Tuesday, November 25, 2014

HARVARD | Admissions, The Asian Quota (Comment)

Harvard Yard, 18th century.
A NY Times op-ed today, "Is Harvard Unfair to Asian-Americans?" by Yascha Mounk suggests that the answer to the question is yes, Harvard is unfair, or at very least opaque.

Based on the circumstantial evidence that Asians are a flat share of undergraduates for 20 years, despite the fact that Asians are "the fastest-growing racial group in America", there seems to be a quota for Asian applicants at Harvard.

If so, it would be reminiscent of the quota in place for Jews in the 1920s-1950s.

The topic of the admissions process to Ivy League colleges is of intense interest to Americans. The most-read article in the history of the New Republic was one in July this year on this subject, questioning the process.

The op-ed takes us back to Harvard President A. Lawrence Lowell, who warned that the "Jewish invasion" of Harvard College in 1922 (when Jews were 21.5 percent of freshmen, three times the rate at the turn of the century), would "ruin the college." He wanted the Jewish quota to be 15 percent.  The faculty objected, so he imposed a de facto cap on Jewish admissions by taking sports and character (and geographical distribution) into account in the admissions process that was pursued into the 1950s.

I wrote a letter to the East Hampton Star about this two years ago, where I noted the changes that had taken place in the 1950s. Catholics may have been very briefly favored as the Kennedy family rose in importance. Portsmouth Priory (now Abbey) School, with a senior class then of 35 boys, had seven alumni in the Harvard Class of 1962–more than twice the number of African Americans in the class. (Portsmouth also had an exceptionally qualified headmaster, Fr. Leo van Winkle, an atomic scientist with a Yale doctorate.)

Our Harvard class elected the first African American Class Marshal–Haywood Burns. Affirmative action for Catholics, if that is what occurred, was overtaken in the 1960s by a realization that Harvard should participate in affirmative action for African Americans. This in turn was overtaken by the huge upheaval when women pressed for equal status. The Harvard Class of 1963 was the first to offer Radcliffe students a Harvard diploma. The same objections to women were raised that had been raised about Jews. The quota today, I suggested in my letter, was applied to Asians.

The op-ed by Mounk accepts that the admissions process does not lend itself to "one right answer" to the problem of allocating spaces in elite schools. The author simply asks for more transparency about the process...

Comment

I frankly doubt we will voluntarily get much more transparency about the process currently in place– the choices are too difficult and controversial–but we might get more transparency about the past and about the process, which will be helpful in addressing the issues being raised today.

Meanwhile, an advocacy group has filed a law suit against Harvard for discriminating against Asians. The Economist doubts it will win, but if it makes a good case it may lead to changes in policies of Harvard and other Ivy League colleges–if not changes in the selection criteria and process, then perhaps more disclosure of the criteria.

Friday, November 21, 2014

INCOMES | U.S. Counties, 2013

The dark blue areas have seen significantly higher personal incomes in 2013.
The brown areas have been flat or have seen declines. The others are
closer to the national average increase of 1.3 percent.
Personal income data by county for 2013 were just released by the Bureau of Economic Analysis.

Of course, the numbers being for the year 2013 when it's nearly 2015, the reaction could be: Ho hum.

But the delay is made up for by the reliability and importance of the numbers. They provide a good comparative look at the counties.

Personal income data by county tell us how their neighbors are doing over time and relative to the rest of the country.

The Bureau of Labor Statistics numbers on jobs and unemployment are released faster because we agree to suspend disbelief in our eagerness to have the data in our hands. The BLS tries to adjust the data to mask the variability of the data resulting from seasonality and small sample coverage in the household survey. The BEA's per capita personal income numbers by county are closer to a final answer to the important question: "How are we doing, money-wise?"

Some regional data are in the table below, abbreviated to make it easier to talk about and handle (the full table is at www.bea.gov under "Regional"). The top line is the summary for the United States. Per capita income grew from $42,332 in 2011 to $44,765 in 2013. There was quite a jump between 2011 and 2012, 4.4 percent, but the increase between 2012 and 2013 was less than one-third the rate, 1.3 percent.

Knowing that the country as a whole had a spurt of personal income in 2012 and then fell back to a lower rate in 2013 helps interpret how each state and county was doing.

The states vary considerably among themselves and between years in the change in their share of the increased income. Arkansas jumped 6.8 percent in 2012 and then rose only 0.8 percent in 2013. The District of Columbia actually saw a loss of per capita income in 2013, but at $75,950 in 2012 that was a loss from a high base.

The variability is even greater among counties. Indian River County (Vero Beach) saw its per capita income grow 5 percent in 2012, but then it declined in 2013. Palm Beach grew 4.8 percent from its high base of $57,252 and kept growing in 2013 at the national rate of 1.3 percent.

Table 1. Per Capita Personal Income by County, 2011-2013
Per capita personal income and Rank in State
Change and Rank in State
Dollars
Rank
Percent change
Rank
2011
2012
2013
2013
2012
2013
2013
United States
42,332
44,200
44,765
--
4.4
1.3
--
Alabama
35,010
35,942
36,481
--
2.7
1.5
--
Alaska
48,181
49,906
50,150
--
3.6
0.5
--
Arizona
35,512
36,624
36,983
--
3.1
1.0
--
Arkansas
34,089
36,423
36,698
--
6.8
0.8
--
California
44,749
47,505
48,434
--
6.2
2.0
--
Colorado
44,183
46,315
46,897
--
4.8
1.3
--
Connecticut
57,547
60,223
60,658
--
4.7
0.7
--
Delaware
42,696
44,031
44,815
--
3.1
1.8
--
District of Columbia
74,103
75,950
75,329
--
2.5
-0.8
--
Florida
40,215
41,041
41,497
--
2.1
1.1
--
Indian River (Vero)
51,890
54,501
54,448
5
5.0
-0.1
62
Miami-Dade
38,242
39,467
39,880
17
3.2
1.0
41
Orange (Orlando)
36,333
37,479
37,844
24
3.2
1.0
44
Palm Beach
54,616
57,252
57,985
2
4.8
1.3
34
St. Lucie
31,289
30,932
31,182
40
-1.1
0.8
49
Georgia
36,422
37,229
37,845
--
2.2
1.7
--
Hawaii
42,989
44,578
45,204
--
3.7
1.4
--
Idaho
33,677
35,142
36,146
--
4.4
2.9
--
Illinois
44,169
46,009
46,980
--
4.2
2.1
--
Cook
46,966
48,948
49,661
7
4.2
1.5
90
Indiana
36,367
38,136
38,622
--
4.9
1.3
--
Iowa
42,656
44,014
44,763
--
3.2
1.7
--
Kansas
42,098
43,380
44,417
--
3.0
2.4
--
Kentucky
34,568
35,857
36,214
--
3.7
1.0
--
Louisiana
38,501
40,617
41,204
--
5.5
1.4
--
Maine
38,802
39,863
40,924
--
2.7
2.7
--
Maryland
52,191
53,659
53,826
--
2.8
0.3
--
Baltimore
51,886
53,835
54,009
6
3.8
0.3
14
Baltimore City
42,071
43,386
44,053
15
3.1
1.5
10
Massachusetts
54,235
56,713
57,248
--
4.6
0.9
--
Suffolk
55,608
57,491
57,660
6
3.4
0.3
14
Michigan
37,163
38,585
39,055
--
3.8
1.2
--
Minnesota
45,220
47,377
47,500
--
4.8
0.3
--
Mississippi
32,108
33,446
33,913
--
4.2
1.4
--
Missouri
38,016
39,933
40,663
--
5.0
1.8
--
Montana
36,890
39,142
39,366
--
6.1
0.6
--
Nebraska
43,721
45,914
47,157
--
5.0
2.7
--
Nevada
37,445
39,229
39,235
--
4.8
0.0
--
New Hampshire
47,664
50,056
51,013
--
5.0
1.9
--
New Jersey
53,323
54,932
55,386
--
3.0
0.8
--
Atlantic
41,397
42,288
42,425
19
2.2
0.3
18
Bergen
67,248
69,281
69,495
4
3.0
0.3
19
Burlington
49,471
51,149
51,638
12
3.4
1.0
12
Camden
44,229
45,063
45,544
15
1.9
1.1
9
Cape May
50,908
53,070
53,932
9
4.2
1.6
3
Cumberland
35,413
35,468
35,825
21
0.2
1.0
10
Essex
53,597
54,318
54,606
7
1.3
0.5
15
Gloucester
43,488
44,833
45,169
16
3.1
0.7
13
Hudson
49,111
49,978
50,172
13
1.8
0.4
17
Hunterdon
69,717
74,534
75,523
2
6.9
1.3
6
Mercer
53,037
55,933
56,906
6
5.5
1.7
2
Middlesex
50,267
51,730
52,291
11
2.9
1.1
7
Monmouth
59,875
61,997
62,901
5
3.5
1.5
4
Morris
71,914
74,826
75,054
3
4.0
0.3
20
Ocean
42,121
43,016
43,214
18
2.1
0.5
16
Passaic
43,853
44,600
44,688
17
1.7
0.2
21
Salem
41,138
41,550
41,997
20
1.0
1.1
8
Somerset
72,704
76,918
77,685
1
5.8
1.0
11
Sussex
50,800
52,592
52,958
10
3.5
0.7
14
Union
52,297
53,638
54,382
8
2.6
1.4
5
Warren
46,070
48,115
49,040
14
4.4
1.9
1
New York
51,941
54,099
54,462
--
4.2
0.7
--
Albany
50,275
52,587
53,515
7
4.6
1.8
16
Bronx
32,565
32,680
32,852
60
0.4
0.5
44
Dutchess
47,083
49,378
49,627
10
4.9
0.5
48
Erie
42,925
45,063
45,496
16
5.0
1.0
35
Essex
36,268
38,392
39,309
34
5.9
2.4
4
Kings
41,038
42,211
42,306
24
2.9
0.2
53
Nassau
68,979
72,460
72,549
3
5.0
0.1
58
New York
116,329
120,382
121,632
1
3.5
1.0
34
Putnam
55,844
58,865
58,955
4
5.4
0.2
56
Queens
43,140
44,431
44,966
19
3.0
1.2
27
Rensselaer
41,961
43,371
44,152
22
3.4
1.8
15
Richmond
49,839
51,223
51,328
9
2.8
0.2
55
Rockland
54,294
56,746
56,657
6
4.5
-0.2
60
Suffolk
53,774
56,819
56,940
5
5.7
0.2
54
Ulster
41,818
44,045
44,527
20
5.3
1.1
30
Westchester
75,291
80,505
80,363
2
6.9
-0.2
61
North Carolina
36,508
38,538
38,683
--
5.6
0.4
--
North Dakota
47,868
56,310
53,182
--
17.6
-5.6
--
Steele
55,909
81,029
53,918
16
44.9
-33.5
50
Ohio
38,631
40,230
41,049
--
4.1
2.0
--
Oklahoma
38,980
41,399
41,861
--
6.2
1.1
--
Oregon
37,707
39,258
39,848
--
4.1
1.5
--
Pennsylvania
43,806
45,577
46,202
--
4.0
1.4
--
Rhode Island
44,571
46,257
46,989
--
3.8
1.6
--
South Carolina
34,079
35,347
35,831
--
3.7
1.4
--
South Dakota
44,439
45,676
46,039
--
2.8
0.8
--
Tennessee
37,151
39,002
39,558
--
5.0
1.4
--
Texas
41,016
43,271
43,862
--
5.5
1.4
--
Utah
34,235
35,891
36,640
--
4.8
2.1
--
Vermont
42,968
44,443
45,483
--
3.4
2.3
--
Virginia
47,076
48,715
48,838
--
3.5
0.3
--
Washington
44,565
47,055
47,717
--
5.6
1.4
--
West Virginia
33,954
35,140
35,533
--
3.5
1.1
--
Wisconsin
40,780
42,475
43,244
--
4.2
1.8
--
Wyoming
49,260
52,469
52,826
--
6.5
0.7
--
1. Per capita personal income was computed using Census Bureau midyear population estimates. Estimates reflect county population estimates available as of March 2014.
2. Percent change was calculated from unrounded data.
Source: Bureau of Economic Analysis

The New York State counties are interesting. The average incomes confirm the reputation of Manhattan (New York County) as dwarfing neighboring county incomes. The next-highest skip over the "Outer Boroughs" and go to Westchester, Nassau, Putnam, Suffolk and Rockland counties in New York State and a similar group of high rollers in New Jersey. Westchester saw a big 6.9 percent leap in personal income in 2012, but it and Rockland both saw a small decline in per capita incomes in 2013.

Notice the oddly high variability of per capita income in Steele County, ND. A 45 percent increase in 2012 and a one-third decrease in 2013. However, the explanation is simple. The total population of the county is under 2,000 people, so a few families with highly correlated incomes - for example, all dependent upon revenue from the same oil well - could skew the county numbers.