What's Happening with the Pension Fund? -- Part 27

by Charles Schwartz, Professor Emeritus, University of California, Berkeley
schwartz@physics.berkeley.edu                                            August 10, 2010
>> This series is available on the Internet at   http://socrates.berkeley.edu/~schwrtz

Debating the Record of Investment Performance

It is now 10 years since the UC Regents ousted their Treasurer, Patricia A, Small, and brought in a new era of investment strategy that featured outsourcing much of the management of the pension and endowment funds.  This is a suitable time to try to evaluate their success or failure. Here is a bit of that.

     One ongoing controversy is the request from the state Legislature that The Regents allow "shared governance" with employees over management of their pension fund - the University of California Retirement Plan (UCRP).  This is presently the subject of intense negotiation with employee unions; and there is a letter from UC President Mark Yudof in which he claims that the fund has been well managed and therefore the Regents' critics are wrong. That will be the focal point for this review.

President Yudof's Views on UCRP

     Here is the relevant paragraph from Yudof's July 8, 2010, letter addressed to LaKesha Harrison, President of AFSCME Local 3299, representing the UC Union Coalition.

"There can be no argument, however, that the Plan enjoyed strong performance during this time.  Over the past 20 years, the assumed return was 7.5 percent.  Yet UCRP achieved actual returns of 8.9 percent.  Relative to its policy benchmarks, UCRP performance in the past decade actually outpaced those of the prior two decades.  For the decade ended June 30, 2009, UCRP's total return exceeded that of the benchmark by 30 percent, whereas for the previous decade the return exceeded benchmarks by only 4 percent. And while I agree that the treasurer achieved impressive overall performance through 2000, it must be considered in the context of unusually favorable capital markets."

     In Table 1, below, I have assembled official data about UCRP investment performance over the past two decades.  These are 10 Year Annualized Total Returns: that means average return on investment looking back over the past ten years at each fiscal year end - sometimes this is called a "rolling" or "trailing" average. These numbers come from the annual reports of the UC Office of the Treasurer (UCOT).

Table 1.  UCRP investment performance - 10-Year Trailing Returns
Fiscal Year
10 Year Annualized
Total Return
            
Fiscal Year
10 Year Annualized
Total Return
1990
16.0%/11.6% *
2000
15.6%
1991
10.7%/11.5% *
2001
13.9%
1992
17.5%

2002
11.2%
1993
14.5%

2003
10.1%
1994
14.8%

2004
11.89%
1995
14.3%

2005
10.40%
1996
12.8%

2006
  9.04%
1997
13.0%

2007
  8.40%
1998
16.1%

2008
  5.66%
1999
15.7%

2009
  2.30%
2000
15.6%

2010
  2.83% **
* These are 5-Year and 20- or 21-Year data
** Ten Year data as of March 31, 2010 (The 6/30/10 results have not yet been published)

     What we see in Table 1 above is that over the decade 1990-2000, UCRP had an average annual return on investments of over 15% and over the decade 2000-2010 the average annual return on investments was at most 3%. Those two numbers are very much in contrast. Yet, President Yudof merely averages them and claims that the overall 20-year return of 8.9% shows that the fund has been well managed.  That is quite disingenuous in light of the widely recognized controversy over the Regents' ouster of their Treasurer, Patricia Small, in 2000 and their making major changes in investment strategy for the subsequent decade.

     Next, Yudof turns to talk about benchmarks.  The Treasurer's annual reports gives this data (shown in Table 2) for UCRP returns and benchmarks, averaged over the ten year periods ending at the indicated dates:

Table 2. UCRP 10-year annualized returns compared to Benchmarks
Date
UCRP Return
Benchmark
Difference
June 30, 2009
  2.30%
  1.77%
 + 0.53%
June 30, 2001
13.9%
13.3%
 + 0.6%





     The 2009 data shows that UCRP performance beat the benchmark by 0.53% and Yudof sees this as 30% of the benchmark (1.77%), while in 2001 the difference is just about the same, 0.6%, but he notes this is only about 4% of the benchmark for that year (13.3%).  This is complete nonsense. No respectable financial analyst would do what Yudof has done here - divide the above-benchmark performance by the benchmark itself. Let me explain that. Suppose it was a bad year for the markets and the benchmark performance came out zero; if you achieved performance any tiny bit better than the benchmark, this ratio would be infinity percent! (In really bad years the benchmark performance will be a negative number.) The only sensible way to look at this data is to focus on the differences themselves; and here we see that the two decades show nearly identical results.  (If we look at the most recently published data from the Treasurer's Office, dated March 31, 2010, then we find that the 10-year comparison between UCRP performance and the policy benchmark has dropped to a difference of only + 0.08%, which is much smaller than the number he used, +0.53%.)

     You might ask why I chose the year 2001 in the above comparison rather than 2000 or 1999. The answer is that 2001 is the earliest year in which one finds any data about the "policy benchmark" in the Treasurer's annual reports. Previously, they did not use any "policy benchmark" but rather compared UCRP investment performance to the average performance of a large peer group. The positive differences seen in those earlier reports were considerably larger than the numbers shown here.  One can argue about which is a better comparison - benchmarks or peer groups - but what Yudof has done here is really bad data manipulation. In fact, the so called "policy benchmark" reported in 2001 as a "Historical" benchmark is a fiction, manufactured after-the-fact by The Regents and their advisors at that time.

     Finally, Yudof tries to dismiss the outstanding investment performance of UCRP (under Treasurer Small) before 2000 by referring to the favorable market performance of that decade. The real facts become clear when one looks at UCRP performance compared to other comparable investment funds. While almost everyone did very well in the decade 1990-2000, UCRP's record was far ahead of their peers. This was acknowledged very candidly in the report from Wilshire Associates, which the Regents adopted as their new investment policy in March 2000. I have gathered and published further data showing this record of extraordinary accomplishment in this series of papers, "What's Happening with the Pension Fund?"

     I have previously published some data on such peer comparisons for the more recent decade; and there one saw that UCRP performance had slipped quite badly. More of this data is assembled below. Recently, the current Treasurer of The Regents has published a paper denigrating the use of peer comparisons.  My own view is that it is best to provide a variety of performance measures; each one may show a different aspect to be considered by fiduciaries and beneficiaries of the pension fund. I favor more light, not less, and if people want to debate the relative value of different measures, that is healthy. I do, however, condemn the cherry picking and distortions employed by President Yudof.

Some Comparative Data on Investment Performance

Data from TUCS (Trust Universe Comparison Service) acquired from UCOT
The numbers in these tables show the percentile ranking of UCRP performance for each peer group for the indicated time period (looking backward from the stated date).
Top Quartile is 25th percentile or lower; Median is 50th percentile; Lowest Quartile is 75 or more.

UCRP Percentiles as of 6/30/04  (43% above the median; 7% in lowest quartile)
Peer Group
1 year
2 years
3 years
4 years
5 years
7 years
10 years
All MasterTrusts
65
41
74
69
62
43
  9
All Trusts > $1 B







Public Funds
60
33
79
67
57
35
  3
Public > $1 B








UCRP Percentiles as of 6/30/05  (38% above the median; 10% in lowest quartile)
Peer Group
1 year
2 years
3 years
4 years
5 years
7 years
10 years
All MasterTrusts
39
58
40
68
67
59
17
All Trusts > $1 B
53
69
48
78
72
64
16
Public Funds
30
54
29
76
64
50
  2
Public > $1 B








UCRP Percentiles as of 6/30/06  (20% above the median; 24% in lowest quartile)
Peer Group
1 year
2 years
3 years
4 years
5 years
7 years
10 years
All MasterTrusts
72
54
56
48
69
59
28
All Trusts > $1 B
88
79
77
71
86
72
37
Public Funds
73
58
59
54
76
61
18
Public > $1 B








UCRP Percentiles as of 12/31/07  (62% above the median; 5% in lowest quartile)
Peer Group
1 year
2 years
3 years
4 years
5 years
7 years
10 years
All MasterTrusts
27
31
42
39
41
67
45
All Trusts > $1 B
40
53
64
65
67
82
65
Public Funds
20
27
36
38
42
70
39
Public > $1 B








UCRP Percentiles as of 12/31/08  (0% above the median; 70% in lowest quartile)
Peer Group
1 year
2 years
3 years
4 years
5 years
7 years
10 years
All MasterTrusts
77
71
69
72
68
76
65
All Trusts > $1 B
80
75
81
83
84
92
75
Public Funds
84
83
75
77
77
84
66
Public > $1 B
84
78
86
87
89
93
70

Note:  In this data collection one sees top quartile performance for UCRP seven times - six of those in the 10-year boxes in 2004 - 2006 and one in a 1-year box in 2007. 

The overall picture from this data is that there was much better performance, relative to peers, in the earlier years than there has been in the last decade. These tables cover the earliest and the latest data I have.  One can ask UCOT for the more recent TUCS data.

     In the past I have made repeated requests that the Treasurer publish these data routinely, along with other data showing the performance of individual external investment managers, and also data showing fees paid to external investment managers and consultants - all to no avail. That experience contradicts President Yudof's claim, found in the abovementioned letter: "As a public institution, the University is committed to transparent and open governance of the assets for which it is responsible."