Schelling's Model of Segragation

Schelling's Model of Segregation

Explanatory power

How the Schelling model improves our understanding:

• Every Asian nationality except Japanese is more segregated from whites than are Asians as a broad category, 1990-2010 census data
• Has been little change in the last two decades for Chinese and Indian segregation with whites (Logan and Zhang, 2013)
• Scholars generally agree that income differences between blacks and whites only explain a small fraction of current segregation levels (Muth, 1986; Clark, 1986)

Explanatory power

It gives us a non-intuitive result -

Seemingly small individual preferences can lead to strikingly large overall effects.

Simplifications of reality

• Claim - 2 ways to support it (cross-sectional and temporal)
• Temporal - take case study of U.S cities
• Housing segregation - particularly of African Americans - continues to be a dominant feature of most American cities, and it has been linked to a wide range of urban ills (Massey and Denton, 1994)
• Newsweek poll (Apr 2012): over 4 years, majorities of whites and blacks say segregation has either stayed the same or gotten worse.
• Index of dissimilarity: black/white segregation rose in American cities from an average of .6 in 1910 to .90 in 1940

E.g. Shape of grid not important, so choose the simplest, a square.

We already understand many complicating factors – so why study them in the model?

Counterarguments

Conclusion

Necessary to walk through counterarguments before addressing them.

It's a Counterexample:

No.

Does Schelling's model apply to any city?

Does Schelling's model apply to any one person?

Is segregation always a sign of blatant discrimination?

The model tells us – Not necessarily

The model shows there is at least one case where segregation is not the result of explicit discrimination

Does Schelling's model apply to any group of people?

Video Demonstration

Assumptions

Schelling’s checkerboard model continued

Perfect mobility

Two groups of people

No Transaction cost

People live at one of the location in a grid

People are happy if their neighborhood made up by the different colour type of people is under a certain ratio

Squared City

Preference Assumption

if the ratio becomes too high, they will feel unhappy and eventually move out to the next closest location that makes him happy,

Prof. J. McKenzie Alexander, LSE

The Racial Dot Map

Schelling’s checkerboard model

Background information of segregation in the US

Structure of our presentation

Schelling is interested in the racial segregation in the US

• Intro & Explanation

Descriptive Statistics for Residential Segregation Indexes for Blacks or African

Americans: 1980, 1990, and 2000

• Assumptions(including video run-through)

There are many types of segregation, including racial segregation, religious segregation, residential segregation and more

He attempts to construct an explanation of this in terms of a hypothesis of individual preferences.

‘’This chapter is about the kind of segregation … that results from discriminatory individual behaviour (Schelling, 1977,p.144)

• Counterarguments

• Why is it useful
• Explanation
• Understanding

His model shows that a rather weak assumption of individual preference can produce a sharp residential pattern in the aggregate [Little, 1991]

Proposed causes of segregation:

Housing discrimination

Concentrated inner city poverty

School locations

Racial prejudice

• Counter example
• Conclusion

Racial segregation, commonly measured by dissimilarity index.

Source: Descriptive Statistics for Residential Segregation Indexes for Blacks or African

Americans: 1980, 1990, and 2000

http://www.coopercenter.org/demographics/Racial-Dot-Map, Cable, University of Virginia, 2013