Where Have All the People Gone…and Can We Get Them Back?
(Kurt Metzger, Feb. 1, 2016)
When one attempts to understand Michigan migration patterns – movement into and out of the state, not within (that’s another blog) – we can look at a variety of sources – both public and private. Let us start with our federal government sources.
The Census Bureau produces annual population estimates which incorporate the components of population change (births, deaths and migration – domestic and international). These estimates track population at mid-year, the most recent being July 1, 2015. The latest estimate, released in December of 2015, showed that Michigan experienced a slight population increase between 2014 and 2015, representing the fourth straight year of growth. The increase of 6,270 residents was the smallest in this 4-year cycle, a cycle that reversed seven straight years of population loss and was generated by the relatively low, historically, net out-migration to other parts of the country. The smaller than anticipated gain was driven by a surprising increase in net domestic out-migration from 28,679 in 2013-14 to 38,911 in 2014-15. Figure 1 shows net domestic migration from 2000 through 2015. Michigan’s population fortunes truly depend on this variable more than any other.[1] The volume tends to track with employment trends – as unemployment rises, as it did in the decade of the 2000’s, the numbers grow – hitting 109,257 in 2007-08. The opposite trend was in play up until last year – employment was rising and fewer people were leaving. However, the increase in net out-migration of 10,000 persons came as a quite a surprise as the state trumpeted an employment rate in sync with the national average. We will have to leave the possible reasons for another blog. The major recipients of our former residents were Ohio, Florida, Arizona, North Carolina and Texas. We have Alaska, Indiana, Arkansas, New York and Massachusetts to thank for their contributions.
So where is everyone going? To answer that question I move to my other go-to public source – the Internal Revenue Service. IRS matches income tax forms from year to year to determine migration. They produce annual files that identify flows of households, persons and incomes between states. (They also produce annual county-to-county flows that allow you to better identify local origins and destinations). Their most recent file, covering 2013-14, was released and then pulled back. While I am not sure what the issues are, I can tell you what the data say. Recipient states numbered 30, including the District of Columbia. Serving as the major recipients of our former residents were, in order, Texas, Florida, Georgia, California, Arizona, Tennessee, and the Carolinas – South then North. While we had 20 donor states who gave more than they received, their contributions tended to be much smaller than what we provided to the recipients. Major thanks go out to Illinois (not everyone is going to Chicago), New York, New Jersey and Pennsylvania.
A source that I have found to be quite interesting, but by no means comprehensive, is that of national moving companies. Several of them – United, Atlas and Allied – have made it a point of tracking the number of shipments they carry in and out of states. While we have no idea of the ultimate destination of these shipments, we can use the data as a gauge of a state’s success, or lack thereof, in attracting new households. The interesting conclusion to be drawn from these sources is that, while each is recording better days for Michigan over the last couple years, the three differ in their results. This could just be an indication that certain movers are stronger in different states.
Let us begin with United Van Lines which has been tracking migration patterns annually since 1977. Their 2015 report has 52.2 percent of their 5,331 Michigan-related moves being outbound. This represents an improvement over recent years (2012 – 57.5% out; 2013 – 53.8%; 2014 – 55.4%), and a rise in ranking from 36th to 27th. Michigan ranked well above its Midwest neighbors of Ohio, Indiana, Illinois and Wisconsin. Based on the average household size in Michigan, United moved about 511 more people out of Michigan than in.
The 2015 Atlas Van Lines Migration Patterns study marks their 23rd year of tracking state-to-state patterns. They must be favored by Michigan wannabes (and a friend of state government) because they have been tracking positive toward movers into Michigan over the last 4 years. As recently as 2009, moves to Michigan represented just 39.9 percent of the total. In the latest study, which recorded the highest share of in-movers over the 12 years for which data are available, their 3,528 Michigan-related moves favored inbound travelers at 54.6 percent. This was up from 51.4 percent in 2014, and equates to a net in-migration count of about 807 new residents.
We now look to Allied Van Lines to break the inbound/outbound tie. Allied has produced its 48th annual magnet states report for 2015, and provides the bleakest look of all for Michigan. Of the 3,188 Michigan-related moves tracked by Allied, only 38.2 percent represented in-migrants (even worse than the low point for Atlas in 2009). While the share of moves in was down from 39.8 percent in 2014, Michigan’s rank rose to 45th, up from 47th in both 2013 and 2014. The bottom line, based on average household size, was an estimated loss of 1,868 residents.
Taken together, we can account for only 1,572 Michigan out-migrants. We can assume that the additional 37,000 or so that the Census Bureau estimated used other moving companies or did it themselves with friends and a U-Haul.
This brings me to my last non-governmental migration indicator – U-Haul rates between Detroit and other metros around the country. While this doesn’t exactly help us understand statewide migration patterns, it does help us understand where the Detroit area stands in the migration cue. This is important because Detroit must become a true magnet if Michigan is going to prosper.
I started tracking the cost of renting a 17 ft. U-Haul truck back in 2010. What I found was that the cost of going from Detroit to anywhere else was more expensive than the return trip. [For instance, it cost $714 to pick up a U-Haul in Orlando and drive it to Detroit. It cost $1,445 to pick up that same 17 ft. truck in Detroit and drive it to Orlando.] While I would love to think that this was an organized plan by U-Haul dealerships to keep Detroiters from leaving, my conclusion was that there was a dearth of trucks in Detroit – many had already left and the returns were much fewer – such that scarcity drove costs up.
I have made it a habit to check prices at the beginning of each year (some folks just don’t have enough to keep them busy), and I find that January 2016’s rates offer a mixed, but hopeful, result. I first must address an anomaly in the 2016 data. It appears that U-Haul is no longer quoting prices for 17 ft. trucks on any of our routes except returns from Charlotte and Orlando. As a result, Table 1 will utilize the 15 ft. truck price for all cities. While the total cost may be slightly affected, comparisons on differential should still hold up.
The Census Bureau produces annual population estimates which incorporate the components of population change (births, deaths and migration – domestic and international). These estimates track population at mid-year, the most recent being July 1, 2015. The latest estimate, released in December of 2015, showed that Michigan experienced a slight population increase between 2014 and 2015, representing the fourth straight year of growth. The increase of 6,270 residents was the smallest in this 4-year cycle, a cycle that reversed seven straight years of population loss and was generated by the relatively low, historically, net out-migration to other parts of the country. The smaller than anticipated gain was driven by a surprising increase in net domestic out-migration from 28,679 in 2013-14 to 38,911 in 2014-15. Figure 1 shows net domestic migration from 2000 through 2015. Michigan’s population fortunes truly depend on this variable more than any other.[1] The volume tends to track with employment trends – as unemployment rises, as it did in the decade of the 2000’s, the numbers grow – hitting 109,257 in 2007-08. The opposite trend was in play up until last year – employment was rising and fewer people were leaving. However, the increase in net out-migration of 10,000 persons came as a quite a surprise as the state trumpeted an employment rate in sync with the national average. We will have to leave the possible reasons for another blog. The major recipients of our former residents were Ohio, Florida, Arizona, North Carolina and Texas. We have Alaska, Indiana, Arkansas, New York and Massachusetts to thank for their contributions.
So where is everyone going? To answer that question I move to my other go-to public source – the Internal Revenue Service. IRS matches income tax forms from year to year to determine migration. They produce annual files that identify flows of households, persons and incomes between states. (They also produce annual county-to-county flows that allow you to better identify local origins and destinations). Their most recent file, covering 2013-14, was released and then pulled back. While I am not sure what the issues are, I can tell you what the data say. Recipient states numbered 30, including the District of Columbia. Serving as the major recipients of our former residents were, in order, Texas, Florida, Georgia, California, Arizona, Tennessee, and the Carolinas – South then North. While we had 20 donor states who gave more than they received, their contributions tended to be much smaller than what we provided to the recipients. Major thanks go out to Illinois (not everyone is going to Chicago), New York, New Jersey and Pennsylvania.
A source that I have found to be quite interesting, but by no means comprehensive, is that of national moving companies. Several of them – United, Atlas and Allied – have made it a point of tracking the number of shipments they carry in and out of states. While we have no idea of the ultimate destination of these shipments, we can use the data as a gauge of a state’s success, or lack thereof, in attracting new households. The interesting conclusion to be drawn from these sources is that, while each is recording better days for Michigan over the last couple years, the three differ in their results. This could just be an indication that certain movers are stronger in different states.
Let us begin with United Van Lines which has been tracking migration patterns annually since 1977. Their 2015 report has 52.2 percent of their 5,331 Michigan-related moves being outbound. This represents an improvement over recent years (2012 – 57.5% out; 2013 – 53.8%; 2014 – 55.4%), and a rise in ranking from 36th to 27th. Michigan ranked well above its Midwest neighbors of Ohio, Indiana, Illinois and Wisconsin. Based on the average household size in Michigan, United moved about 511 more people out of Michigan than in.
The 2015 Atlas Van Lines Migration Patterns study marks their 23rd year of tracking state-to-state patterns. They must be favored by Michigan wannabes (and a friend of state government) because they have been tracking positive toward movers into Michigan over the last 4 years. As recently as 2009, moves to Michigan represented just 39.9 percent of the total. In the latest study, which recorded the highest share of in-movers over the 12 years for which data are available, their 3,528 Michigan-related moves favored inbound travelers at 54.6 percent. This was up from 51.4 percent in 2014, and equates to a net in-migration count of about 807 new residents.
We now look to Allied Van Lines to break the inbound/outbound tie. Allied has produced its 48th annual magnet states report for 2015, and provides the bleakest look of all for Michigan. Of the 3,188 Michigan-related moves tracked by Allied, only 38.2 percent represented in-migrants (even worse than the low point for Atlas in 2009). While the share of moves in was down from 39.8 percent in 2014, Michigan’s rank rose to 45th, up from 47th in both 2013 and 2014. The bottom line, based on average household size, was an estimated loss of 1,868 residents.
Taken together, we can account for only 1,572 Michigan out-migrants. We can assume that the additional 37,000 or so that the Census Bureau estimated used other moving companies or did it themselves with friends and a U-Haul.
This brings me to my last non-governmental migration indicator – U-Haul rates between Detroit and other metros around the country. While this doesn’t exactly help us understand statewide migration patterns, it does help us understand where the Detroit area stands in the migration cue. This is important because Detroit must become a true magnet if Michigan is going to prosper.
I started tracking the cost of renting a 17 ft. U-Haul truck back in 2010. What I found was that the cost of going from Detroit to anywhere else was more expensive than the return trip. [For instance, it cost $714 to pick up a U-Haul in Orlando and drive it to Detroit. It cost $1,445 to pick up that same 17 ft. truck in Detroit and drive it to Orlando.] While I would love to think that this was an organized plan by U-Haul dealerships to keep Detroiters from leaving, my conclusion was that there was a dearth of trucks in Detroit – many had already left and the returns were much fewer – such that scarcity drove costs up.
I have made it a habit to check prices at the beginning of each year (some folks just don’t have enough to keep them busy), and I find that January 2016’s rates offer a mixed, but hopeful, result. I first must address an anomaly in the 2016 data. It appears that U-Haul is no longer quoting prices for 17 ft. trucks on any of our routes except returns from Charlotte and Orlando. As a result, Table 1 will utilize the 15 ft. truck price for all cities. While the total cost may be slightly affected, comparisons on differential should still hold up.
As can be seen, price differentials between Detroit and metros in the South and West, while still rather large, are slowing closing, while those for areas in the Midwest and Northeast are showing great improvement. These results fit well with the patterns we see in the Census data. If we can continue to reverse rates to warmer areas, we will know we are on a trajectory of growth. Let’s take a look.
Well, I have run out of sources but am always on the lookout for more. The Census Bureau has been slowly developing some migration tables from the American Community Survey that help us look at some of the demographic characteristics of movers. The IRS has also changed their methodology such that they can now produce state level data that looks at movement by selected age categories and income ranges. I will be following up with both of these in the near future.
The bottom line is, Michigan has always been a net loser in the domestic migration sweepstakes. In order to grow, we must work to see that loss continues to decrease – and to try to understand what the latest numbers mean. Regardless of the final numbers, the most important concept is the demographic characteristics of those we are attracting. While we can always expect retirees to choose warmer climates (and our aging demographic will exacerbate this trend), we must work hard to retain and attract young, educated individuals.
We have a great story to tell. We just need to understand how best to get people to believe it.
[1] The others are natural increase due to births and deaths and immigration. Each of these variables tends to have much less variability than domestic migration.
Kurt Metzger is the Mayor of Pleasant Ridge and the founder of Data Driven Detroit. Metzger has provided insights into voter turnout for several elections. Check out previous stories at:
https://oaklandcounty115.com/2015/12/01/analysis-marriage-and-divorce-trends-in-oakland-county/
https://oaklandcounty115.com/2015/11/05/analysis-of-voter-turnout-in-south-oakland-county-2015/
https://oaklandcounty115.com/2015/08/05/pontiac-election-a-perfect-example-of-why-voter-turnout-matters/
https://oaklandcounty115.com/2015/03/17/guest-view-lack-of-voting-not-registration-is-the-problem/
https://oaklandcounty115.com/2014/11/08/huntington-woods-tops-pleasant-ridge-with-66-voter-turnout/
https://oaklandcounty115.com/2014/10/29/a-look-at-low-voter-turn-out-rates-in-the-aug-5-primary-2/
Well, I have run out of sources but am always on the lookout for more. The Census Bureau has been slowly developing some migration tables from the American Community Survey that help us look at some of the demographic characteristics of movers. The IRS has also changed their methodology such that they can now produce state level data that looks at movement by selected age categories and income ranges. I will be following up with both of these in the near future.
The bottom line is, Michigan has always been a net loser in the domestic migration sweepstakes. In order to grow, we must work to see that loss continues to decrease – and to try to understand what the latest numbers mean. Regardless of the final numbers, the most important concept is the demographic characteristics of those we are attracting. While we can always expect retirees to choose warmer climates (and our aging demographic will exacerbate this trend), we must work hard to retain and attract young, educated individuals.
We have a great story to tell. We just need to understand how best to get people to believe it.
[1] The others are natural increase due to births and deaths and immigration. Each of these variables tends to have much less variability than domestic migration.
Kurt Metzger is the Mayor of Pleasant Ridge and the founder of Data Driven Detroit. Metzger has provided insights into voter turnout for several elections. Check out previous stories at:
https://oaklandcounty115.com/2015/12/01/analysis-marriage-and-divorce-trends-in-oakland-county/
https://oaklandcounty115.com/2015/11/05/analysis-of-voter-turnout-in-south-oakland-county-2015/
https://oaklandcounty115.com/2015/08/05/pontiac-election-a-perfect-example-of-why-voter-turnout-matters/
https://oaklandcounty115.com/2015/03/17/guest-view-lack-of-voting-not-registration-is-the-problem/
https://oaklandcounty115.com/2014/11/08/huntington-woods-tops-pleasant-ridge-with-66-voter-turnout/
https://oaklandcounty115.com/2014/10/29/a-look-at-low-voter-turn-out-rates-in-the-aug-5-primary-2/