Declining U.S. Internal Migration Rates

Declining U.S. Internal Migration Rates

Mobility periodTotal, 1 year old and

overSame residence

(non-movers) Total movers

Different residence in the United States

Total Same county

Different county

Total Same State Different State

NUMBER                

2009-2010 300,074 262,534 37,540 36,594 26,017 10,577 6,252 4,326

2008-2009 297,182 260,077 37,105 36,017 24,984 11,034 6,374 4,660

2007-2008 294,851 259,685 35,167 34,022 23,013 11,009 6,282 4,728

2006-2007 292,749 254,068 38,681 37,490 25,192 12,298 7,436 4,862

2005-2006 289,781 249,945 39,837 38,541 24,851 13,690 8,010 5,679

2004-2005 287,148 247,261 39,888 38,023 22,736 15,287 7,847 7,441

2003-2004 284,367 245,372 38,995 37,723 22,551 15,172 7,842 7,330

2002-2003 282,556 242,463 40,093 38,824 23,468 15,356 7,728 7,628

2001-2002 278,160 237,049 41,111 39,548 23,712 15,836 8,066 7,770

2000-2001 275,611 236,605 39,007 37,251 21,918 15,333 7,550 7,783

1999-2000 270,219 226,831 43,388 41,642 24,399 17,242 8,814 8,428

1998-1999 267,933 225,297 42,636 41,207 25,268 15,939 8,423 7,516

1997-1998 265,209 222,702 42,507 41,304 27,082 14,222 7,867 6,355

1996-1997 262,976 219,585 43,391 42,088 27,740 14,348 7,960 6,389

1995-1996 260,406 217,868 42,537 41,176 26,696 14,480 8,009 6,471

1994-1995 258,248 215,931 42,317 41,539 27,908 13,631 7,888 5,743

1993-1994 255,774 212,939 42,835 41,590 26,638 14,952 8,226 6,726

1992-1993 250,210 208,162 42,048 40,743 26,212 14,532 7,735 6,797

1991-1992 247,380 204,580 42,800 41,545 26,587 14,957 7,853 7,105

1990-1991 244,884 203,345 41,539 40,154 25,151 15,003 7,881 7,122

1989-1990 242,208 198,827 43,381 41,821 25,726 16,094 8,061 8,033

1988-1989 239,793 197,173 42,620 41,153 26,123 15,030 7,949 7,081

1987-1988 237,431 195,258 42,174 40,974 26,201 14,772 7,727 7,046

1986-1987 235,089 191,396 43,693 42,551 27,196 15,355 8,762 6,593

1985-1986 232,998 189,760 43,237 42,037 26,401 15,636 8,665 6,971

1984-1985 230,333 183,863 46,470 45,043 30,126 14,917 7,995 6,921

1983-1984 228,232 188,853 39,379 38,300 23,659 14,641 8,198 6,444

1982-1983 225,874 188,465 37,408 36,430 22,858 13,572 7,403 6,169

1981-1982 223,719 185,592 38,127 37,039 23,081 13,959 7,330 6,628

1980-1981 221,641 183,442 38,200 36,887 23,097 13,789 7,614 6,175

Period Interstate intercounty

1947-1948 0.030941 0.06378

1948-1949 0.030245 0.058033

1949-1950 0.026569 0.056363

1950-1951 0.035032 0.070658

1951-1952 0.034113 0.066504

1952-1953 0.036258 0.066633

1953-1954 0.032544 0.064526

1954-1955 0.031043 0.065994

1955-1956 0.031472 0.067964

1956-1957 0.031049 0.062808

1957-1958 0.033484 0.067401

1958-1959 0.029855 0.061765

1959-1960 0.031819 0.064795

1960-1961 0.032622 0.063769

1961-1962 0.031106 0.061647

1962-1963 0.03658 0.068048

1963-1964 0.032783 0.066348

1964-1965 0.032872 0.068151

1965-1966 0.033075 0.066213

1966-1967 0.034325 0.067367

1967-1968 0.036387 0.070566

1968-1969 0.033932 0.066276

1969-1970 0.035795 0.067457

1970-1971 0.034737 0.065727

1971-1972

1972-1973

1973-1974

1974-1975

1975-1976/2 0.029673 0.064015

1976-1977 0.029673 0.064015

1980-1981 0.028026 0.062584

1981-1982 0.029771 0.0627

1982-1983 0.02743 0.060348

1983-1984 0.028369 0.064454

1984-1985 0.030235 0.065166

1985-1986 0.030074 0.067455

1986-1987 0.028182 0.065635

1987-1988 0.029827 0.062532

1988-1989 0.029711 0.063065

1989-1990 0.033381 0.066878

1990-1991 0.029249 0.061614

1991-1992 0.028867 0.06077

1992-1993 0.027308 0.058384

1993-1994 0.026425 0.058744

1994-1995 0.022306 0.052942

1995-1996 0.02498 0.055898

1996-1997 0.024416 0.054832

1997-1998 0.024071 0.05387

1998-1999 0.028202 0.059808

1999-2000 0.031392 0.064222

2000-2001 0.02842 0.055989

2001-2002 0.028091 0.057253

2002-2003 0.027118 0.054592

2003-2004 0.025892 0.053593

2004-2005 0.026083 0.053585

2005-2006 0.019686 0.047455

2006-2007 0.016676 0.04218

2007-2008 0.016098 0.037483

2008-2009 0.015738 0.037265

2009-2010 0.014462 0.035359

Note:Different state 2004-05 7,441 major declines in each type 2005-06 5,679 2006-07 4,862Different county, 2005-06 8,010 same state 2006-07 7,436 2007-08 6,282Different county 2004-05 15,287 2005-06 13,690 2006-07 12,298 2008-09 11,009

Different residence 2004-05 22,736 but residential mobility 2005-06 24,851 increases 2006-07 25,192

Some Possible Reasons for the Recent Decline in U.S. Internal Migration Rates

1. The business cycleThe current recession began at the end of 2007, but

migration rates began to fall one or two years earlier. Moreover, even though the current recession has been much more serious than previous recessions, the drop in migration is somewhat smaller, according to MSW. Dismiss as a cause, according to MSW. The claim that the drop in migration is somewhat smaller during the current recession is open to question.

check 1990-92 2000-02 1988-92 1978-83

MSW extrapolate trends in migration, 1996-2006 and 1986-2006. Based on the 10-year extrapolation, migration rates would have fallen 0.23 percentage points between 2007 and 2009. Based on the 20-year extrapolation, they would have fallen by 0.25 percentage points. They actually fell by 0.36 percentage points, so MSW claim that very little of the drop remains to be explained by the recession. This type of comparison I like. Still dismiss as a cause.

2. The housing-market contraction This contraction occurred at about the same time as the drop in migration. What are the possible mechanisms?

When housing prices drop, homeowners may find themselves upside down in their mortgages (owe more than the current value of their property). They become less likely to move.

We have no specific evidence to support this hypothesis, so the rationale is very speculative.

Falling house prices cause pessimism concerning future house prices, so renters do not buy homes and owners do not move up in terms of housing quality. Migration falls. Again,very speculative with no supporting empirical evidence.

Both hypotheses suggest that homeowner migration rates should have fallen more than those of renters. In percentage point terms, neither interstate nor intercounty rates fell more for homeowners. The percentage decline was larger for homeowners, but homeowners generally have lower migration rates than renters. Dismiss as a cause.

3. Retirement migration is down

Check the data for 2009-10 relative to 1999-2000. The big decline had not yet begun in 1999-2000.

Intercounty diff 1999-20 less 2009-10

Interstate diff 1999-20 less 2009-10 Intercounty relative diff Interstate relative diff

AGE

0.0333 0.0190 0.8282 0.4730 1 to 4 years

0.0296 0.0176 0.8969 0.5319 5 to 9 years

0.0277 0.0176 0.9546 0.6077 10 to 14 years

0.0270 0.0188 0.9594 0.6689 15 to 17 years

0.0436 0.0278 1.0234 0.6529 18 to 19 years

0.0557 0.0324 0.8857 0.5143 20 to 24 years

0.0580 0.0280 0.9592 0.4626 25 to 29 years

0.0328 0.0176 0.8060 0.4332 30 to 34 years

0.0312 0.0174 0.9670 0.5387 35 to 39 years

0.0261 0.0171 0.9336 0.6140 40 to 44 years

0.0215 0.0118 1.0563 0.5769 45 to 49 years

0.0158 0.0090 0.8749 0.5000 50 to 54 years

0.0169 0.0119 0.8664 0.6112 55 to 59 years

0.0184 0.0117 0.8825 0.5614 60 to 61 years

0.0107 0.0106 0.6509 0.6438 62 to 64 years

-0.0010 0.0018 -0.0953 0.1610 65 to 69 years

0.0095 0.0074 0.7974 0.6227 70 to 74 years

0.0097 0.0078 0.9196 0.7364 75 to 79 years

0.0145 0.0080 1.2102 0.6689 80 to 84 years

0.0100 0.0036 0.9851 0.3518 85+ years

Intercounty Intrastate 1999-20

Intercounty Intrastate 2009-10

Intrastate difference

Interregional 1999-20

Interregional 2009-10

Interregional difference

AGE

1 to 4 years 0.0439 0.0296 0.0143 0.0156 0.0107 0.0048

5 to 9 years 0.0321 0.0201 0.0121 0.0125 0.0081 0.0044

10 to 14 years 0.0269 0.0169 0.0101 0.0109 0.0063 0.0046

15 to 17 years 0.0238 0.0156 0.0082 0.0105 0.0042 0.0064

18 to 19 years 0.0459 0.0302 0.0158 0.0129 0.0075 0.0053

20 to 24 years 0.0722 0.0489 0.0234 0.0228 0.0148 0.0080

25 to 29 years 0.0724 0.0423 0.0300 0.0251 0.0155 0.0095

30 to 34 years 0.0461 0.0309 0.0152 0.0146 0.0120 0.0026

35 to 39 years 0.0342 0.0204 0.0138 0.0126 0.0076 0.0051

40 to 44 years 0.0255 0.0166 0.0089 0.0084 0.0058 0.0026

45 to 49 years 0.0237 0.0139 0.0098 0.0058 0.0046 0.0012

50 to 54 years 0.0186 0.0118 0.0068 0.0051 0.0044 0.0006

55 to 59 years 0.0170 0.0120 0.0050 0.0072 0.0042 0.0031

60 to 61 years 0.0162 0.0095 0.0067 0.0066 0.0051 0.0015

62 to 64 years 0.0118 0.0116 0.0001 0.0065 0.0029 0.0036

65 to 69 years 0.0080 0.0108 -0.0028 0.0051 0.0043 0.0008

70 to 74 years 0.0089 0.0068 0.0021 0.0061 0.0020 0.0040

75 to 79 years 0.0092 0.0072 0.0019 0.0054 0.0013 0.0042

80 to 84 years 0.0114 0.0048 0.0065 0.0056 0.0017 0.0038

85+ years 0.0124 0.0060 0.0065 0.0064 0.0045 0.0019

All age classes are down. Retirement migration is down, and it is down slightly more than other age classes. However, in itself, retirement migration is not of sufficient magnitude to cause the overall rate to fall so substantially. Dismiss as a basic cause, but could be a slight contributing cause.

4. Migration to the West is down substantially.

5. Migration to or from other regions is down substantially.

Let’s look at the in- and out-migration experiences of the various regions before and after the decline in migration rates.

In-migration (thousands) toNortheast Midwest South West

1999-2000 363 722 1,258 764

2009-2010 246 414 897 631

(-32.2%) (-42.7%) (-28.7%) (-17.4%)

Out-migration (thousands) from

1999-2000 615 640 1,031 819

2009-2010 376 637 698 477

(-38.9%) (-0.5%) (-32.3%) (-41.8%)

4. In-migration to the West is down less than in-migration to the other regions. Dismiss as a cause.

5. However, out-migration from the West has fallen more than out-migration from any other region. Why might this have happened? Out-migration from the Midwest has hardly changed. What is happening in the Midwest to make its out-migration experience so different than the other regions?

Counterfactual based on 1980-1985 propensities; actual is 2000-2005

From To Counterfactual ActualNortheast Midwest 415 288 -30.6% South 1,524 1,512 - 0.8 West 518 514 - 0.8Midwest Northeast 365 228 -37.5 South 2,177 1,382 -36.5 West 1,274 719 -43.6South Northeast 873 710 -18.7 Midwest 1,145 930 -18.8 West 1,375 1,062 -22.8West Northeast 353 352 - 0.3 Midwest 991 691 -30.3 South 1,607 1,230 -23.5 12,617 9,618 0.0473 0.0361

Something appears to be going on in the Midwest. Have the most mobile already departed? Has migration from rural areas diminished because the young and most mobile have already moved?

6. The system has moved closer to being an equilibrium system.If this is the case, then presumably people have sorted themselves in such a way that migration is being driven by shocks to the system and by long-term trends that change the spatial pattern of economic opportunities and demands for location-specific amenities.

Discuss the neoclassical model: movement of people and capital.

This could be a cause, but I am skeptical because the migration-rate changes occurred over a very short span of time. Interregional migration has been occurring since the founding of the nation, and if this factor were driving the changes, we would, I think, expect the movement toward long-term equilibrium to have occurred gradually over a long period of time.

7. Demographic change

It is possible that changes in the age structure of the U.S. population have contributed to declining internal migration rates. If this were to have occurred, it most likely would have been due to population maturing out of the ages for which migration rates are highest (18-39). To test this hypothesis, I did a counterfactual. I distributed the 2009 population as it was age distributed in 1999. Then I applied 2009-10 migration rates to the hypothetical 2009 population. The result was almost no change in the national migration rate. Dismiss as a cause.

However, over longer periods of time, demographics do matter. They also can matter over shorter periods of time, such as during the 1970s and 1980s when the baby boom was maturing into the most highly mobile age classes and later out of these age classes. The baby boom has now aged into relatively low migration-rate ages, and thus over short periods of time is unlikely to have much effect on national rates.

8. Method of measuring migration.MSW discuss several migration data sources. In their work, they aggregate from microdata in order to avoid various measurement issues. This is a good practice. But does it avoid the so-called “hot deck” imputation procedures that some think have biased the migration rates down so severely in recent years? Open question.

Bottom line: ????