Win Some; Lose Some

The U.S. Census released its annual estimates of state population change late last week. As usual the papers in the state took the opportunity to paint a rosy picture of the changes. The Reformer, for example, characterized the changes as"relatively small" and qualified its admission of the overall decline by quoting  Will Sawyer of the Vermont State Data Center who says "it's too soon to tell whether the new estimates reflect a trend."

Well, if you're  looking at only one year of data, I suppose it is difficult to see a trend. That's why, when I'm looking for trends, I look at the historical data, too. Lucky for us, the U.S. Census provides lots of historical data. And when I look at the recent data alongside the last few years, I do see a trend.

Looking at the net population change - even if done by county - doesn't show the whole picture. To better understand what's happening you need to look at the components of change.  That is, you need to break down the total change into the sources of change. The U.S. Census provided data for three components: Natural Change, (births and deaths), International Change, (immigrants), and Domestic Migration, (US residents).

The table below is a compilation I made of the last few years of Census data for Vermont. The most striking thing about this data is the clear trend of domestic migration out of the state. This is significant because it speaks to the truth of Vermonter's leaving.  This says much more about the desirability of Vermont as a place to live than does people who have never lived here deciding to move in.

The clear trend is that people who have experienced the state are leaving ... and at an increasing rate.

Comments

[Cairn Cross]

Greg am I doing my math right when I estimate that Vermont has 600,000 people and approximately 2,000 left last year (not counting the fact that there were natural changes and international migration which offset these folks leaving)and that 2000 divided by 600,000 is about 3/10ths of 1%? Just checking to make sure that I am right with my math.

[Greg Decker]

Cairn, that sounds about right. The Census states the relative changes in the terms of people per 1000 population. Vermont is listed as having an internal migration of -2.9 Comparatively speaking, this places Vermont as number 14 for the rate of decline coming in behind:

.New York -9.8
.Rhode Island -9.5
.Michigan 9.4
.New Jersey -8.0
.Hawaii -7.6
.California -7.2
.Maryland -6.5
.Connecticut -5.5
.Massachusetts -5.5
.District of Columbia -5.4
.Illinois -4.7
.Ohio -4.5
.Alaska 4.1

The large scale trend in a decline in population across the north east with Vermont fairing better than most in the region.

Interestingly, the states losing people all rank near the bottom of the Public Policy Institutes economic freedom index.

[Jon Harrison]

Your last point, Greg, is obviously no coincidence. I also think you made a very shrewd point in highlighting the fact that those who know the state are leaving; any influx we get is made up of people who don't really know the Vermont score (you really have to live here for a while to understand).

[Cairn Cross]

Jon: Not to be picky but what statistics would show that those who don't know the state are leaving rather than as you say "those who know the state are leaving"? If you could say that the preponderance of those who left had lived here for 20 years that's one thing but I don't think that is really how to interpret these statistics. The truth is the statistics don't tell us how long people lived here nor how well they "knew the state" before leaving. I think the stats Greg reports are one thing. I think his interpretation is anecdotal though rather than factual.

[Greg Decker]

Cairm your correct, as usual, that the data doesn't proves the claims. Unfortunately, there are few instances where data will prove anything. Its always up to the researcher to make some hypotheses about what's really happening. The hypotheses about the demographics of Vermont put forth by the Tiger is well know to our regular reader (yourself included) This single example of the internal migration data should not bee seen as proof of that hypothesis but rather, as additional supporting evidence.

One important piece of data supporting my interpretation that was left unstated is an assessment of the states who are gaining in population. Texas, NC and SC combined have gained about half of all people who left other states. When I Google these states I find lots of articles about growth and opportunity. A anecdotal way of interpreting this would be to say people are not leaving Vermont to escape the lack of opportunity they are leaving to take advantage of the apparent greater opportunity elsewhere.

[Cairn Cross]

Greg: Just to play devil's advocate might the out migration in recent years have something to do with the first wave of baby boomers retiring and leaving Vermont for warmer climes rather than say people in their early or peak earning years leaving for opportunities. The study below published by the Fed in Boston has some interesting slides. Take a look at the Chittenden county MSA slide and where people are going to and where they are coming from when they move from/to Chittenden County. There are some usual suspects but some counter intuitive data emerges.

http://www.bos.frb.org/economic/neppc/dp/2006/neppcdp0601.pdf

[Jon Harrison]

Cairn, you are indeed correct. I plead guilty to shooting from the lip on this one.

[Greg Decker]

Cairn, that's a possibility. However, the experience in my community has been a pattern of baby boomer's moving in from out of state for their first retirement. As a member of the town's planning commission I (we) have a pretty good understanding of the local changes. How these changes map to the rest of the state is pure conjecture.

The paper you cited was prepared using the same Census data sources I've been reviewing. If your interested, the 2007 data in now available which shows the trend is accelerating.

Another observation is that if you map the migration trends using red and blue you get a close approximation to the 2004 election results which is likewise a close approximation to the economic freedom index. While your hypothesis of warmth seekers is possible, there are outliers in the data such as Utah that is not warm but has been attracting people as an astonishing rate.

[Jon Harrison]

Utah of course is a special case. It is perceived (correctly?) as a stable, relatively crime-free place. This is in large part because of the high percentage of Mormons in its population. Rightly or not, the Mormon lifestyle is perceived as promoting "family values" and social stability. At the same time, Utah offers many of the economic advantages typical of blue states. So, many people are now choosing it over Arizona and Florida. And it's as easy to get to Vegas from Utah as it is from Arizona.