# Simulation To Reproduce Fallbrook, CA Area Demographics

The observed demographics of the Fallbrook area show some significant differences from the demographics of San Diego County, explored in that link.

This page presents a simple detailed numeric simulation I have performed that demonstrates that a few simple plausible effects can reproduce the observed data. However, just because a model can reproduce observed data does not establish that the model is correct. It would require a lot of detailed investigation in order to prove that these are the true causes.

Nonetheless, my experience as a Fallbrook resident leads to the strong belief that the effects simulated here must play an important role in producing the observed Fallbrook demographics. Certainly the precise numbers found must be considered correct only to within some large uncertainty.

The simulation started with the San Diego County demographics (green curves in the plots below), and tried to match the observed Fallbrook demographics (blue curves), for both the age and ethnic distribution of the population.

 (Click on graph for bigger and better image.) (Click on graph for bigger and better image.)

My simulation modified the San Diego County demographics by:

• adding Hispanic workers with ages spread uniformly between 20 and 40, with a smaller number up to age 55;

• adding children of those workers at the rate of 0.48 children per worker, with most uniformly spread between ages 0 and 15 and a lesser number between ages 15 and 20;

• subtracting Fallbrook residents between the ages of 18 and 50 (which is equivalent to adding such residents to the San Diego County curve); and

• adding Fallbrook residents past the age of 50. I didn't attempt to model the Hispanic percentage of these residents since it is arbitrary how to allocate the Hispanic percentage of native Fallbrook seniors versus the Hispanic percentage of seniors relocating to Fallbrook.

The plot on the left below show the components of the model needed to reproduce the data, as well as the final simulated demographic curve for Fallbrook. The components are given in the same units as the plots above, as percentage of the total population within a 5 year age bracket. The gray curve is the input San Diego County curve, which becomes the blue curve by adding the other plotted components. Note that because the demographic curves are normalized so that the integral is 100%, the final curve has been renormalized after the components were added, so that it is slightly different than just adding the curves directly.

The plot on the right shows the simulated Hispanic percentage (green curve) compared to the actual Fallbrook percentage (blue curve). Note that the model made no attempt to match the Hispanic percentage past age 60, although it would be easy, but arbitrary, to do so.

 (Click on graph for bigger and better image.) (Click on graph for bigger and better image.)

Let's consider these components one at a time.

"Workers" and their children. The only component of the model above that affects the Hispanic percentage of the population is the added "workers" and their children. For simplicity, I assumed a uniform age distribution between ages 20 and 40, with smaller numbers at older ages. I then simply adjusted the total number of these workers to fit the Hispanic percentage of the Fallbrook population between ages 20 and 40. The plot below on the right shows that this simple model is able to change the San Diego County Hispanic percentage into the Fallbrook percentage for those ages.

Some of those workers have to have children. Hence I continued my simplistic model by assuming an average number of children per worker, and assuming those children had a uniform age distribution between ages 0 and 15. I then adjusted that average number to reproduce the Hispanic percentage of the population with ages less than 15. Again, the plot on the right shows that this simple model works well. (Recall I made no attempt to match the Hispanic percentage past age 55.)

So with only two simple adjustable parameters, the total number of workers and the average number of children per worker, the entire Hispanic demographics below age 50 was matched.

Amazingly enough, the addition of these children then automatically caused the demographics for children under the age to 20 to match the Fallbrook curve, without further modification, a sign of confidence in the correctness of this model.

The number of workers needed is a total of 6% of the population, or about 2,500 workers. Fallbrook has 81,111 acres, so this works out to 3 workers per 100 acres. In my neighborhood, roughly half the people have a worker who comes one day a week to help tend their 2 acre properties. If the workers work 5 days a week on average, this is a "density" of 5 workers per 100 acres. Since San Diego County as a whole must have some number like 1-2 workers per 100 acres on average, the expected excess in Fallbrook, due to our large parcel sizes and equivalent need for agricultural workers, is very similar to the number derived by the model.

Migration from and to Fallbrook. The other two components operationally work in the model similarly as a percentage of people to subtract (the out-migration between ages 18 and 50) or a percentage of people to add (the in-migration past the age of 50). Therefore I simply subtracted or added people in each 5 year bin to best match the Fallbrook curve. So even though I described two components above, I in fact did not attempt to model each component smoothly or separately and simply derived the change needed in each bin.

Note the very smooth curve produced by the adjustments needed for each bin ("migration" and "retirees" in the plot on the left above). The percent of the population that must be subtracted is essentially a constant percentage between the ages of 20 and 40, exactly what would be expected by the idea that Fallbrook residents leave Fallbrook at the age of 18 to seek employment elsewhere. Beginning at age 40, in-migration begins to mitigate the loss of those people, until at age 50-55, in-migration has balanced the loss. Past the age of 55, many people are relocating to Fallbrook.

These results again match my knowledge of Fallbrook. My wife and I personally migrated to Fallbrook at the age of 44; I know several residents of my neighborhood who relocated to Fallbrook at the age of 40 or older; and I know of quite a few retirees who relocated to Fallbrook past the age of 55.

The plot on the right below expresses these components as a percentage of each age bracket. It shows that the net loss of the Fallbrook population between ages 20 and 40 is almost 30% . Because the "workers" have added almost 20% of this age bracket, the net loss of the non-"worker" Fallbrook population approaches 50%. Because some people ages 20-40 do indeed relocate to Fallbrook (school teachers, for example), this implies that well over 50% of native Fallbrook residents leave after age 18, as expected.

The plot also shows that ~40% of the people over age 70 in Fallbrook are people relocating to Fallbrook, over and above those that relocate to San Diego County as a whole. Since undoubtedly a number of retirees relocate to San Diego County as a whole, at least half of all the retirees in Fallbrook probably have relocated to Fallbrook from elsewhere. This again matches my personal experience in that at least half of the people over age 60 are not Fallbrook natives.

So how well does this plausible model match the data? The plot on the left shows the comparison. It should come as no surprise that the model matches the data for ages above 20, since I have tweaked the model bin by bin to do so. The belief in the model for those ages must then come out of the reasonableness of the derived model parameters, which I have shown above to be very plausible. The agreement below age 20 is entirely a success of the model, since no tweaking was done to match those data at all.

 (Click on graph for bigger and better image.) (Click on graph for bigger and better image.)

## Conclusions

A very plausible model of migration of people to and from Fallbrook is able to explain the demographics of Fallbrook exceptionally well. Although this model is able to quantify these effects, it must be kept in mind that this is an extremely simplistic model which is probably only reliable in its basic outline and order of magnitude for its numbers. Conclusions from the model:

• 6% of the total Fallbrook population are "workers" who have migrated to Fallbrook, attracted by agricultural jobs and work tending the estates of Fallbrook. These workers make up nearly 30% of the Fallbrook population between the ages of 20 and 40. The children under age 18 of these workers add another 3% of the total population, and around 10% of their age bracket.

• Over 50% of the native Fallbrook population leaves Fallbrook at around age 20 to take jobs elsewhere. Except for the "workers", no significant net migration to Fallbrook occurs until about age 40, leaving a significant deficit of people with ages 20-40.

• The result is a net loss of almost 30% of the Fallbrook population between ages 20 and 40.

• Beginning at around age 40, migration into Fallbrook begins to "fill in" for the missing native population, until at around age 50, migration in has replaced the missing native population.

• Migration to Fallbrook has almost doubled the number of people over age 70 in Fallbrook.

Please note that I have presented and analyzed the Hispanic percentage of the population solely because these are fascinating data, and no bias or evil intent is present! Note that the Hispanic percentage was key to understanding some of the effects in the demographic curves.

This article was inspired by the discussion of Jason Anderson, who authored the SANDAG Report on Fallbrook commissioned by the Fallbrook Chamber of Commerce.

Go To: