lines cross? We have an official forecast from the California Department of Finance. The data that I used for this post is here. Sometime in summer of 2015, the CA Department of Finance projected robust population growth through 2060. Their projection seems to serve as the basis of all planning forecasts for the state. This graphic appeared on the web about that time. The growth of non-white groups was emphasized in the media:


It's ironic now in 2016 that these inland counties forecast for high population growth have some of the worst drought-related problems in the state. Another connection: the California Energy Code Title 24.6 focuses especially on new tract housing in these warmer inland climate zones.

However, something's not right. If you look at the CA Dept. of Finance numbers in a spreadsheet and back out their assumed growth rates, they are unexpectedly high: above 0.7% through 2040 and 0.5% in the two decades after. What's the basis of these high rates? They have no relationship to the trend data. Mainstream economic models appear to not handle negative growth, so it makes sense that the California Department of Finance would just assume a small positive number for future projections, in lieu of zero or negative. How about instead, we extend the trend growth rate linearly through zero to negative. This fits the data, and fits the headlines. Here is a comparison chart of the two different growth rate scenarios:
Figure 3. Comparison of CA Dept. of Finance Projection with Trend Rate
Figure 3 shows a linear projection of the current trend, and a comparison to the 2015 Dept. of Finance projection which breaks a 100-year trend to take a new and different line. I'm not sure why they did this. Now compare the resulting estimates of future California population in Figure 4:
Figure 4. Comparison of Estimates of Future California Population
Figure 4 shows a huge difference between the trend and the official forecast of nearly 20 million people by 2060. This has significant impacts for all kinds of estimates and policy for the state. Relevant to my work in my work in buildings and energy — and why I became interested in this data — we won't need to build as much new energy infrastructure if our population and demand will be smaller in the future. We won't need as many buildings. Our greenhouse gas (GHG) emissions will not be as high as estimated assuming a growing population. We won't need to choose between the lesser evils climate change or nuclear energy because we have population and economic contraction as a third option.

I'm inclined to believe the declining curve rather than the Dept. of Finance projection, because it seems more supported by widespread evidence:

1) For example, look at this study showing fewer babies since the Great Recession, "The Hidden Cost of the Recession: Two Million Fewer Births and Still Counting [PDF]":
Professor Johnson calculated that if births had continued at the pre-recession rate, Americans would have had 3.4 million additional babies in 2008–2015. He found no evidence to suggest that this trend changed in 2016, so the total is likely near 4 million by now. (link)

2) Young people don't have high-enough paying jobs, can't afford housing, have big college loans, and are living at home. In December 2014 the LA Times reported "California millennials living longer with parents, marrying less". Will they be having lots of babies while living at home? It's possible. We had our second child while living with parents! We couldn't afford mortgage & childcare & property tax on our two incomes. :-)

3) Corresponding to these headlines, the Dept. of Finance E-2 data actually show that the residents' natural rate of increase (births minus deaths) has plunged since the 2009 recession:
Figure 5. Natural Increase = Births minus Deaths of resident population
 At this trend rate of decline shown in Figure 5, deaths will outnumber births by 2032. The date when "natural increase" turns negative — when deaths outnumber births — is similar to the overall population trend date turning negative, shown in Figure 3 above and Figure 6 below:
Figure 6. Trend Projection to California "Natural Increase" going Negative
4) Emigration to other states is increasing again. Here is a story from the San Jose Mercury news about exodus of long-time residents. "California’s skyrocketing housing costs, taxes prompt exodus of residents." It's interesting that the exodus of residents is increasing again, but the state is still attractive to foreign immigrants, who offset this emigration. Figure 7 below from historical Dept. of Finance data shows that residents are increasingly leaving for other states, but foreign immigration has held the growth rate up. The annual population growth has already been declining since 2000, and only saw a brief uptick from 2010-2014.
Figure 7. Components of California Population Change 2000-2016
Finally, compare the number of persons added between 2020 and 2060 using the Dept. of Finance forecast and this alternative estimate that I created by extrapolating the current growth rate, Figure 8 below. Which better fits the data and observations of current affairs?
Figure 8. Comparison of Forecasts for Persons Added in California 2020-2060.
My prediction: California population will shrink by 1 million people in the 2030s, 2 million in the 2040s, and shrinking will continue thereafter. To put this another way, this prediction is a forecast for permanent contraction of the economy, electricity consumption, gasoline consumption, housing construction, mortgage loans, auto sales, student loans, demand for professional services, air travel, vehicle miles traveled, need for schools, college enrollment, and GHG emissions starting in the 2030s (about 15 years from now).

Now what do you want to invest money in? Health care for the elderly! :-)

What factors would need to change to make this prediction wrong?

The decline in birthrates is happening all over the industrialized world, with Japan in the lead. Read more about demographics at Econimica. Why are birthrates declining? Theory: Because of diminishing returns on human labor. Young people are finding that it requires two incomes to support a lower standard of living than their parents had from one income. The fundamental reason for diminishing returns on labor may be that industrialized countries have been unable since the 1970s to continue leveraging increasingly inexpensive energy sources (petroleum) to improve labor productivity, and this decline has increased since 2000. At the same time, cost of living has shot up due to asset price inflation. Read more about the energetic basis for falling returns on human labor at Our Finite World.

Economists and banks seem upset about demographic slowdown and reversal. There will be fewer consumers to take out new loans and support asset prices.

My take is this: Population contraction is great news! Fantastic! Whew, what a relief that it's finally here! Thank goodness that crazy petroleum interlude is coming to an end. We'll be able to spend more time with family. There may be a resurgence of wildlife. Air quality may improve. The stars may shine brighter.