Mar 12: Demography, Labor Migration, Displacement

Mar 12: Demography, Labor Migration, Displacement

Tiebout, Charles M. "A Pure Theory of Local Expenditures." Journal of Political Economy, Vol. 64, No. 5, October 1956, pp. 416-424. [Library reserves]

Frey, William H. Immigration and Internal Migration "Flight from US Metropolitan Areas: Toward a New Demographic Balkanization." Urban Studies, Vol. 32, No. 4-5, May 1995, pp. 733-757. [Library reserves]

Florida, Richard. 2002. Bohemia and economic geography, Journal of Economic Geography 2 (Jan): 55-71. [c-tools "Resource" section]

Florida, Richard. 2009. How the Crash will Reshape America. The Atlantic Monthly; March. [online]

see also:

Myers, Dowell and Lee Menifee . "Population Analysis," in The Practice of Local Government Planning, 3rd edition, edited by Charles J. Hoch, Linda C. Dalton and Frank S. So International City/County Management Association, 2000, pp. 61-86. [Library reserves]

excerpts from Bill Bishop, The Big Sort [link]

links:US Census:Geographical Mobility/Migration • Migration Data and Reports • migration tables in the 2009 Statistical Abstract • United Van Lines 2008 Migration Studythe American Moving & Storage AssociationCS Monitor: "Patchwork Nation"Andrea Coombes, Retirees Who Relocate Often Opt For Homes in Metropolita Areas, Wall Street Journal, March 31, 2006. [link]

What is demography?

FertilityMortalityeconomic demographymigration

- Be born- give birth - migrate- (attract a migrant)- die

time

spac

e

birth

death

Imagine a two time-space dimensional world (one time dimension, one space dimension)

migration migrationmigration

migration

migration

migration

migration

college

Comes HomeAftercollege

Moves to another city for job

Moves to retirement

community

birth

The next generation is born…

time

spac

e

birth

death

Imagine a two time-space dimensional world (one time Dimension, one space dimension)

migration migrationmigration

migration

migration

migration

migrationbirth

time

spac

e

birth

death

Do you need to know the path or just the start and finish?

birth

Demography: the study of life EVENTS.Individuals are “at risk” of experiencing these events.

The frequency of events: events happen at rates:

number of events of a specific type in a given time period

----------------------------------------------------------

Number of people at risk of experiencing that type of event in the given time period.

Rate =

The frequency of events: events happen at rates:

occurrence

--------------

exposureRate =

EXAMPLE: CRUDE DEATH RATE

Total number of deaths in a given year

----------------------------------------------------------

Total population in that yearCDR =

EXAMPLE: CRUDE BIRTH RATE

Total number of births in a given year

----------------------------------------------------------

Total population in that yearCBR =

EXAMPLE: CRUDE BIRTH RATE (per 1,000 pop)

Total number of births in a given year

-------------------------------------------------

Total population in that yearCBR = * 1,000

Total number of births in a given year

-------------------------------------------------

Total population in that yearCBR =

* 1,000

A comparison of:

CRUDE VERSUS AGE-SPECIFIC RATES

General Fertility Rate

Age-Specific Fertility Rate

Number of Births_____________Number of Women ages 15 - 49

Number of Births to womenages 20 - 24______________Number of Women ages 20 - 24

Total number of births in a given year

_______

Total population in that year

Crude Birth Rate

“The Demographic Transition”

http://www.geo.oregonstate.edu/classes/geo300/trans/demot.jpg http://www.populationaction.org/resources/factsheets/images/demographicTransit_fs.jpg

TOTAL FERTILITY RATE (TFR)

TOTAL FERTILITY RATE (TFR)

Source: CDC http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5449a5.htm

Fecundity: The physiological capacity of a woman to produce a child [a theoretical upper limit]

Fertility: The actual reproductive performance of an individual, a couple, a group, or a population. [fecundity adjusted by social/cultural practices and events, including birth control, nutrition, delayed childbirth, etc.]

Demography: the study of life EVENTS.Individuals are “at risk” of experiencing these events.Events occur in a temporal-spatial context.

Right now, each of you can be located at a specific:Tn time (1 dimension)Sn space (3-dimensions [x,y,z], or 2 on a flat surface

[x,y], such as a map or the earth’s surface)

When you were born, you were located at a specific:T0 timeS0 spaceFrom these two we can get:Your age: Tn - T0

Your net migration: Sn - S0 (and with more, intermediate time points, we could also determine the specific legs of your migratory path to Ann Arbor)

- if we follow age groups (cohorts) over time, then this is known as cohort data

Here: randomly pick three every 10 years

- special case: if we follow the same individuals in a cohort over time, then this is known as panel data

- Here: pick the same three people every ten years

1970 1980 1990 2000 1970 1980 1990 2000

Birth Age 10

Age 20

Age 30

Birth Age 10

Age 20

Age 30

Having data over time is known as longitudinal data

"Give me the child at seven, and I will give you the man.”

Example of a cohort study

Mortality

10 Leading Causes of Death in 2001

1.heart disease2.cancer3.stroke4.chronic lower respiratory disease5.accidents6.diabetes7.pneumonia/flu8.Alzheimer's disease9.kidney disease10.SuicideSource: Centers for Disease Control and Prevention (CDC), National Center for Health Statistics

http://www.cdc.gov/nchs/data/ahcd/agingtrends/06olderpersons.pdf

Life Expectancy

Life Expectancy

€

e0 = T0

l0

€

ex = Tx

lx

= Total number of person - years lived above age xnumber of persons surviving to year x

From birth…

€

e0 = T0

l0€

ex = Tx

lx

= Total number of person - years lived above age xnumber of persons surviving to year x

From birth…

Life expectancy of a villageOf ten people…

674289

2798361175993

592

592 person-years___________10 persons

= 59.2 years

National Vital Statistics Reports, Vol. 54, No. 14, April 19, 2006

mortality in the United States, 1998

Death Rates, by Age Group, in the United States, 1998

Age 10 is the safest time of life

The first year of life is high risk

Once you get through the teenage years, the risk of death stops increasing -- at least for a few years.€

qx = dx

lx

e(0) IF YOU HAVE SURVIVED TO THE BEGINNING OF THIS AGE GROUP

75.0

80.0

85.0

90.0

95.0

100.0

105.0

0 10 20 30 40 50 60 70 80 90 100

If you have reached this birthday ...

… you will likely live to this age

e(0) IF YOU HAVE SURVIVED TOTHE BEGINNING OF THIS AGEGROUP

€

ex = Tx

lx

= Total number of person - years lived above age xnumber of persons surviving to year x

Life Expectancies (at birth) in the United States, in years (e0) for the year 2001, by Race and Sex

Male Female TOTALAll Races 74.4 79.8 77.2White 75.0 80.2 77.7Black 68.6 75.5 72.2

Source: National Vital Statistics Reports, Vol. 52, No. 14, February 18, 2004 33 url: http://www.cdc.gov/nchs/data/dvs/nvsr52_14t12.pdf

http://www.cdc.gov/nchs/fastats/pdf/nvsr50_06tb12.pdf; FROM http://www.cdc.gov/nchs/fastats/lifexpec.htmNational Vital Statistics Report,Vol.50,No.6,March 21,2002 33Table 12.Estimated life expectancy at birth in years,by race and sex:Death-registration States,1900 –28,and United States,1929 –99[For selected years,life table values shown are estimates;see Technical notes.Beginning 1970 excludes deaths of nonresidents of the United States;see Technical notes ]

Demography for planners?

absolute population levels in the future

but more importantly:

The components of population.

Demand for services vary by age, sex, family status, immigration status, etc.

Examples: demand for housing ,roads, schools, child care, special-needs housing.

Demographers thus speak of

Age-specific rates

(e.g., age-specific migration rates)

What is the implication for social policy (e.g., school funding, job training, pension plans) of dramatic differences in age-specific racial/ethnicity composition in places like California and Florida?

US Census 2000 (SF3)TM-P017. Median Age: 2000Source: American FactFinder Thematic Mapping

The basic components of demographic change:Components of Change: births, deaths, migrationExample:Population 2000 = Population1990 + Births - Deaths + Inmigration - Outmigration Or Pt+10 = Pt+ B - D + IN - OUT

Often you only know “net migration”, so this becomesPt+10 = Pt+ B - D + Net Migration

You can turn this formula around to solve for Net Migration:Net Migration = (Pt+10 - Pt) - (B - D)

Population change

Natural increase

Example:If starting Population Pt = 100,000Ending population Pt+10 = 150,000Births in time period B = 30,000Deaths D = 20,000

Then...Net Migration = (Pt+10 - Pt) - (B - D)

Net Migration = (150,000 - 100,000) - (30,000 - 20,000) = 50,000 - 10,000 = 40,000

Panel data is the ideal data for migration studies:(following the same people over time)

Panel data is the ideal data for migration studies:(following the same people over time)

Unfortunately, however, we don’t often have panel data.

From the U.S. Census form, there are two key questions of interest for migration studies:1) Where were you born?2) Where did you live 5 years ago? (not a great measure of migration over one’s lifetime, but a start)

1) Where were you born?2) Where did you live 5 years ago?

P21. PLACE OF BIRTH BY CITIZENSHIP STATUS [15] - Universe: Total populationData Set: Census 2000 Summary File 3 (SF 3) - Sample Data

United States

Los Angeles

city, California

Atlanta city,

Georgia

Ann Arbor city,

Michigan

Detroit city,

MichiganTotal: 100.0% 100.0% 100.0% 100.0% 100.0%

Native: 88.9% 59.1% 93.4% 83.4% 95.2%Born in state of residence 60.0% 40.2% 58.5% 49.5% 70.6%Born in other state in the United States: 27.7% 18.0% 34.1% 32.6% 24.0%

Northeast 6.6% 5.1% 7.0% 9.0% 1.6%Midwest 7.8% 5.3% 6.4% 12.1% 3.2%South 9.0% 5.5% 18.3% 6.9% 18.3%West 4.3% 2.1% 2.3% 4.5% 0.9%

Born outside the United States: 1.3% 0.9% 0.8% 1.4% 0.7%

Puerto Rico 0.5% 0.1% 0.1% 0.1% 0.3%U.S. Island Areas 0.1% 0.1% 0.1% 0.1% 0.0%Born abroad of American parent(s) 0.7% 0.7% 0.6% 1.2% 0.3%

Foreign born: 11.1% 40.9% 6.6% 16.6% 4.8%Naturalized citizen 4.5% 13.8% 1.6% 4.7% 1.6%

Source: US CENSUS American FactFinder

1) Where were you born?

Source: US CENSUS American FactFinder

P24. RESIDENCE IN 1995 FOR THE POPULATION 5 YEARS AND OVER --STATE AND COUNTY LEVEL [18] - Universe: Population 5 years and overData Set: Census 2000 Summary File 3 (SF 3) - Sample Data

United States

Los Angeles

city, California

Atlanta city,

Georgia

Ann Arbor city,

Michigan

Detroit city,

MichiganTotal: 100.0% 100.0% 100.0% 100.0% 100.0%

Same house in 1995 54.1% 49.5% 44.2% 36.7% 60.0%Different house in 1995: 45.9% 50.5% 55.8% 63.3% 40.0%

In United States in 1995: 43.0% 44.0% 51.7% 55.1% 37.8%Same county 24.9% 36.9% 26.9% 19.7% 32.0%Different county: 18.1% 7.0% 24.8% 35.5% 5.8%

Same state 9.7% 3.2% 11.6% 18.1% 2.8%Different state: 8.4% 3.8% 13.2% 17.3% 3.0%

Northeast 1.7% 1.0% 2.3% 4.8% 0.4%Midwest 1.8% 0.7% 2.0% 5.5% 0.7%South 2.9% 1.1% 7.5% 3.8% 1.3%West 2.1% 1.0% 1.4% 3.2% 0.6%

In Puerto Rico in 1995: 0.1% 0.0% 0.0% 0.0% 0.1%Same municipio 0.0% 0.0% 0.0% 0.0% 0.0%Different municipio 0.0% 0.0% 0.0% 0.0% 0.0%

Elsewhere in 1995: 2.8% 6.5% 4.1% 8.1% 2.1%

2) Where did you live 5 years ago?

A challenge: how to represent data on migration?

ANSWER:Simplify: Show only selective movesShow only the major flows (moves by many people)Show only long-distance moves (e.g., interregional)

Visually Representing Migration Flows: the Migration of Scientists and Engineers, 1982 - 1989 (single, before-after flows; only crossing regions)

Visually Representing Migration Flows: looking just at those scientists and engineers who worked in the Pacific Region in 1989.

Visually Representing Migration FlowsRegional migration, the break-up of states and geopolitical realignment are changing the shape of Africa. The continent is pulling apartand reforming under the combined effects of demography, massive urbanisation and economic, military and religious ambitions. The conflicts and population movements rarely fit into a pattern based on the state, but a confused picture is emerging on which the continent's new frontiers are being drawn. Three main territorial groups are taking shape. The first comprises the two ends of the continent, the second the war zones in the Horn of Africa, the Great Lakes region and the Congo. The third is emerging under the impact of the internationalisation of trade and new ways of exploiting resources.

http://MondeDiplo.com/maps/africambembemdv51

Map-Making Workshop at Sciences Po (L'Institut d'Etudes Politiques de Paris) http://www.ttc.org/maps/migra.htm

Map-Making Workshop at Sciences Po (L'Institut d'Etudes Politiques de Paris) http://www.ttc.org/maps/winte.htm#Top

Population Forecasting: from simple to complex(in each step, you are using more information about the past and existing population)

1. simplest: assumes it stays the same2. linear increase (could use trendline)3. exponential increase(go to Excel file on growth rates)-----------4. age-specific changes (cohort-survival analysis using matrix algebra)

The Cohort-Survival Method(taking the age distribution of the population into account)Or more specifically:Age-specific fertility and Age-specific mortality.“Cohort survival”: how each cohort survives. use matrix algebra to figure it out.Remember: COHORT = AGE GROUPS

Age Group (years)

Population in Year 2000 Survival Rate Birth rate

0 - 9 100 0.95 0 10 - 19 95 0.98 0.220 - 29 90 0.97 130 - 39 85 0.95 0.840 - 49 80 0.9 0.450 - 59 70 0.8 060 - 69 60 0.7 070 - 79 50 0.5 080 - 89 40 0.3 090 and above 30 0.1 0

Why use cohorts?

Because one can have more accurate estimates of fertility, mortality, migration (and thus population levels) if one breaks the population down into cohorts, since behavior is often age-specific. (e.g., cohort survival rates)

as distinctive from crude rates (e.g., mortality).

So, each member of a specific age-cohort (e.g., 20 - 30 year olds) is “at risk” of dying based on age-specific mortality rates. * some people survive into the next cohort* others die

so: persons in the next age group = persons in this age group * survival rate

(the models are descriptive, not explanatory, since no causal inferences are made about WHY population is increasing)

In addition, each female member of a cohort is “at risk” of giving birth. The chances of this are based on age-specific fertility rates. Generally these are highest in the age 20-30 cohort.

so, for each age group, each person is "at risk" for either giving birth or dying. (or also moving -- but we will only include migration in the analysis later)

AND WHY DO WE USE MATRICES, IF THEY SEEMS SO COMPLICATED?

IT GIVES US THE OPPORTUNITY TO INCLUDE IN A SINGLE EQUATION BOTH AGE-SPECIFIC BIRTHS AND DEATHS.

Use Matrix algebra to estimate population

where P1 is the new population level (by age)and C is the matrix (the sum of the Survival and Birth matrices, S and B). C is the components of change (here just B and S)

In general,

where n is the number of time periods into the future. (the trick is to raise the components of change matrix to the n power).

P1

= CP0

Pn

= C

n

P0

Migration: Adding the Migration Component

Where M is net migration

See the Oppenheim reading for details

P1 =CP0 +M

Question: How do you decide the optimal allocation of public services?

HINT: look at your feet…

Vote with your feet

Class Exercise (handouts)United Van Lines DATA

Growth from inside and outside

Growth from outside

Growth from insideLosing everyone

+

-

CONCEPTS

MEASURES

DIVERSITY TALENT

Gay Index Percent BA/BS Degrees

Intermediate variable

dependent variable

Assumes cause-effect flows this way -->>

Direct effect of diversity on high-tech

Indirect effect -- mediated via talent

No place in the United States is likely to escape a long and deep recession. Nonetheless, as the crisis continues to spread outward from New York, through industrial centers like Detroit, and into the Sun Belt, it will undoubtedly settle much more heavily on some places than on others. Some cities and regions will eventually spring back stronger than before. Others may never come back at all. As the crisis deepens, it will permanently and profoundly alter the country’s economic landscape. I believe it marks the end of a chapter in American economic history, and indeed, the end of a whole way of life.