Asset Allocation Review

Arizona Permanent State Land Fund
Asset Allocation Review
January 2012

Stephen P. McCourt, CFA Laura B. Wirick, CFA, CAIA
Managing Principal Vice President

M E K E T A I N V E S T M E N T G R O U P
5796 ARMADA DRIVE SUITE 110 CARLSBAD CA 92008
760 795 3450 fax 760 795 3445 www.meketagroup.com
L:ASTODoc20111201-Full Doc.doc

Arizona Permanent State Land Fund Asset Allocation Review

Table of Contents

1. Introduction

2. Asset Allocation Overview

3. Proposed Policy Options

4. Risk/Return Analysis
− Mean-Variance Optimization
− Deterministic Economic Scenario Analysis
− Risk Analytics
− Liquidity Analysis

5. Analysis of Distribution Policy

6. Analysis of Fiscal Year Change

7. Conclusions

8. Appendices
− Detail of Deterministic Scenario Analysis
− Additional Risk Analytics
− Asset Class Definitions
− Risk/Return Analysis Assumptions
Asset Allocation
Review
Prepared by Meketa Investment Group for the Arizona Permanent State Land Fund
i

Introduction

Introduction

• This document presents a range of asset allocation options for the Fund, on both a restricted and unrestricted
basis.
• We provide various approaches to assessing the risk in each policy option in order to provide a “mosaic” of the
risks faced by the Fund.
• The goal of this review is not to declare one portfolio the “right” choice or the only prudent choice, but to
highlight the risk and return tradeoffs of different policy portfolios.
• Over long periods of time, riskier assets, such as equities, are likely to produce relatively high rates of return.
Consequently, higher allocations to risky assets increase the likelihood of the Fund maintaining and improving the
real value of its corpus over the long term. However, riskier assets increase volatility in the short term.
• The asset allocation review process highlights the natural tension between long term goals and short term risks,
and should allow the Trustees to make more informed decisions regarding portfolio positioning.
• This document also reviews the current distribution policy of the Fund, and evaluates two alternative distribution
policies.
• A brief analysis of the historical relevance of different fiscal year end dates is provided.

Asset Allocation
Review
1

Asset Allocation Overview

Asset Allocation

What is Asset Allocation?
Asset allocation refers to the distribution of assets across a number of asset classes that exhibit modest or low correlations
with each other. Each asset class exhibits a unique combination of risk and reward. The expected and realized long-term
returns vary by asset class, as do the interim volatility of those returns. Some asset classes, like equities, exhibit high
degrees of volatility, but also offer high returns over time. Other asset classes, like cash, experience very little volatility, but
offer limited return potential.

Why is Asset Allocation important?
The distribution of assets across various asset classes exerts a major influence on the return behavior of the aggregate pool
over short and long time periods.

How does Asset Allocation affect aggregate performance?
In addition to exhibiting unique characteristics, each asset class interacts differently with other asset classes. Because of
low correlations, the likelihood that any two asset classes will move together in the same direction is limited, with the
movement of one asset class often partially offsetting another’s. Combining asset classes allows investors to control more
fully the aggregate risk and return of their portfolio, and to benefit from the reduction in volatility that stems from
diversification.

Asset Allocation
Review
3


Asset Allocation Review Process

Fund’s Investment Objectives and Constraints

Traditional Asset Risk and Scenario Distribution Policy
Allocation Study (MVO) Analysis Analysis

Liquidity Analysis

Policy Portfolio

Asset Allocation
Review
4


Current Investment Constraints

• What is the overall time horizon for the State Land Fund?
− Perpetual.
• What are the liquidity needs of the State Land Fund?
− The current annual distribution is the average total rate of return for the previous five fiscal years less the
average of the annual percentage change in the GDP price deflator for the previous five fiscal years,
multiplied by the trailing 60-month average market value of the Fund.
• What are the legal and regulatory constraints under which the Fund operates?
− Constitutional Constraints
No more than 60% of the Fund (at cost) may be invested in equities.
No more than 5% of equity securities may be invested in any one firm.
All equities purchased must be listed on a national exchange (no private equity).
Fixed income securities must be rated investment grade (BBB or higher).
Exchange-traded funds ("ETFs") are allowed, but only if their underlying holdings are allowed by
state law.
− Limitations by Statute
Fixed income securities are limited to firms organized to do business in the United States.
ETFs are allowed, but are classified based on their underlying holdings.

Asset Allocation
Review
5


The Secular Decline in Investment Returns1
18% 100%

16%

80%
14%

12%

Percent Return
60%

Probability
10%

8%
40%
6%

4%
20%

2%

0% 0%
1980 1985 1990 1995 2000 2005 2010
65/35 Eq/Bond Exp. Ret. 15.1% 13.8% 9.1% 7.8% 4.7% 5.3% 5.8%
Probability of earning 8% 97% 93% 56% 41% 15% 18% 22%
Actual 10-year Return 15.5% 12.8% 14.3% 10.8% 2.4%

• A portfolio comprised of 65% domestic stocks (Russell 3000) and 35% investment grade bonds (Barclays
Aggregate) has produced diminishing returns over the past twenty years (along with diminished forward-looking
expected returns), as interest rates have declined, equity valuations have inflated, and equity market returns have
disappointed.

1
Expected return assumptions for 1) Bonds equals the yield of the ten-year Treasury plus 100 basis points, and 2) Equities equals the dividend yield plus the earnings yield of the S&P 500 index (using the inflation-
adjusted trailing 10-year earnings). Probability calculation is for the subsequent ten years.
Asset Allocation
Review
6


Expected Return and Volatility for Major Asset Classes1

12

Private Equity EM Equity
10
Natural Resources

U.S. Equity Infrastructure
8 EAFE Equity
Expected Return (%)

Core Real Estate
EM Bonds
Hedge Funds High Yield Commodities
6

4 TIPS
Bonds
Cash

2

0
0 5 10 15 20 25 30

Standard Deviation (%)

• A positive correlation exists between long-term return expectations and the level of risk accepted.

1
Expected return and standard deviation are based upon Meketa Investment Group’s 2011 Annual Asset Study.
Asset Allocation
Review
7

Proposed Policy Options

Review of Proposed Asset Allocation Policies

• We reviewed three alternative policy portfolios that conform with current restrictions, as well as three policy
portfolios that would require changes to current investment restrictions.
• In the case of both the Unrestricted and Restricted Policies, we recommend that the Trustees consider reducing
domestic equity exposure, while increasing developed foreign and emerging markets equity exposure. We also
recommend considering the addition of dedicated Treasury Inflation-Protected Securities ("TIPS") exposure.
• In the case of Unrestricted Policy U-1, we added a target allocation to investment grade non-U.S. fixed income,
in addition to the non-U.S. equity securities that we added to the Restricted Policies. The policy could be
adopted with a change to existing statutes.
• In the case of Policies U-2 and U-3, we also added target allocations to below investment grade fixed income and
to private market alternatives (i.e., private equity, real estate, and infrastructure) that would require a
constitutional change.

Asset Allocation
Review
9


Restricted Asset Allocation Policy Options1
Current
Policy Policy R-1 Policy R-2 Policy R-3
(%) (%) (%) (%)
Equity 60 49 48 50
U.S. Equity 60 37 27 24
Developed Foreign Equity 0 6 8 8
Emerging Market Equity 0 6 12 15
Frontier Market Equity 0 0 1 3

Investment Grade Fixed Income 40 42 40 40
Investment Grade Bonds 40 30 20 20
TIPS 0 12 20 20

Real Assets (Public Equity) 0 9 12 10
REITs 0 3 4 2
Natural Resources (public) 0 3 4 5
Infrastructure (public) 0 3 4 3
Expected Return (%) 6.4 6.4 6.7 6.9
Standard Deviation (%) 11.0 10.2 10.8 11.2
Sharpe Ratio 0.58 0.63 0.62 0.61
Target Illiquid Assets (%) 0 0 0 0
Target Non-U.S. Assets (%) 0 12 21 26

• Alternative policies R-1, R-2, and R-3 produce more efficient mean-variance outcomes than the current policy,
largely due to the diversification benefits of investing outside the U.S.

1
Expected return and standard deviation are based upon Meketa Investment Group’s 2011 Annual Asset Study. Throughout this document, returns for periods longer than one year are annualized.
Asset Allocation
Review
10


Unrestricted Asset Allocation Policy Options
Complies With Constitution Requires Change to Constitution
Current Policy Policy U-1 Policy U-2 Policy U-3
(%) (%) (%) (%)
Equity 60 48 36 47
U.S. Equity 60 21 12 16
Developed Foreign Equity 0 14 6 5
Emerging Market Equity 0 12 7 12
Frontier Market Equity 0 1 1 1
Private Equity 0 0 10 13
Investment Grade Fixed Income 40 30 20 15
Investment Grade Bonds 40 13 10 7
TIPS 0 17 10 8
Credit 0 10 19 15
High Yield 0 0 7 6
Bank Loans 0 0 3 2
Foreign Debt (Investment Grade) 0 10 1 0
Foreign/Emerging Market Debt (Below IG) 0 0 8 7
Real Assets 0 12 25 23
Core Real Estate 0 0 6 4
REITs 0 2 0.5 0.5
Value-Added Real Estate 0 0 2 2.5
Opportunistic Real Estate 0 0 2 2.5
Infrastructure (private) 0 0 4 3
Infrastructure (public) 0 5 0.5 0.5
Natural Resources (private) 0 0 4 5
Natural Resources (public) 0 5 2 1
Commodities 0 0 4 4
Expected Return (%) 6.4 6.8 7.5 8.0
Standard Deviation (%) 11.0 11.0 11.0 12.7
Sharpe Ratio 0.58 0.62 0.68 0.63
Target Illiquid Assets (%) 0 0 28 30
Target Non-U.S. Assets (%) 0 37 23 25

• The unrestricted policies meaningfully improve the mean-variance outcome for the Fund, as they take advantage
of a variety of different (and illiquid, in the case of Policies U-2 and U-3) asset classes.
Asset Allocation
Review
11


Peer Asset Allocation Target Comparison1
Arizona New Mexico Texas
(%) (%) (%)

U.S. Equity 60 31 28

Developed Foreign Equity 0 10 18

Emerging Market Equity 0 5 4

Hedge Funds 0 8 17

Private Equity 0 10 6

Fixed Income2 40 16 15

Real Estate 0 10 6

Commodities/Natural Resources 0 10 6

Expected Return (%) 6.4 7.6 7.3
Standard Deviation (%) 11.0 12.6 12.2
Sharpe Ratio 0.58 0.60 0.60
Target Illiquid Assets (%) 0 38 18
Target Non-U.S. Assets (%) 0 15 22

• Relative to similar vehicles in the states of New Mexico and Texas, the Arizona State Land Fund is invested more
cautiously and across fewer asset classes.

1
Peer targets are approximations based on available public data.
2
Both New Mexico and Texas are allowed to invest in non-U.S. fixed income. New Mexico's non-U.S. investments are limited to 15% of the portfolio, though there is a bill before the legislature to remove that
requirement.
Asset Allocation
Review
12

Mean Variance Optimization


Asset Allocation
Review
14



• Mathematically determines an “efficient frontier” of policy portfolios with the highest risk-adjusted returns.
• All asset classes exhibit only three characteristics, which serve as inputs to the model:
− Expected return
− Expected volatility
− Expected co-variance with all other assets
• The model assumes:
− Normal return distribution
− Stable volatility and co-variances over time
− Returns are not serially correlated
• The MVO Model tends to underestimate the risks of large negative events.

Asset Allocation
Review
15


The Efficient Frontier

100% Stocks

60% Stocks,
Return

40% Bonds

100% Bonds

Risk

• Combining uncorrelated assets produces an “efficient frontier.” Different combinations of assets (e.g., 60% stocks
& 40% bonds) will lie along this efficient frontier.
• By combining assets that are not highly correlated with each other, the Fund can produce a higher expected
return for a given level of risk than it could by investing in perfectly correlated assets. Alternatively, it can
experience lower risk for a given level of expected return.

Asset Allocation
Review
16


Asset Allocation Policy Options

8.5
U-3 Theoretical Efficient Frontier
without Constitutional
constraints
Expected Return (%) U-2
7.5
Current
Efficient
Frontier
U-1
R-3

6.5 R-2

R-1 Current

5.5
9 10 11 12 13 14
Expected Risk (%)

• Based on the MVO model inputs, each Restricted Policy (and Unrestricted Policy U-1) falls on an efficient frontier
that expresses greater investment efficiency than current policy. The efficient frontier shifts upward (more
efficient) for Policies U-2 and U-3, which include illiquid asset classes.1

1
See the appendix for Meketa Investment Group’s MVO assumptions.
Asset Allocation
Review
17


Asset Allocation Policy Options
Current
Policy Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3
(%) (%) (%) (%) (%) (%) (%)
Expected Return (%) 6.4 6.4 6.7 6.9 6.8 7.5 8.0
Expected Standard Deviation (%) 11.0 10.2 10.8 11.2 11.0 11.0 12.7
Sharpe Ratio 0.58 0.63 0.62 0.61 0.62 0.68 0.63

20-Year Projected Return (Percentiles)
95th 9.9 9.7 10.1 10.4 10.4 11.0 11.9
75th 7.5 7.5 7.8 8.0 7.9 8.6 9.1
50th 5.8 6.0 6.2 6.3 6.3 6.9 7.2
25th 4.2 4.4 4.6 4.6 4.6 5.3 5.4
5th 1.9 2.3 2.4 2.3 2.3 3.0 2.7

• Each asset allocation policy option expresses a distinct expected return and expected volatility, as shown in the
efficient frontier on the previous page.
• While the "expected return" represents the "median" of all possible outcomes, we also show the 20-year projected
return across the 95th, 75th, 50th, 25th, and 5th percentiles to emphasize that the range of potential returns
varies widely.

Asset Allocation
Review
18


20-Year Opportunity Cost of the Current Allocation as Compared to Alternate Policies
Based on $3 Billion Beginning Value at Selected Percentile Returns1
Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3
Ending Wealth Percentiles ($ millions) ($ millions) ($ millions) ($ millions) ($ millions) ($ millions)

95th Percentile (highest value) -520 1,040 2,140 1,870 4,480 9,190

75th Percentile 60 810 1,250 1,180 2,910 4,710

50th Percentile 250 660 850 850 2,150 2,890

25th Percentile 330 530 570 610 1,580 1,720

5th Percentile (lowest value) 350 390 330 390 1,030 760

• The power of compounding makes small differences in average annual returns into large differences in
end-of-period value.
• In the "median" expectation (50th percentile), each of the alternative policies would add between $250 million
and $2.9 billion to the value of the Fund over 20 years (assuming no cash inflows or outflows).

1
See the appendix for Meketa Investment Group’s MVO assumptions. Ending market values assume no inflows to, or outflows from, the Fund.
Asset Allocation
Review
19

Deterministic Economic Scenario Analysis


Asset Allocation
Review
20


• Where mean-variance optimization captures a probabilistic forecast of portfolio returns, deterministic economic
scenario analysis allows one to look at “path dependent” scenarios. These scenarios can incorporate many of the
real world risks that are often overlooked when using only traditional mean-variance optimization.
• With history as a guide, we have developed five realistic economic scenarios, ranging from very pessimistic to
very optimistic.
− Our methodology is strict but also allows some creativity in how we construct our twenty-year forecasts:
essentially, we use history to model five-year sub-periods with different economic characteristics, then
we string these periods together in a way that captures possible future scenarios.
− At the same time, we ensure that the entire twenty-year history passes reasonability checks. We believe
that this methodology allows us to be cognizant of historical relationships between assets and economic
factors, while at the same time ensuring that we are not presenting a set of scenarios dominated by
(possibly unrepeatable) periods in history.

Asset Allocation
Review
21


Deterministic Economic Scenario Analysis Summary

• Over the next twenty years, the investment outcome of the State Land Fund will be more determined by the
economic conditions that unfold than the asset allocation policy selected.

Average Annual Return
Very Very
Pessimistic Pessimistic Moderate Optimistic Optimistic Range
(%) (%) (%) (%) (%) (%)

Current Policy 3.2 4.7 6.8 8.8 10.9 7.7

R-1 3.6 5.3 7.2 8.7 10.5 6.9

R-2 3.9 5.7 7.6 8.9 10.4 6.6

R-3 3.9 5.5 7.4 8.5 10.0 6.1

U-1 4.1 5.9 7.7 8.7 10.1 6.0

U-2 2.4 6.0 7.6 9.2 11.6 9.2

U-3 2.2 5.9 7.9 9.5 12.1 9.9

• Policy U-1 has the narrowest range of outcomes across economic scenarios.

Asset Allocation
Review
22

Risk Analytics

Risk Analytics

Asset Allocation
Review
23

Risk Analytics

Historical Stress Test: Global Financial Crisis (4Q07 thru 1Q09)
(Cumulative Return)
0%

-5%

-10%

-15%

-20%

-20.9% -21.4%
-25% -24.2% -24.1%
-25.8% -25.6%

-30%
-29.8%

-35%
Current Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3

• In an extended down market environment (e.g., the GFC), only Policy R-1 would produce a higher relative return
than the current allocation.

Asset Allocation
Review
24

Risk Analytics

Historical Stress Test: Stagflation (January thru March 1980)
(Cumulative Return)
0%

-1%

-2%

-3%

-4%

-5%

-5.3% -5.3%
-6%

-6.6% -6.5% -6.4%
-7%
-6.9%
-7.3%
-8%

-9%

-10%

• In an environment of high inflation but low growth (e.g., early 1980), the alternative policies protect better than
the current allocation, due to higher allocations to real assets and foreign assets.

Asset Allocation
Review
25

Risk Analytics

Stress Testing: Impact of Market Movements
Expected Returns1

Current
Allocation Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3
What happens if (over a 12-month period): (%) (%) (%) (%) (%) (%) (%)

10-Year T-Bond rates rise 100 bp 5.9 7.3 8.6 8.1 9.0 8.6 9.8
10-Year T-Bond rates rise 200 bp 6.6 7.3 8.5 7.2 8.8 8.5 10.6
10-Year T-Bond rates rise 300 bp -0.8 -0.6 -0.1 -3.3 -0.7 1.9 3.0

BBB Spreads widen by 100 bp, HY by 200 bp 3.6 4.2 4.9 5.1 4.7 4.2 4.5
BBB Spreads widen by 300 bp, HY by 1000 bp -26.6 -23.0 -22.0 -21.6 -21.9 -23.4 -26.9

Trade-weighted US$ gains 10% 5.3 3.8 2.7 2.3 0.6 1.4 0.9
Trade-weighted US$ gains 20% 10.6 7.7 5.4 4.5 1.3 2.9 1.9

Equities decline 10% -5.2 -5.1 -5.5 -5.0 -5.9 -5.6 -6.5

• Each policy portfolio has a different sensitivity to four major risk factors: interest rates, credit spreads, currency
values, and equity values.
• The Fund’s primary risk factor would continue to be a decline in the equity markets, regardless of the policy.

1
Assumes that assets not directly exposed to the factor are effected nonetheless. See the Appendix for further details.
Asset Allocation
Review
26

Risk Analytics

Risk Analytics Summary

• Using mean-variance optimization, only the current allocation has a more than a 10% chance of producing a
negative return over a five-year period. Over the 20-year period, the current allocation has the lowest probability
of earning a 5% return.
• In most historical market scenarios, the alternative policies would outperform the current allocation. The
exceptions are a stronger U.S. Dollar or a repeat of the 2008 Global Financial Crisis, when all assets suffered
except for high quality bonds and cash.
• In each portfolio, equity risk dominates the risk profile of the portfolio, and the Fund's primary risk factor would
continue to be a decline in the equity markets, though the alternative policies reduce equity risk somewhat.

Asset Allocation
Review
27

Liquidity Analysis

Liquidity Analysis

Asset Allocation
Review
28

Liquidity Analysis

• Liquidity risk is a meaningful risk that is generally not captured in traditional asset allocation processes.
• The Fund should maintain adequate liquidity to avoid having to sell illiquid assets at distressed prices to satisfy
any distribution needs.

Asset Allocation
Review
29

Liquidity Analysis

Liquidity Profile
Daily Monthly Quarterly Not Liquid
100%

90%

80%

70%

60%

50%

40%

30%

20%

10%

0%
Current Policy Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3

• Using strict liquidity assumptions, each policy portfolio except for Unrestricted Policies U-2 and U-3 has at least
75% daily liquid assets.
Asset Allocation
Review
30

Analysis of Distribution Policy


Historical Distributions

90

80

70

60

50
Millions

40

30

20

10

0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

• Over the last 13 years, the State Land Fund’s distributions have averaged approximately $40 million per year,
with a high in 2012 near $80 million and a low in 2010 of zero.
Asset Allocation
Review
32


Distribution Policy

Current Policy
• The current annual distribution is the average total rate of return for the previous five fiscal years less the average
of the annual percentage change in the GDP price deflator for the previous five fiscal years, multiplied by the
average market value over the previous five years.

Possible Alternative Policies

Alternative Distribution Policy A
• The Fund could modify the current return-oriented distribution policy to one that is based on a static proportion
of the value of the corpus. A static proportion distribution would further smooth the annual distribution rate.
− For example, 2.5% of the average five-year market value. This policy would result in a distribution of
approximately $67 million in 2012.

Alternative Distribution Policy B
• The Fund could set an absolute value as its distribution policy; i.e., a flat dollar amount. This policy would allow
beneficiaries to plan on a more consistent distribution amount for budgeting purposes.
− For example, $80 million in 2012, increased by the GDP inflator each year.

Asset Allocation
Review
33


Peer Distribution Policies
• New Mexico State Investment Council Land Grant Permanent Fund: Annual Distributions are equal to 5% of the
five-year average market value. As this rule is being phased in, the distribution will be 5.8% for fiscal years 2005
to 2012, 5.5% for fiscal years 2013 to 2016, and 5% thereafter.
• Texas Permanent School Fund: Distributions are determined by the State Board of Education based on the
projected return of the current fiscal year, as well as the realized returns during the nine previous fiscal years.
Any one-year distribution shall not exceed 6% of the average market value of the total fund.
• Utah School and Institutional Trust Lands: Interest and dividend income are distributed, while market gains
remain in the fund. Each of twelve beneficiaries receives a specific percentage of total interest and dividend
income.
• Wyoming Permanent Land Funds: A specific acreage of trust lands is assigned to each beneficiary, and the
revenue generated from those lands is deposited into the corresponding fund. The state legislature can vote to
distribute a portion of the Public School Permanent Land Fund each year.

Asset Allocation
Review
34


Projected Future Distributions Under Very Pessimistic Case1
Current Policy Alternative A Alternative B

1,200
Total
Distributions
($ mm)

1,000 Current Policy 1,078
Alternative A 1,686
Annual Distibution ($ Millions)

Alternative B 2,366
800

600

400

200

0
3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2
1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3
20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
• Using the very pessimistic scenario, distributions under the Current Policy would vary from zero to nearly
$200 million during the twenty-year period.
1
Actual data used for 2007-2011, and projected data used thereafter.
Asset Allocation
Review
35


Projected Future Distributions Under Pessimistic Case1

1,200 Total
Distributions
($ mm)

Alternative A 1,636

Alternative B 2,452

800

600

400

200

0
3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2
1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3
20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
• Using the pessimistic scenario, distributions under the Current Policy would vary from zero to nearly $600 million
during the twenty-year period.
1
Asset Allocation
Review
36


Projected Future Distributions Under Moderate Case1

1,200 Total
Distributions
($ mm)

Alternative A 1,974

Alternative B 2,725

800

600

400

200

0
3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2
1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3
20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
• Using the moderate scenario, distributions under the Current Policy would vary from zero to nearly $800 million
during the twenty-year period, and would total over $1 billion more than the alternative policies.
1
Asset Allocation
Review
37


Projected Future Distributions Under Optimistic Case1

1,200 Total
Distributions
($ mm)

Alternative A 2,161

Alternative B 2,454

800

600

400

200

0
3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2
1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3
20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
• Using the optimistic scenario, distributions under the Current Policy would vary from zero to over $800 million
during the twenty-year period, and would be nearly double those of the alternatives over the entire period.
1
Asset Allocation
Review
38


Projected Future Distributions Under Very Optimistic Case1

1,200 Total
Distributions
($ mm)

Alternative A 2,793

Alternative B 2,722

800

600

400

200

0
3

4

5

6

7

8

9

0

1

2

3

4

5

6

7

8

9

0

1

2
1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3
20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20

20
• Using the very optimistic scenario, distributions under the Current Policy would vary from zero to nearly
$800 million during the twenty-year period.
1
Asset Allocation
Review
39


Projected Future Distribution Characteristics Under Each Economic Scenario, Using Current Distribution Policy
Very Pessimistic Scenario Current
20-Year Period Policy Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3

Average Annual Distribution ($ millions) 54 53 60 57 60 39 43

Standard Deviation of Distribution ($ millions) 67.1 67.8 76.0 72.7 76.8 57.5 59.8

Range of Distribution ($ millions) 175 184 212 206 217 208 211

Number of Years With No Distribution 10 10 10 10 7 10 10

Pessimistic Scenario Current





Moderate Scenario Current





Asset Allocation
Review
40


Projected Future Distribution Characteristics Under Each Economic Scenario (continued)
Optimistic Scenario Current





Very Optimistic Scenario Current





• Under the current distribution policy, there is little difference in the size or variability of distributions as a result of
the asset allocation policy selected.

Asset Allocation
Review
41

Analysis of Fiscal Year Change

Analysis of Fiscal Year Change

Fiscal Year Analysis1
50% Equity/50% Bond Allocation since 1926
Standard
Average Deviation Maximum Minimum
Annual of Annual Annual Annual
Return Return Return Return
Fiscal Year Ending (%) (%) (%) (%)

March 31 8.6 12.8 45.3 -30.0

June 30 8.5 13.7 77.2 -40.9

September 30 8.3 10.7 30.9 -23.2

December 31 8.3 10.7 28.7 -23.7

• Historically, a fiscal year of June 30 has proved to be the most volatile end point of the four calendar quarters.

1
Equity returns reflect the S&P 500 Index, while bond returns reflect the intermediate government bond yield from 1926 to 1976, and the Barclays Aggregate Index return thereafter, rebalanced monthly.
Asset Allocation
Review
43

Conclusions

• The Trustees should assess the current risk profile of the Fund and reconfirm, or modify, the asset allocation
policy based on their determination of the appropriate level of risk.
• The Trustees should consider diversifying the Fund into additional asset classes to benefit from modest
improvement to the Fund’s long term risk/return profile.
− Shifting a modest amount of the Fund's U.S. equity exposure to non-U.S. equities moderately improved
the Fund's investment efficiency.
• The Trustees should evaluate the pros/cons of lifting some of the current restrictions on the investment of the
Fund. The unconstrained policies tend to exhibit stronger risk/return attributes.
− A change in statutory restrictions would allow the Fund to diversify into non-U.S. debt investments,
slightly improving the Fund's investment efficiency.
− A change in the state Constitution would be required to allow private market strategies, below
investment grade debt investments, and commodities.
• The Trustees should consider modifying the distribution policy, if more stable and predictable distributions over
time are an objective of the Fund.

Asset Allocation
Review
45

Detail of Deterministic Scenario Analysis


Asset Allocation
Review
47


Scenario A – Very Pessimistic Case

Economic Conditions
• The economy suffers from a major recessions intermittently throughout the twenty-year period. Unemployment
becomes a major economic concern, reaching double-digit levels for multiple years. Unemployment averages
7.4% during the period.
• Inflation is at its peak in the period’s early years, and remains relatively low. The U.S. dollar declines in value by
3.1%.

Asset Class Returns
• Bonds produce a relatively attractive return during the period, with Treasuries and corporates outperforming the
U.S. equity market over the entire period.
• Equity markets are volatile throughout the period, and produce an average annual return of only 2.6%, less than
the rate of inflation. Twice during the period, in 2016 and 2026, domestic equities fall by nearly 40%.
Developed and emerging market foreign stocks perform poorly as well.

Asset Allocation
Review
48


Scenario A – Very Pessimistic Case
Current

Average Annual Return (%) 3.2 3.6 3.9 3.9 4.1 2.4 2.2

Standard Deviation (%) 11.8 10.7 11.2 10.7 10.9 11.2 12.5

Maximum Drawdown (%) -24.5 -22.6 -24.0 -23.3 -23.6 -26.2 -30.3

Ending Period Value ($ millions)1 5,607 6,089 6,480 6,450 6,721 4,866 4,648

• In a very pessimistic scenario, the more conservative policies outperform the riskier policies.

1
Assumes no cash inflows into, or out of, the Fund.
Asset Allocation
Review
49


Scenario B – Pessimistic Case

Economic Conditions
• Weak and inconsistent economic growth characterizes the period. A prolonged recession occurs in 2021-2023,
followed by a recovery. Domestic unemployment remains an economic problem throughout the period,
averaging 6.4%.
• Inflation peaks around 12% in 2022, during the recession. Long-term Treasury yields are low at the beginning of
the time period, then increase after the recession and remain high until the end of the period.

Asset Class Returns
• Bond returns vary considerably during the period. As yields rise in the second half of the period, bonds produce
a healthy return.
• Equities produce a negative return in 2016, and again in 2021-2022 during the recession. They experience a
“bull run” from 2027-2031.

Asset Allocation
Review
50


Scenario B – Pessimistic Case
Current





• In a pessimistic scenario, the current policy underperforms the alternative policies.

1
Assumes no cash inflows to, or outflows from, the Fund.
Asset Allocation
Review
51


Scenario C – Moderate Case

Economic Conditions
• Overall, GDP grows at a normal rate, with frequent but shallow economic cycles. Unemployment is held largely
in check, peaking around 8% during cyclical downturns.
• Inflation is high (10%) at the beginning of the time period, then moderates after that. Long-term Treasury yields
move within a range of 3% to 9%.

Asset Class Returns
• Bond returns reflect the moderate interest rate environment. Negative returns occur only during a few periods of
economic stress.
• Equities exhibit normal levels of volatility and produce an average annual return of about 8%. A “bear market”
afflicts equity investors in 2016-2017.

Asset Allocation
Review
52


Scenario C – Moderate Case
Current





• In a moderate case, Policy U-3 has the highest ending value, approximately $2.5 billion more than the Current
Policy.

1
Asset Allocation
Review
53


Scenario D – Optimistic Case

Economic Conditions
• Economic growth is strong and persistent, interrupted only by a short recession in 2016. Unemployment is
regularly maintained at or below the 6% level.
• Inflation is generally low and controlled, except for one high inflation year in 2022. Strong economic growth
prevents long-term interest rates from going too low.

Asset Class Returns
• Bond returns reflect the low and controlled inflation environment, averaging over 6% for both Treasuries and
corporates.
• The strong economy drives equity markets both in the U.S. and overseas. Over the entire period, U.S. stocks
produce an average return of 9.8% per year, outpacing inflation, while private equity returns are very high.
Foreign emerging equities outpace foreign developed equities.

Asset Allocation
Review
54


Scenario D – Optimistic Case
Current





• In the optimistic scenario, all policies perform very well.

1
Asset Allocation
Review
55


Scenario E – Very Optimistic Case

Economic Conditions
• The global economy enters a period of enhanced productivity and economic growth, very much like that
experienced by America in the 1950s and 1960s. U.S. real GDP grows at an average rate of 3.5%. After
recovering from the current environment, the economy spends most of its time at or beyond what is now
typically termed “full employment.” The unemployment rate averages 5.3%.
• After a couple of high years in 2012 and 2016, inflation is controlled throughout the period. Treasury yields are
stable, reflecting low inflation and a stable economy.

Asset Class Returns
• Bond returns are strong, with Treasuries averaging 6.8% and corporate bonds averaging 8.9%.
• Equities produce a superb “real” return, as the economy’s increased productivity results in high earnings growth
for corporate America. For the entire period, U.S. stocks return 13.7% per year. Riskier market segments of the
equity markets (private equity, high yield bonds, real estate, and emerging market equities) perform impressively
as well.

Asset Allocation
Review
56


Scenario E – Very Optimistic Case
Current




Ending Period Value ($ millions) 1 23,862 22,044 21,833 20,109 20,542 26,832 29,208

• In the very optimistic scenario, the least constrained asset allocation policy performs best.

1
Assumes no cash inflows into, or out of, the Fund.
Asset Allocation
Review
57

Additional Risk Analytics


Asset Allocation
Review
58


Types of Risk Analysis Addressed

• Risk budgeting1
− Attributes overall portfolio risks to specific asset classes
− Highlights the source and scale of portfolio-level risk
• MVO-based risk analytics
− Includes worst-case return expectations
− Relies on assumptions underlying MVO
• Stress Testing
− Stress tests policy portfolios using actual historical examples
− Stress tests policy portfolios under specific hypothetical scenarios

1
Risk budgeting seeks to decompose the aggregate risk of a portfolio into different sources (in this case, by asset class), with risk defined as standard deviation.
Asset Allocation
Review
59


Risk Budgeting Analysis1
Capital Allocation vs. Risk Allocation
Equities Fixed Income Real Assets

Current R1 R2 R3 U1 U2 U3
100%

80%

60%
Allocation

40%

20%

0%
Capital Risk Capital Risk Capital Risk Capital Risk Capital Risk Capital Risk Capital Risk

· Assets with low relative volatility, such as fixed income, contribute less to risk than their asset weights imply.
1
Risk allocation is calculated by multiplying the weight of the asset class by its standard deviation and its correlation with the total portfolio and then dividing this by the standard deviation of the total portfolio.
Asset Allocation
Review
60


Risk Budgeting Analysis1
Absolute Contribution to Risk
14

12
Standard Deviation (%)
10

8

6

4

2

0
Current
R1 R2 R3 U1 U2 U3
Policy
Real Assets 0.0 1.1 1.4 1.2 1.4 2.7 2.4
Fixed Income 1.0 1.0 0.9 0.8 1.1 1.8 1.4
Equities 10.0 8.2 8.5 9.2 8.5 6.5 8.9

· In each policy option, equity risk dominates the risk profile of the portfolio, though the alternative policies reduce
equity risk somewhat.
1
Contribution to risk is calculated by multiplying the weight of the asset class by its standard deviation and its correlation with the total portfolio.
Asset Allocation
Review
61


“Worst Case” Return Projections1
Current Allocation Policy R-1 Policy R-2 Policy R-3
Policy U-1 Policy U-2 Policy U-3
0%

-5%

-10%
Return

-15%

-20%

-25%
1 Year 3 Years 5 Years 10 Years

Current Allocation Policy R-1 Policy R-2 Policy R-3 Policy U-1 Policy U-2 Policy U-3
(%) (%) (%) (%) (%) (%) (%)
One Year -16.5 -15.0 -15.8 -16.5 -16.2 -15.5 -18.2
Three Years -8.6 -7.6 -8.0 -8.5 -8.3 -7.6 -9.3
Five Years -5.7 -4.9 -5.2 -5.5 -5.4 -4.7 -6.0
Ten Years -2.6 -1.9 -2.1 -2.3 -2.2 -1.5 -2.4

· Policies R-1 and U-2 best defend the portfolio in a “worst case” scenario, as defined by MVO model assumptions.

1
“Worst Case” Return Projections encompass >99% of possible outcomes.
Asset Allocation
Review
62


Probability of Experiencing Negative Returns
30%

25%

20%
Probability

15%

10%

5%

0%
1 Year 3 Years 5 Years 10 Years

(%) (%) (%) (%) (%) (%) (%)
One Year 26.9 25.0 25.3 25.8 25.5 23.3 25.3
Three Years 16.1 13.8 14.1 14.7 14.4 12.0 14.2
Five Years 10.4 8.3 8.6 9.2 8.8 6.7 8.7
Ten Years 3.9 2.7 2.8 3.1 2.9 1.8 2.9

· Only the current allocation has a more than a 10% chance of producing a negative return over a five-year period.

Asset Allocation
Review
63


Probability of Achieving a 5% Annualized Return1
100%

90%

80%

70%

60%
Probability

50%

40%

30%

20%

10%

0%
1 Year 5 Years 20 Years

(%) (%) (%) (%) (%) (%) (%)
One Year 53.4 54.2 54.9 55.1 55.2 57.8 57.9
Five Years 55.4 56.8 57.9 58.2 58.3 62.4 62.6
Ten Years 56.9 58.6 60.0 60.4 60.6 66.9 65.9
Twenty Years 63.3 66.5 69.1 69.8 70.1 78.5 78.9

· Policy U-3 has the highest likelihood of producing a 5% annualized return over all time periods.

1
Represents the probability of achieving a 5% return over the specified time horizon.
Asset Allocation
Review
64


Historical Stress Test: Interest Rate Spike (1994)
(Cumulative Return)
3.0% 2.8%

2.5%

2.0%
1.7%

1.5%

1.0%
0.6%
0.5% 0.4%
0.1%
0.0%

-0.5% -0.4%

-1.0%
-1.1%
-1.5%

• In a period of rising interest rates (e.g., 1994), Policy U-3 produces the highest return, as it has the lowest
allocation to investment grade bonds and the highest allocation to private equity.

Asset Allocation
Review
65


Historical Stress Test: Crash of 1987 (September thru November 1987)
(Cumulative Return)
0%

-2%

-4%

-6%

-8% -7.3%

-10% -8.9%

-12% -11.1%
-11.6%
-12.0%
-13.2%
-14%

-16%

-18% -17.0%


• In a down market environment that differs from 2008 (e.g., the Crash of ‘87), the alternative policies would
produce higher returns relative to the current allocation.

Asset Allocation
Review
66


Historical Stress Test: Bursting of the dot.com Bubble (2Q00 thru 3Q02)
(Cumulative Return)
0%

-2%

-4%
-4.1%

-6%

-6.9%
-8% -7.6%
-8.1%
-8.9%
-10%
-10.2%
-12%

-14%
-14.4%
-16%

• In another down market environment that differs from 2008 (e.g., the bursting of the dot.com bubble), the
alternative policies produce higher returns relative to the current allocation.

Asset Allocation
Review
67


Historical Stress Test: Strong US Dollar (1Q81 through 3Q82)
(Cumulative Return)
6.0%

5.3%
5.2%
5.0%

4.2%

4.0% 3.6%
3.4%

3.0%
2.5%

2.0%

1.0%
0.6%

0.0%

• In a period of an appreciating U.S. dollar (e.g., the early 1980’s), the alternative policies generally produce lower
returns relative to the current allocation due to their allocations to foreign assets.

Asset Allocation
Review
68


Historical Stress Test: Weak US Dollar (January 1986 thru August 1987)
(Cumulative Return)
35%
32.0% 32.2%

29.6% 29.1%
30%

25.9% 26.1%
25%
22.3%

20%

15%

10%

5%

0%

• In a period of a declining US dollar (e.g., the middle 1980’s), the alternative policies produce higher returns
relative to the current allocation due to their allocations to foreign assets.

Asset Allocation
Review
69


Historical Stress Test: Stagflation (1Q73 thru 3Q74)
(Cumulative Return)
0%

-5%

-10%

-15%

-17.3%
-20%

-21.1% -21.3%
-21.6% -21.4%
-21.8%
-23.6%
-25%

• In an extended environment of high inflation but low growth (e.g., the middle 1970’s), the alternative policies
protect better than the current allocation due to their allocations to real assets.

Asset Allocation
Review
70

Asset Class Definitions


Asset Allocation
Review
71



Domestic Equity
• Investments in publicly traded U.S. companies. There are over 6,000 investable U.S. stocks.
• Historically, have provided the main engine of growth for investors.
• Includes huge multinational companies and very small operations. Some investors consider large cap and small
cap stocks to be different asset classes, due to their varying performance cycles.

International Developed Equity
• International equity investing refers to investing in companies domiciled outside of the U.S.
• Developed foreign markets predominantly include the countries of Western Europe and Japan.
• International equity markets provide opportunities to add value through active management, by allocating
holdings across markets and within markets.

Emerging Markets Equity
• Proponents of emerging markets support the thesis that the most rapid economic growth in the coming decades
will occur in less developed nations (e.g., the BRIC countries).
• Emerging markets have produced higher returns than developed markets historically.
• However, emerging markets have been far more volatile, having experienced three calendar years with declines
in excess of 25% since 1998.

Asset Allocation
Review
72


Frontier Markets Equity
• Frontier markets represent stock markets in underdeveloped countries. Frontier markets feature young and
fast-growing populations, and these countries are generally net natural resource exporters.
• Frontier markets are characterized by illiquidity, low transparency, low levels of foreign investment, and immature
regulation.
• Of more than 115 stock markets worldwide, over 60 may be characterized as frontier markets. As of 2009, these
60 carried a market capitalization of $1.1 trillion, or 2% of world market capitalization.

Private Equity
• Investments in privately held companies; generally structured in the form of partnerships, consisting of ten to
twenty investments.
• Historically, private equity has returned 2% to 3% per year more than public equity.
• Although they are self-liquidating, usually over periods of eight to ten years, private equity partnership interests
are not generally traded on a short-term basis.

Investment Grade Bonds
• Reduce portfolio volatility, provide stability in crisis environments, provide diversification benefits, and serve as a
source of liquidity.
• An investment grade bond portfolio can be structured to target a certain sector, duration, and quality rating.
• The intermediate-term duration associated with a broad bond market index is likely to provide an investor with
the best risk/reward tradeoff.

Asset Allocation
Review
73


Treasury Inflation-Protected Securities (TIPS)
• Inflation-linked bonds offer investors a guaranteed return over inflation, if held to maturity.
• Investors receive an explicit annual coupon plus a variable adjustment based on the rate of inflation, providing
inflation protection.
• TIPS should produce returns very close to nominal Treasuries.
• Since their inception in 1997, TIPS have exhibited a level of volatility similar to that of nominal Treasuries.
• TIPS should be uncorrelated with public equities and only modestly correlated with nominal bonds.

High Yield Bonds
• High yield bonds, also known as “junk bonds,” are usually issued by corporations rated below investment grade.
• Three main types of risk affect the high yield bond market: liquidity risk, interest rate risk, and default risk.
• Even though they are often referred to as “junk” bonds, high yield bonds have historically been less risky than
public equities.
• High yield bonds have been only modestly correlated with investment grade bonds and public equities.

Asset Allocation
Review
74


Bank Loans
• Investable bank loans are senior floating-rate loans made to speculative-grade issuers that theoretically constitute
a safer alternative to high yield bonds. Bank loans are typically secured by company assets.
• Because bank loans pay a floating interest rate, they provide a hedge against rising short-term interest rates and
potentially a hedge against inflation.
• The secondary bank loan market has grown rapidly in the last decade, due primarily to their popularity as a
source of financing for mergers and acquisitions. Like high yield bonds, they experienced a sharp sell-off in 2008.

Emerging Market Debt
• Emerging market debt can be divided into two broad categories: “External” debt is issued in currencies other than
the country’s home currency (i.e., U.S. Dollars or Euros); “Local” debt is issued in the local currency of the issuing
country or company.
• The addition of an emerging markets debt allocation provides diversification benefits to a portfolio, either on a
stand-alone basis or as a complementary strategy to a high yield bond portfolio.
• The risks of investing in emerging market debt include political, event (i.e., crises) and currency volatility.

Core Real Estate
• Real estate investing entails the direct or indirect ownership of physical property or land.
• This is a hybrid asset, exhibiting both equity and fixed income characteristics; it provides increased diversification
and helps moderate aggregate returns over time. Relative to stocks and bonds, real estate offers moderate
inflation protection.
• Core real estate is generally liquid and developed, and achieves a high percentage of its return from income.
Asset Allocation
Review
75


Value-Added Real Estate
• Value-added real estate includes real estate that achieves a significant portion of its return from an appreciation in
value.
• Value-added real estate typically uses a higher degree of leverage and has higher volatility of returns than core
real estate. Value-added real estate can also include core properties that are not highly leased.
• Major property types include specialty retail, hotels, assisted living, storage, and low-income housing.

Opportunistic Real Estate
• Opportunistic real estate is expected to derive most of its return from property appreciation.
• Relative to core and value-added real estate, opportunistic real estate typically uses a higher degree of leverage
and has higher volatility of returns.
• Opportunistic real estate includes nontraditional property types, including speculative development and land.

Infrastructure
• Infrastructure is the underlying foundation of basic services, facilities, and institutions upon which a community
depends.
• Infrastructure investments include utilities, transportation, communications, and social institutions like hospitals
and prisons.
• These types of investments typically have large barriers to entry and long-duration contracts, which can be
advantageous for matching the long-term liabilities or spending needs of institutional investors.

Asset Allocation
Review
76


Natural Resources
• Natural resources are essentially products of the Earth (e.g., oil, coal, wheat, timber, water, wind, etc.). Natural
resources investments are holdings in companies which are involved with the following activities:
− Extracted resources: oil, natural gas, coal, industrial and precious metals
− Harvested resources: agricultural production, ownership of farm- or timberland
− Renewable energy: solar, biofuels, wind, hydro, and geothermal
• Historical data suggest (1) that public market natural resources strategies should perform best during periods of
high inflation, and (2) that natural resources should produce returns in excess of those of broader public equities.

Commodities
• Commodities are generally physical goods or raw materials.
• Commodities may provide three benefits: increased portfolio diversification, a modest hedge against consumer
price inflation, and a hedge against unique economic risks-including currency devaluation and armed conflict.
• Empirical work suggests that trend-following and momentum strategies can add value in commodities.

Asset Allocation
Review
77

Risk/Return Analysis Assumptions


Asset Allocation
Review
78


Scenario Return Inputs
Asset Class Benchmark Used1
Investment Grade Bonds Barclays Aggregate
TIPS Barclays U.S. TIPS
EM Bonds JPM GBI-EM Global Diversified
Bank Loans CSFB Leveraged Loan
High Yield Bonds Barclays High Yield
Core Real Estate NCREIF Property
Value-Added RE NCREIF Townsend Value Added
Opportunistic RE NCREIF Townsend Opportunistic
REITs NAREIT Equity
Infrastructure (private) S&P Global Infrastructure
Natural Resources (private) S&P Global Natural Resources
Commodities Summer Haven Commodity
US Equity Russell 3000
Public Foreign Equity (Developed) MSCI EAFE
Public Foreign Equity (Emerging) MSCI Emerging Markets
Long-short Equity HFRI Equity Hedge
Private Equity Venture Economics Private Equity Composite

1
For U.S. Equity, we used the S&P 500 prior to 1979; for Investment Grade Bonds, we used Ibbotson’s US Intermediate Government series prior to 1976; for EM Bonds, we used the JP Morgan EMBI+ prior to
2003; for Infrastructure, we used the S&P Utilities prior to 2002; for Natural Resources, we used a 75/25 mix of S&P Energy and S&P Diversified Metals and Mining prior to 2003.
Asset Allocation
Review
79


Notes and Disclaimers
1
The returns shown in the Policy Options and Risk Analysis sections rely on estimates of expected return, standard deviation, and correlation
developed by Meketa Investment Group. To the extent that actual return patterns to the asset classes differ from our expectations, the results
in the table will be incorrect. However, our inputs represent our best unbiased estimates of these simple parameters.

2
The returns shown in the Policy Options and Risk Analysis sections use a lognormal distribution, which may or may not be an accurate
representation of each asset classes’ future return distribution. To the extent that it is not accurate in whole or in part, the probabilities listed
in the table will be incorrect. As an example, if some asset classes’ actual distributions are even more right-skewed than the lognormal
distribution (i.e., more frequent low returns and less frequent high returns), then the probability of the portfolio hitting a given annual return
will be lower than that stated in the table.

3
The standard deviation bars in the chart in the Risk Analysis section do not indicate the likelihood of a 1, 2, or 3 standard deviation
event-they simply indicate the return we expect if such an event occurs. Since the likelihood of such an event is the same across allocations
regardless of the underlying distribution, a relative comparison across policy choices remains valid.

Asset Allocation
Review
80


Overview of Annual Asset Study Methodology

• In order to construct an optimal portfolio from a risk-return standpoint, conventional financial wisdom dictates that
one develops return, volatility, and correlation expectations over the relevant investing horizon.
• Given the uncertainty surrounding financial and economic forecasts, expectations development is challenging, and
any of several methodological approaches may meaningfully contribute to this complex task.
• Meketa Investment Group’s process relies on both quantitative and qualitative methodologies.
• First, we employ a large set of quantitative models to arrive at a set of baseline expected ten-year annualized returns
for major asset classes.
• These models attempt to forecast a gross “beta” return for each asset class.
• Our models may be econometrically derived (that is, based on a historical return relationship with current
observable factors), factor-based (that is, based on a historical return relationship with predicted factors), or
fundamentally based (that is, based on some theoretically defined return relationship with current observable
factors).
• Some of these models are more predictive than others, for this reason, we next overlay a qualitative analysis, which
takes the form of a data-driven deliberation among the asset allocation team (comprised of senior members of the
firm).
• As a result of this process, we form our ten-year annualized return expectations, which serve as the primary
foundation of our longer-term, twenty year expectations.
• We form our twenty-year annualized return expectations by systematically considering historical returns on an asset
class by asset class level.

Asset Allocation
Review
81


Overview of Annual Asset Study Methodology (continued)

• Depending on our confidence in the historical average, we will modify the weightings of the ten-year forecast and
historical average returns.
• Finally, we develop our twenty-year volatility and correlation expectations, relying primarily on various historical
averages -- qualitative adjustments, when applied, usually serve to increase the correlations and volatility over
and above the historical estimates (e.g., using the higher correlations usually observed during a volatile market).
• These volatility and correlation expectations are then combined with our twenty-year return expectations to assist
us in subsequent asset allocation work, including mean-variance optimization and scenario analyses.
• We review and make modifications to the inputs, based on changing market dynamics, on a quarterly basis.
• Throughout the process, we remind ourselves of our overarching goals:
− Consistency of results with historical experience and fundamentals
− Consistency of results with macroeconomic reality
− Consistency of results across asset classes
− Recognition of forecasting error and its implications

Asset Allocation
Review
82


Meketa Investment Group 2011 Annual Asset Study
Inputs Summary: 20-Year Assumptions
Annualized Annualized
Return Standard Deviation
Asset Class (%) (%)
Fixed Income
Cash Equivalents 3.1 1.5
Short-Term Investment Grade Bonds 3.3 2.5
Investment Grade Bonds 3.8 5.5
Long-term Government Bonds 4.4 8.3
TIPS 4.2 7.0
High Yield Bonds 6.4 12.0
Bank Loans 5.9 11.0
Foreign Bonds (unhedged) 4.9 12.0
Emerging Market Bonds (local; unhedged) 7.3 15.0
Equities
Public Domestic Equity 8.1 17.0
Public Foreign Equity (Developed) 8.1 19.0
Public Foreign Equity (Emerging) 10.5 26.0
Private Equity 10.5 25.0
Hedge Funds 6.4 10.5
Real Assets
Real Estate 8.4 17.0
REITs 6.6 19.0
Core Private Real Estate 7.8 13.0
Value Added Real Estate 8.8 17.0
Opportunistic Real Estate 10.4 25.0
Natural Resources (Private) 9.6 22.0
Commodities 6.0 23.0
Infrastructure (Private) 8.5 18.0

Asset Allocation
Review
83


Meketa Investment Group 2011 Annual Asset Study
Inputs Summary: Correlations for Major Asset Class
Public Public
Investment High Natural Public Foreign Foreign
Grade Yield Resources Domestic Equity Equity Private Infrastructure Hedge Real
TIPS Bonds Bonds (private) Commodities Equity (developed) (emerging) Equity (private) Funds Estate

TIPS 1.00

Investment
0.80 1.00
Grade Bonds

High Yield
0.30 0.35 1.00
Bonds

Natural
Resources 0.20 0.20 0.45 1.00
(private)

Commodities 0.30 0.15 0.05 0.7 1.00

Public
0.00 0.25 0.65 0.35 0.05 1.00
Domestic Equity

Public
Foreign Equity 0.10 0.20 0.55 0.30 0.10 0.80 1.00
(developed)
Public
Foreign Equity 0.10 0.00 0.50 0.50 0.50 0.75 0.80 1.00
(emerging)
Private
0.00 0.15 0.65 0.50 0.05 0.80 0.70 0.60 1.00
Equity
Infrastructure
0.30 0.35 0.45 0.50 0.20 0.40 0.35 0.35 0.50 1.00
(private)
Hedge
0.10 0.25 0.60 0.30 0.30 0.70 0.70 0.65 0.55 0.35 1.00
Funds

Real Estate 0.10 0.25 0.50 0.50 0.10 0.40 0.35 0.30 0.45 0.45 0.35 1.00

Asset Allocation
Review
84


Stress Test Return Assumptions1
BBB BBB
Rates Rates Rates Spreads Spreads USD USD Equities Equities Equities
rise rise rise widen by widen by Gains Gains Decline Decline Decline Rates fall Rates fall
100 bp 200 bp 300 bp 50 bp 300 bp 10% 20% 10% 25% 40% 100 bp 200 bp
Public Domestic Equity 11.3% 15.8% 6.9% 6.0% -42.0% 3.5% 7.0% -10.0% -25.0% -40.0% 10.5% 8.4%
Public Foreign Equity (Developed) 20.3 23.6 4.8 5.5 -33.0 -7.0 -14.0 -10.5 -26.3 -42.0 0.5 10.0
Public Foreign Equity (Emerging) 20.3 23.6 4.8 5.0 -39.0 -7.0 -14.0 -11.0 -27.5 -44.0 4.4 9.0
Long-Short Hedge Funds 11.5 12.8 4.2 6.5 -21.0 2.1 4.2 -6.0 -15.0 -24.0 13.6 6.8
Private Equity 11.3 15.8 6.9 6.0 -42.0 3.5 7.0 -8.0 -20.0 -32.0 10.5 8.4
Core Real Estate 11.4 12.2 17.4 9.5 -12.0 4.0 8.0 -5.0 -12.5 -20.0 5.5 5.2
REITs 19.3 12.8 16.8 0.5 -36.0 1.0 2.0 -9.5 -23.8 -38.0 14.9 7.4
Non-Core Real Estate 8.3 13.0 17.4 11.5 -24.0 4.0 8.0 -8.0 -20.0 -32.0 6.9 7.2
Infrastructure (private) 14.0 6.6 5.7 3.5 -24.0 3.0 6.0 -5.0 -12.5 -20.0 7.2 7.1
Natural Resources (private) 11.4 18.4 14.4 2.0 -16.5 -3.1 -6.2 -5.0 -12.5 -20.0 5.0 0.8
Natural Resources (public) 22.8 36.8 28.8 4.0 -33.0 -6.2 -12.3 -9.5 -23.8 -38.0 10.0 1.6
Commodities 12.6 9.6 -0.6 -0.5 -21.0 -15.0 -30.0 -7.0 -17.5 -28.0 1.8 -4.8
Long-Term Government Bonds -12.7 -29.6 -46.5 12.0 15.0 10.0 20.0 5.0 12.5 20.0 21.1 38.0
TIPS -2.3 -7.6 -12.9 8.5 12.0 8.0 16.0 1.0 2.5 4.0 8.3 13.6
Investment Grade Bonds -2.3 -7.3 -12.3 -0.1 -3.6 8.0 16.0 2.0 5.0 8.0 7.7 12.7
Investment Grade Corporate Bonds -2.9 -9.5 -16.1 -0.3 -14.9 8.0 16.0 -1.5 -3.8 -6.0 10.3 16.9
Foreign Developed Bonds -4.1 -10.7 -17.3 0.4 -2.4 -6.3 -12.6 -2.0 -5.0 -8.0 9.0 15.5
Emerging Market Bonds (external) 0.0 -6.2 -12.4 -1.4 -25.7 5.0 10.0 -2.0 -5.0 -8.0 12.4 18.6
Emerging Market Bonds (local) 1.8 -2.8 -7.4 3.0 -10.5 -6.3 -12.6 -3.0 -7.5 -12.0 11.0 15.6
High Yield Bonds 2.5 -2.0 -6.4 -3.3 -33.5 4.5 9.0 -6.0 -15.0 -24.0 11.5 16.0
Bank Loans 12.0 12.0 15.0 2.5 -30.0 4.5 9.0 -6.0 -15.0 -24.0 4.3 -2.8
Hedge Funds 6.6 8.4 3.6 3.5 -18.0 5.0 10.0 -5.0 -12.5 -20.0 8.1 4.4
TAA 10.2 11.4 2.6 6.5 -22.2 3.2 6.4 -7.0 -17.5 -28.0 8.8 12.2
Risk Parity 7.3 5.0 -2.5 5.6 -12.0 1.6 3.3 -2.0 -5.0 -8.0 10.2 12.3

1
Assumptions are based on performance for reach asset class during historical periods that resembled these situations.
Asset Allocation
Review
85

Asset Allocation Review

More Related Content

Similar to Asset Allocation Review (20)

More from dougducey (9)

Asset Allocation Review