Desired Target Return® An Upside Potential/Downside Risk Framework

Optimal Blending of Active and Passive Investments

A Desired Target Return® Briefing

Table of Contents DESIRED TARGET RETURN® OVERVIEW...................................................................................................3 Modern Portfolio Theory (MPT) Meets Post Modern Portfolio Theory (PMPT) Five Fundamental Investment Beliefs

DESIRED TARGET RETURN® PORTFOLIO CONSTRUCTION METHODOLOGY ....................9 THE DTR® CALCULATION................................................................................................................................10 Identifying the DTR® Calculation Risk/Return Assessment and Measurements Upside Potential Downside Risk Deviation DTR®-Alpha

SURZ STYLE PURE® INDICES.......................................................................................................................12 Better Built Indices/Better Built Portfolios Mutually Exclusive, Exhaustive, Investable and Macro Consistent Significance of Centric — Segment Between Value and Growth — “Blend” Definition Difference

BOOTSTRAPPING TECHNIQUE...................................................................................................................14 Manager Statistical Data Data Interpretation

DESIRED TARGET RETURN® SCREENING PROCEDURES ............................................................16 DESIRED TARGET RETURN® PORTFOLIO CONSTRUCTION INTEGRATION.......................18 SUMMARY ..............................................................................................................................................................21 GLOSSARY...............................................................................................................................................................22

PAGE 2 • DTR®

DISCLAIMER ..........................................................................................................................................................27

D E S I R E D TA R G ET R ET U R N

®

Overview Desired Target Return® portfolio construction provides a revolutionary advancement in the evolution of portfolio construction. This advanced Upside Potential/Downside Risk framework of portfolio construction methodology is based on three Nobel Prize winning theories and other advancements in Post Modern Portfolio Theory (PMPT) as briefly explained in this white paper. The Desired Target Return® portfolio construction focus begins with identifying the DTR® calculation that links the investor’s assets to their financial goals or liabilities. Sortino Investment Advisors (SIA) apply the Upside Potential/Downside Risk framework as an overlay, which fortifies any Global Investment Committee’s strategic asset allocation, tactical asset allocation, consultant’s active manager research and consulting services. Desired Target Return® offers professionally managed portfolios tailored to an Institutional or Private Client’s specific financial goals. The Upside Potential/Downside Risk (UP/DR) framework provides the “Missing Link” to Modern Portfolio Theory (MPT). This is made possible by integrating multiple contributing advancements recognized as Post Modern Portfolio Theory (PMPT) into the Desired Target Return® portfolio construction methodology discussed in this paper.

Five Fundamental Investment Beliefs Desired Target Return® has five basic beliefs that establish the foundation for how we approach and structure client portfolios. These five beliefs are: BELIEF #1: The Desired Target Return® goal is to meet or exceed the investor’s DTR® calculation. The goal is an absolute return objective and not to beat arbitrary market index(es). BELIEF #2: The investor’s primary risk is not achieving the Desired Target Return®. Short term volatility is a secondary risk consideration. BELIEF #3: Most active managers are not style pure. Style mix should be accounted for in the portfolio construction process.

PAGE 3 • DTR®

BELIEF #4: Many “Risk Tolerance” profiling techniques and questionnaires ignore the Desired Target Return®. This investor self-assessed risk profiling technique is subject to their emotional bias, which is influenced by varying perceptions of the capital markets and economic environments. BELIEF #5: The DTR® calculation is event driven by capital markets and/or investor circumstantial change(s). Event Driven Investing (EDI) is designed to help avoid the emotional hazards of market timing and other potentially destructive investment decision making temptations.

Overview

BELIEF #1: The Desired Target Return® goal is to meet or exceed the investor’s DTR® calculation. The goal is an absolute return objective and not to beat arbitrary market index(es), as illustrated in Figure 1. Progress Is Measured to the DTR® Goal Figure 1: Meaningful Return Measurements

...But you miss your DTR® goal

Outperform relative blended Market Return (MR)

Return

P

Desired Target Return (DTR®) Actual Return Relative Blended Market Return (MR)

Time

BELIEF #2: The investor’s primary risk is not achieving the Desired Target Return®. Short term volatility is a secondary risk consideration. Market volatility is secondary and more of an issue to the investor on the downside. The rationale of this second belief is illustrated in Figure 2.

Figure 2: Meaningful Volatility (risk) Measurements

Industry’s Common Risk Definition Variability

Average Return

PAGE 4 • DTR®

Upside and downside risk are treated equally

SIA Advanced Risk Definition Downside Risk Upside Potential

Desired Target Return Emphasis is on downside risk

Overview

BELIEF #3: Most active managers are not style pure. Style mix should be accounted for in the portfolio construction process. Most active managers’ portfolios are an assortment of management styles. This is somewhat intuitive because for active management to achieve excess return, commonly referenced as “alpha”, the active manager’s portfolio must be structured differently than the style benchmark mandate. For example, Figure 3 illustrates a sample of the impurity of active value equity managers. Notice the returns based style analysis of all these large, mid and small cap value equity funds. The first three (1-3) are large value funds, the next three (4-6) are mid value funds and the last three (7-9) are small value funds. Few active managers are pure to their categorized manager style.

Active Managers Are a Mix of Style Indexes Manager

U-P

DTR®

Ratio

Apha

1

0.98

2

LrgVal

LrgCen

LrgGro

MidVal MidCen MidGro MinVal MinCen MinGro

1.8%

30.00% 11.30%

1.40%

0.00%

44.90%

9.10%

0.00%

3.30%

1.27

3.5%

23.00% 1.60%

0.00%

0.00% 29.70% 30.70%

9.90%

0.00%

5.10%

3

1.31

3.7%

21.70% 32.20%

0.00%

0.00%

0.00%

34.60% 11.50%

0.00%

0.00%

4

1.61

5.1%

19.60% 0.00%

0.00%

47.70%

4.30%

1.60%

26.80%

0.00%

0.00%

5

1.30

2.1%

0.00% 18.40%

0.00%

41.50% 26.60%

0.00%

0.00%

13.50%

0.00%

6

1.35

7.1%

0.00%

0.00%

0.00%

0.00%

9.10%

17.70% 59.00% 14.20%

0.00%

7

0.70

6.0%

0.00%

0.00%

52.10%

0.00%

0.00%

0.00%

4.80%

8

0.93

9.5%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00% 42.90% 39.00% 18.10%

9

0.90

4.8%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

0.00%

8.30%

34.80%

41.40% 38.50% 20.10%

Source: Sortino Investment Advisors, 2006

Figure 3: Active Managers Style Blends

PAGE 5 • DTR®

As evidenced by William Sharpe’s Nobel Prize winning work illustrated in Figure 3, none of the portfolios are style pure. This is common for all actively managed mutual funds and separately managed accounts. This is further compounded by the fact that current asset allocation optimizers ignore active managers’ style blend. Said another way, managers are assumed to be 100% pure in their respective style category. As a result, most asset allocation strategies are immediately corrupted upon implementation. When Sortino’s Upside Potential/Downside Risk framework methodology is overlaid in the portfolio construction process this deficiency is better addressed. For example, the two charts illustrated in Figure 4 and Figure 5 (see next page) demonstrate how an asset allocation recommendation becomes corrupted under commonly used optimization with active manager implementation.

Overview

Additionally, there are compounded asset allocation implications to be considered as a result of ignoring this active manager style purity issue. Imagine this one manager’s style analysis multiplied by the number of active managers you currently have in your aggregated diversified portfolio. What is your actual asset allocation and how does it compare to the recommended asset allocation you think you have?

U.S. Equity International Equity

Asset Mix Policy

Fixed Income

Large Value Large Centric Large Growth Mid Value Mid Centric Mid Growth (10% Allocation) Small Value Small Centric Small Growth

Style Mix Policy

Europe ex UK UK Pacific ex Japan Japan Long Bonds 3-5 Ye ear Bonds 1-3 Year Bonds Tbills

Figure 4: How Manager Selection may corrupt the Asset Allocation

Manager Style Perception vs. Reality What you think you are getting*

100% Mid Growth

What you actually got*

40% Large Growth

35% Small Growth 25% Mid Growth

PAGE 6 • DTR®

Only 2.5% of the required 10% filled the mid-growth pie. *MidCap Growth 3 years ending December 31, 2004. Source: Sortino Investment Advisors, 2006

Figure 5: Recommended versus Implemented Asset Allocation

Asset allocation is critically important for attempting to achieve the investor’s Desired Target Return® and developing an effective and efficient DTR® portfolio structure. Since asset allocation strategies revolve around both asset and style mix policies, it is extremely important to recognize that active managers (and perhaps some passive solutions as well) are not pure within their respective style category. If this fact is not acknowledged, the intended asset allocation strategy will be corrupted and not actually implemented.

Overview

BELIEF #4: Many “Risk Tolerance” profiling techniques and questionnaires ignore the Desired Target Return®. This investor self-assessed risk profiling technique is subject to their emotional bias, which is influenced by varying perceptions of the capital markets and economic environments.

Investment Fund A Different Needs, Different Portfolios

Goal: Annual Need: CPI: Expenses: Fund Assets:

Investment Fund B

Fund research $500,000 2.5% 0.5% $10,000,000

Goal: Annual Need: CPI: Expenses: Fund Assets:

DTR® = 8% 4% Req. Return Portfolio

6% Req. Return Portfolio

8% Req. Return Portfolio

Fund research $500,000 2.5% 1.0% $6,000,000

DTR® = 12% 10% Req. Return Portfolio

12% Req. Return Portfolio

These charts are for illustrative purposes only and do not reflect any actual investment product.

Figure 6: Different Desired Target Return® determines portfolio constructs The investor’s Desired Target Return® is more essential and uniquely different than some market benchmark’s volatility, target dated or “age-based” life cycle fund solutions. Falling below the Desired Target Return® is the primary investor’s risk, market volatility is a secondary risk considered more for investor behavioral and emotional factors. Yet asset allocation recommendations are frequently based primarily on market volatility (risk) tolerance to determine an investor’s portfolio structure. It should be apparent that differing Desired Target Returns® would logically dictate different portfolio constructs as illustrated in Figure 6.

BELIEF #5: The DTR® calculation is event driven by capital markets and/or investor circumstantial change(s). Event Driven Investing (EDI) is designed to help avoid the emotional hazards of market timing and other potentially destructive investment decision making temptations.

S&P 500 Market Value

PAGE 7 • DTR®

• Primary risk is not achieving the investor’s DTR® to accomplish their financial goals. • Contingency risk is market volatility. • There is an inverse relationship between substantial market events or investor circumstantial changes and their DTR® calculation as illustrated in Figure 7 below. Decreased DTR®

1800 1600 1400 1200 1000 800 600

Increased DTR®

400

Closing Value

200 0 Sep 99

Sep 00

Sep 01

Sep 02

Sep 03

Sep 04

Sep 05

Sep 06

Month Figure 7: DTR® Event Driven Investing

Sep 07

Sep 08

Overview

Typical Investment Committee’s Portfolio Construction Process Desired Target Return® integrates the typical investment consulting firm’s portfolio construction process with the Sortino Upside Potential/Downside Risk (UP/DR) framework as an overlay. The UP/DR framework is designed to select the best mix of active managers while optimally blending both active and passive management strategies. The unique Upside Potential/Downside Risk (UP/DR) overlay is integrated with a leading Global Investment Committee’s portfolio construction process that is time-tested since 1973 and illustrated in Figure 8.

PAGE 8 • DTR®

Figure 8: A leading consulting firm’s portfolio construction process

D E S I R E D TA R G ET R ET U R N

®

Portfolio Construction Methodology Desired Target Return® approaches portfolio construction based on what we define as the investor’s DTR® calculation for achieving a stated financial goal. The investor’s DTR® calculation is liability driven and considerate of dynamic market events. The Upside Potential/Downside Risk framework used for DTR® combines the innovative ideas of leading financial research and academics that include Nobel Prize winners Harry Markowitz, William F. Sharpe, Daniel Kahneman, and multiple other academics. This Post Modern Portfolio Theory (PMPT), combined with today’s most advanced statistical models and computer technology, is the keystone for Desired Target Return's® revolutionary active manager mix selection and active/passive blending methodology. Dr. Frank Sortino tested and refined the Upside Potential/Downside Risk framework over two decades at the Pension Research Institute. It has been applied at major institutions globally. Furthermore, the quantitative methods employed in Desired Target Return® portfolio construction are based on advanced financial theory developed by some of the leading experts in Investment Theory identified in Figure 9.

GREAT IDEA

GREAT THINKER

Returns-Based Style Analysis

William F. Sharpe* (Stanford University)

Bootstrap Method

Bradley Efron (Stanford University)

3-Parameter Lognormal Distribution

Aitchison/Brown (Cambridge University)

Downside Risk

Peter Fishburn (University of Pennsylvania)

Upside Potential

Daniel Kahneman* (Princeton University)

Source: Sortino Investment Advisors, 2007

PAGE 9 • DTR®

Figure 9: Advancements in Financial Theory

*Nobel Prize winners

D E S I R E D TA R G ET R ET U R N

®

The DTR® Calculation Identifying the DTR® Calculation The foundation to Desired Target Return® portfolio construction is the DTR®. Investors have a rate of return that will achieve their cash flow withdrawals (liability schedule) relative to their asset inventory and cash contribution (funding schedule) as illustrated in Figure 8. The DTR® is the return necessary to achieve the investor’s goal. DTR® identifies the investor’s risk/return profile allowing us to link investment goals to portfolio solutions. DTR® is the primary benchmark we use to measure performance, to analyze managers’ characteristics and dynamically monitor and evaluate the portfolio’s risk/return attributes. Each portfolio is customized to each investor’s unique DTR® profile.

The DTR® risk assessment methodology uses uniquely different risk measurement statistics that are more informative and stringent. The risk measurement statistics are referred to as Upside Potential/Downside Risk and DTR®-Alpha. They are more informative than standard deviation and risk adjusted return ratios because they provide magnitude,

Results

SIA Risk/Return Assessment and Measurements

frequency and whether past returns met or Time exceeded the Desired Target Return®. These important Post Modern Portfolio Theory Figure 10: Liability Driven Investing (LDI) approach to determine and monitor (PMPT) advancements are summarized in Figure 10. They are more stringent because they rely on monthly data and managers are penalized more severely for their downside performance than they are given credit for exceeding the Desired Target Return®. These risk and reward measurements are defined as:

PAGE 10 • DTR®

Upside Potential Ratio The average return above the DTR® measures how often and how far above the Desired Target Return® a portfolio’s returns are likely to occur. Upside Potential, a term coined by Nobel Prize winner Daniel Kahneman, captures investors’ perception of risks concerning gains as opposed to risk concerning losses. Downside Risk Deviation A measure of portfolio risk developed by Peter Fishburn at the University of Pennsylvania, Downside Deviation defines risk as not achieving an investor’s desired target return. By measuring only deviations below the investor’s target return, Downside Deviation distinguishes between “good” and “bad” returns—good returns are greater

The DTR® Calculation

than the target return, and bad returns are below the target return. Only the latter represents risk. In addition, the more returns fall below the target return, the greater the risk. DTR®-Alpha A risk-adjusted return measuring the value added by a manager over and above performance that could have been achieved with its “style benchmark”. A style benchmark, versus market index, is a blended benchmark that is representative of the actual style assortments in the active managers’ portfolio(s). This is accomplished using Sharpe’s style based analysis. The style-blended benchmarks serve as a more precise measurement for identifying manager skill versus random luck. Sortino’s DTR®-Alpha is the difference between the manager’s style blended return and the manager’s customized style blended benchmarks return. Figure 11 below compares Post Modern Portfolio Theory (PMPT) to Modern Portfolio Theory (MPT), which has been the industry standard since 1952. One suspected reason that Nobel Prize winner Harry Markowitz selected a mean variance framework for his Modern Portfolio Theory (MPT) work was due to the practicality at the time of processing any more complex calculations (i.e. semi-variance) on computers with much more limited capacity than today’s computers. Obviously Post Modern Portfolio Theory (PMPT) and the Upside Potential/Downside Risk framework benefit from more efficient computer power and portfolio theory advancements since 1952.

Modern Portfolio Theory (Industry Standard)

Post-Modern Portfolio Theory (Upside Potential/Downside Risk Standard)

Reward

Mean or Expected Return

Upside Potential (exceeding DTR®)

Risk

Standard Deviation, Beta

Downside Risk (falling below DTR®)

Excess

Alpha (Excess over Market)

DTR®-Alpha (Excess over Style Blend)

Forecast

Historical (What did happen)

Bootstrap (What could have happened)

Profiles

Expected Utility Theory

Behavioral Finance

NOTE: DTR® = Desired Target Return

PAGE 11 • DTR®

Figure 11: Post Modern Portfolio Theory (PMPT) Advancements

D E S I R E D TA R G ET R ET U R N

®

Surz Style Pure® Indices As mentioned above, the UP/DR framework concludes through Sharpe’s style based analysis that few active managers’ portfolios are 100% pure to their manager style categories. This is further flawed if using market indices versus a style blended benchmark comparisons for identifying manager skill. Consequently, in order to determine if the manager adds value (DTR®Alpha return) to a diversified portfolio, a manager’s style-blended return needs to be determined and then measured against a similar style blended benchmark. However, this analysis requires the benchmarks used in the analysis to be style pure. If the benchmarks are not style pure then all the analytics based on those benchmarks are misleading.

Surz-Style Indexes Large

Middle

Small

Value

Centric Growth

ive s u l Exc y l l tua u M

ive t s u xha E &

Source: PPCA Inc., 2005

Figure 12: Matrix of Surz Style Pure® (SSP®) Indexes

The popular brand indexes use Price/Book (P/B) to make the determination between value and growth. Low P/B is value and high P/B is growth. Furthermore, not all indexes are constructed using Price/Book. Some use Price Earnings (P/E) combined with other factors such as dividend yield. Here dividend yield is a value measure and P/E is a growth measure. The idea of using both a value and a growth measure is that one confirms the other. A low P/E and high yield indicates value, just as a high P/E and low yield signifies growth. Stocks with off-setting characteristics fall in the middle which Surz Style Pure® (SSP®) indices call “Centric.” The SSP® indices use three factors – Price/Earnings, Dividend Yield and normalized (by sector) Price/Book to create the nine indices matrix illustrated in Figure 12. Centric is defined as the equities in between value and growth. The other major brand indices either assign these Centric stocks proportionately divided between value and growth or eliminate them all together! It matters a lot which factors are used to define equity style classifications.

PAGE 12 • DTR®

For these reasons UP/DR uses the Surz’s Style Pure© indices, as illustrated in Figure 12. Currently this matrix of nine SSP® indices is the purest style benchmarks available today for measuring the style-blended returns of money managers.

Surz Style Pure© Indices

UP/DR Risk/Return Assessment and Measurements The strengths of the Surz Style Pure® indices (SSP®) for portfolio construction purposes are as follows: 1. The SSP® indices used as manager skill benchmarks offer enhanced manager attribution evaluation versus market indexes. 2. SSP® indices utilization ensures better built portfolios under basic tenets of Modern Portfolio Theory (MPT). SSP® indices have greater adherence, than other industry popular indices, to MPT’s important fundamental tenets* of: a. Mutually exclusive—no asset class should overlap with another b. Exhaustive—all securities should fit in the set of asset classes c. Investable—it should be possible to replicate the return of each asset class at relatively low cost d. Macro-consistent—the performance of the entire set should be replicable with some combination of asset classes *Dr. William F. Sharpe, “Determining a Fund’s Effective Asset Mix,“ Investment Management Review, December, 1988, pages 59-69.

PAGE 13 • DTR®

Using the combination of all the above noted academics, Sortino’s Upside Potential/Downside Risk (UP/DR) framework is a more thorough approach to analyzing a manager’s past performance and determining whether a manager is skillful based not on an ill-defined market benchmark but on pure style-blended benchmarks and considering the investor’s Desired Target Return®.

D E S I R E D TA R G ET R ET U R N

®

Bootstrapping Technique Bootstrapping Manager Statistical Data and Interpretation Attempting to forecast money managers’ future returns entails a high degree of uncertainty. Therefore, more data is better than less when seeking the potential for more reliability and meaningful interpretation. Unfortunately managers’ performances are typically categorized into regimented time periods such as 3, 5 and 10year quarterly results. This leaves investment advisors/consultants and investors with a limited number of data points to examine. In addition, the returns are linear so that both good and poor performances fall off (end point bias) after a period of time. This approach increases the risk of not capturing poor or good performance data and could potentially lead to inferior active manager selection. Consequently, Sortino’s UP/DR framework has incorporated a more detailed and informative way to look at past performance. A technique referred to as “bootstrapping” proposed by Effron and Tibshirani (Stanford 1993) allows an analyst to look at what happened but also what best and worse cases could have happened. For example, Figure 13 demonstrates the effectiveness of bootstrapping. Four Years of Fictional Monthly Returns YEAR 1:

YEAR 3:

2 -13 6 11 -5 12 4 7 -6 8 3 1

5 15 0 2 7 14 5 18 6 9 17 -4

YEAR 2:

YEAR 4:

4 7 -6 22 -1 13 0 2 -7 9 4 5

8 1 6 9 7 4 11 21 3 1 13 -12

4 -6 -13 -4 0 -1 -12 1 1 2 -7 -6 Historical Worst Year = +30% Bootstrap Worst Year = -35%

what did happen what could have happened

Repeat process 2,000 times This chart is for illustrative purposes only and does not reflect any actual investment product.

Figure 13: Bootstrapping Trials to Simulate Performance

PAGE 14 • DTR®

In Figure 13 the graphic shows 4 years of monthly returns for a money manager. Typically a consultant will only have 5 years of quarterly returns, possibly more if rolling periods are used and they will be linear. The worst performing year in our example for this manager was a positive + 30%. However, what if we assumed that next year’s return is made by compounding a random sample of 12 of the monthly returns, as shown on Figure 13.

Bootstrapping Technique PAGE 15 • DTR®

Our first random draw might be 4% selected from the sixth month of the fourth year. The first draw is replaced and a second return is randomly selected, in this case negative 6%. Notice this same return is again selected for the last month of one year that could have happened. We repeat this process with 2000 random trials and we now have more data points than using the standard approach. As the graph demonstrated, a negative 35% year is possible. This results in having more downside deviations and upside potential data, and a measurement of the magnitude of such data as well, for attempting to determine the Upside Potential Ratio and DTR®-Alpha Return of the money managers.

D E S I R E D TA R G ET R ET U R N

®

Desired Target Return® Screening Procedures Desired Target Return® portfolio construction begins with the following screening procedures addressing the following requirements for more effective portfolio construction: 1. Identify the DTR®. It is the return that links the client’s assets to their liabilities and financial goals. This is an ongoing review process as these dynamic liabilities, goals and market conditions change. 2. Determine the Desired Target Return® asset and style mix strategy needed to achieve the DTR®. 3. Desired Target Return® optimizes the combination of both active and passive managers to meet or exceed the investors’ DTR®. 4. Desired Target Return® monitors and re-balances the portfolio for dynamic market shifts or changes to the client’s DTR® or financial situation. 5. This Desired Target Return® overlay serves as the mortar to fortify the Global Investment Committee’s strategic and tactical asset allocation, active manager research and portfolio construction process.

SIA Screening Process Non-Proprietary Manager Universes

Manager vs. Style Benchmark R2 > .70?

Funds Drop Out

Funds Stay In

Top Half of Universe Ranked by U-P Ratio?

Funds Drop Out

Funds Stay In

Fund’s DTR®-Alpha > 0%?

Funds Drop Out

Funds are Candidates for Optimizer

Asset and Style Mix Policy

PAGE 16 • DTR®

Client’s DTR® solution set

Figure 14: SIA screening process to best meet investor’s Desired Target Return®

Desired Target Return Screening Procedures

The Predictive Power of DTR®-Alpha

STEP 1. SIA Screening Process Quantitatively analyze over 30,000 plus separately managed accounts, mutual funds and ETFs for their style purity. An R squared value of .70 or higher qualifies for the first screen as illustrated in the top box of Figure 14.

Funds Ranked by Omega Excess Return 1981-1996

70%

2

1 1. Almost 70% of funds in the top quartile in Period 1 stayed in the top quartile in Period 2

STEP 2. Universe Ranking by U-P Ratio Determine the Upside Potential Ratio of the manager with the bottom 50% being eliminated for consideration as illustrated in the second box of Figure 14 the U-P Ratio has historically demonstrated a difference and probably may make a difference in future analysis.

2. Almost 70% of funds in the bottom quartile stayed in the bottom quartile

60% 50% 40% 30% 20%

3. Few top-ranked funds fell to the bottom quartile

4

10% 3

2 Period 1 Rank

0% 1

3

2

3 Period 2 Rank

1

4

Figure 15: DTR®-Alpha historical reliability

STEP 3. DTR®-Alpha Determine whether or not the style blended return of the manager exceeds a similar passive style blended benchmark. If the manager adds value, DTR®-Alpha, they are considered for the portfolio. If not the passive investment option is employed. This step is represented in the third box of Figure 14 and further illustrated in Figure 15. STEP 4. Asset & Style Mix Policy The DTR® links the investor’s profile to the best portfolio combination of active and passive managers. The portfolio is designed to identify the DTR® asset allocation recommendation in adherence to the client’s investment policy guidelines, as illustrated in the fourth box of Figure 14 and in more detail in Figure 16.

The Investor

DTR® = 8%

Goal: Fund liabilities Annual Contribution: $500,000 Annual Distribution: $800,000 Assets:

$35,000,000

The Portfolio Active Fund A Passive A Active Fund B Passive B Fund C 1-3 Year Bonds

$8,575,000 2,940,000 7,105,000 6,825,000 3,255,000 6,300,000

Linking it all together

PAGE 17 • DTR®

Figure 16: Linking the investor’s DTR® Desired Target Return® to a Portfolio Construct

These advanced Upside Potential/Downside Risk framework components are necessary in order to create better portfolio performance results.

D E S I R E D TA R G ET R ET U R N

®

Portfolio Construction Integration The Upside Potential/Downside Risk framework is overlaid in a utilitarian application across non-proprietary manager universes. There are three basic integrating steps to Desired Target Return® portfolio construction, which are as follows:

Global Investment Committee’s Asset Allocation Recommendation

Japan

U.K.

Pacific

International Equity

Europe

Long Bonds

1-3 Year Bonds

3-5 Year Bonds

Tbills

Small Core

U.S. Fixed Income

Small Growth

Small Value

Mid Core

Mid Growth

Mid Value

Large Core

Large Growth

Large Value

U.S. Equity

Figure 17: Step 1—Strategic and Tactical Allocation

STEP 1: Global Investment Committee Asset Allocation

PAGE 18 • DTR®

Global Investment Committee’s Strategic and Tactical Asset Allocation recommendation is the first step of the portfolio construction procedure as illustrated in Figure 17.

Portfolio Construction Integration

STEP 2: Consultant’s Manager Research Manager research and due diligence of traditional Separately Managed Account (SMAs), Exchange Traded Funds (ETFs) or other investment alternatives represents the next step of the portfolio construction procedure, as illustrated in Figure 18. All fund managers in this open architecture consist of non-proprietary Separately Managed Accounts (SMAs) or other investment vehicles.

Investment Advisors over 24,000

Consulting Group Manager Databases representing over 4,700 products 1,430

Search Process Step 1: Initial Qualifications approx.100

Step 2: Quantitative Screens approx.25

Step 3: Qualitative Screens approx.12

Candidate List approx.3-6

Selected Investment Manager(s)

Figure 18: Step 2—Manager Research and Due Diligence

SIA Screens Manager Universe

PAGE 19 • DTR®

STEP 3: Sortino Upside Potential/Downside Risk Framework Overlay Upside Potential/Downside Risk framework serves as an overlay and finishing step for the portfolio construction integration procedure. It serves as the “mortar” for securing the integrity of the Global Investment Committee’s asset allocation and consultant’s manager research/ due diligence “building blocks,” as illustrated in Figure 19 and Figure 20.

Manager Universe

Filter for UP Ratio

Top half of style stay in

Manager Universe Source: Sortino Investment Advisors, 2005

Figure 19: Step 3—SIA Overlay

Portfolio Construction Integration

STEP 4: Identifying Manager Skill

Style Bucket

Small Cap Value

Large Cap Value

The Upside Potential/Downside Risk framework overlay first identifies active managers

Large Value

0.0%

36.4%

Large Centric

0.0%

23.0%

who consistently demonstrate an ability to outperform their style benchmarks on a risk-

Large Growth

0.0%

26.1%

33.3%

7.8%

Mid Centric

0.0%

2.8%

Mid Growth

0.0%

0.0%

Small Value

61.5%

0.0%

Small Centric

0.0%

0.6%

Small Growth

0.0%

3.3%

Europe

0.0%

0.0%

U.K.

0.0%

0.0%

Pacific

0.0%

0.0%

Japan

0.0%

0.0%

Long Bonds

0.0%

0.0%

3-5 Year Bonds

0.0%

0.0%

1-3 Year Bonds

0.0%

0.0%

T-bills

5.2%

0.0%

100.0%

100.0%

adjusted basis. Next, the active managers are optimally mixed to obtain the best combination of active managers for a portfolio. Finally, the overlay seeks to ensure the recommended asset allocation is preserved when these active managers are implemented in a portfolio. This is accomplished by identifying the actual style blends using Surz Style Pure© (SSP®) indices from each active manager’s portfolio. Passive management is then blended where suitable to best align the portfolio to its intended asset allocation recommendation.

Mid Value

R-Squared 90% 90% This overlay enhances the probability of ® meeting or exceeding the investor’s DTR . Style Analysis William F. Sharpe Source: Sortino Investment Advisors, 2005 Each Desired Target Return® custom constructed portfolio seeks to achieve a distinctive Figure 20: Step 3—Active Manager DTR® and utilizes a wide diversification of Style Determination fixed income investments, money market vehicles, indices exchange traded funds, active mutual funds and separately managed accounts. This approach allows us to focus on the individual or institution’s DTR®, and to take a liability driven investment approach to providing investment portfolio strategies. Desired Target Return® seeks to measure success by achieving each client’s DTR®, as opposed to unrelated past performance of market returns, as illustrated in Figure 21.

Active Mix and Passive Blend Process Defensive

PAGE 20 • DTR®

4% Req. Return

High Growth 6% Req. Return

8% Req. Return

9% Req. Return

10% Req. Return

11% Req. Return

12% Req. Return

SIA solves for each desired target return’s optimal combination of: 1. Active Manager mix; 2. Passive blend Source: Sortino Investment Advisors, 2006

Figure 21: DTR® Active Manager Mixing and Passive Blending Process

D E S I R E D TA R G ET R ET U R N

®

Summary The Desired Target Return® portfolio construction process results in an optimal mix of managers and an optimal blend of active and passive investment strategies. The optimal active manager mix and passive blend strategies are designed to ensure that the recommended asset allocation results in the actual implemented asset allocation strategy.

The integration of Modern Portfolio Theory (MPT) with Post Modern Portfolio Theory (PMPT) and today’s technology results in a revolutionary advancement in the evolution of portfolio construction. Desired Target Return®

PAGE 21 • DTR®

portfolio construction is designed to build more effective custom portfolios for better portfolio performance with less portfolio risk for investors.

D E S I R E D TA R G ET R ET U R N

®

Glossary Bootstrap Method A statistical method developed by Bradley Effron of Stanford University that is used by Sortino Investment Advisors (SIA) to develop better estimates of risk and return. With the Bootstrap Method, a manager’s performance is simulated by developing a distribution of possible returns through random sampling with replacement. The process is as follows: Twelve returns are selected at random from historical monthly returns of a manager’s style benchmark and combined to make a single annual return. Each monthly return can be drawn more than once. This is repeated many times to construct a full distribution of possible annual returns. Bootstrapping avoids the dual problems of time sensitivity (beginning/ending date) and limited data because it is not dependent on an arbitrarily selected single period of history. By randomly sampling from a manager’s actual historical monthly returns to simulate future performance, this method measures what could have happened rather than what did happen.

Desired Target Return® (DTR®) The Desired Target Return® an investor must earn in order to accomplish his/her financial goal. While returns above the Desired Target Return® are welcome, returns below this level represent risk to the investor. For example, a 50-year-old investor with $100,000 in retirement assets who is looking to retire at age 65 with $600,000 in assets and is willing to contribute $10,000 a year will need to earn at least an 8% annual return. Thus the Desired Target Return® is 8%; if the actual return falls below 8%, the investor will have to contribute more and/or retire with fewer assets. The following performance measures are calculated based on the DTR®: Downside Deviation, DTR®-Alpha Return, Omega Return, Sortino Ratio, Upside Potential, and U-P Ratio.

PAGE 22 • DTR®

Downside Deviation A measure of portfolio risk developed by Peter Fishburn at the University of Pennsylvania, Downside Deviation defines risk as not achieving an investor’s target return (“TR”). By measuring only deviations below the investor’s TR, Downside Deviation distinguishes between “good” and “bad” returns—good returns are greater than the TR, and bad returns are below the TR. Only the latter represents risk. In addition, the farther returns fall below the TR, the greater the risk. Because it is stated in relation to a TR, a portfolio’s risk, as measured by Downside Deviation, may be perceived differently by investors with a different Desired Target Return®.

Target Return (TR) Risk

Reward

Mean

Lower Returns

Higher Returns

Source: Sortino Investment Advisors

Figure 22

Glossary

Downside Probability The probability, or likelihood, that a portfolio’s return will fall below the DTR®.

Downside Risk See Downside Deviation.

Holdings-Based Style Analysis See Style Analysis.

DTR®-Alpha Return A risk-adjusted return measuring the value added by a manager over and above performance that could have been achieved with its style benchmark. DTR®-Alpha Return is the difference between the manager’s Omega Return and the style benchmark’s Omega Return. Mathematically, DTR Alpha® Return = OmegaMgr – OmegaBenchmark Where, OmegaMgr = Manager’s Omega Return OmegaBenchmark= Style Benchmark’s Omega Return

Omega Return A measure of risk-adjusted performance that indicates whether a manager was compensated for the level of risk taken. Mathematically, Omega Return = RMgr – 3 (Style Beta * DVarBenchmark) Where, RMgr = Manager’s Return DVarBenchmark = Downside Variance of Style Benchmark

PAGE 23 • DTR®

R-Squared A statistical measure that represents the percentage of a manager’s returns that are explained by returns in the manager’s style benchmark. R-Squared ranges from 0 to 100 where 100 means that all movements of a portfolio are completely explained by movements in the style benchmark.

Glossary

Return Distribution

Risk

Reward

Because a portfolio’s return cannot be predicted with certainty, a probability distribution is used to represent many possible outcomes. It incorporates the range of possible returns and the probability of each return occurring. A discrete return distribution has a discrete number of values, each of the discrete values has a certain probability of occurrence that is between zero and one, and the sum of these probabilities must be one (i.e., at least one of the values has to occur).

Average Downside Potential

Upside Potential

Required Return Great Thinkers: Sharpe, Efron, Aitchison, Fishburn, Kahneman

Figure 23

A continuous return distribution has theoretically an infinite number of points over a continuous interval. Probabilities are measured over intervals, not single points as is done for a discrete distribution. That is, the area under the curve between two distinct points (e.g., between two returns) defines the probability for that interval. Sortino Investment Advisors uses the Three-Parameter Lognormal Distribution developed by Aitchison and Brown to fit a continuous curve over a discrete return distribution to improve the accuracy of risk and return estimates. See Three-Parameter Lognormal Distribution.

Returns-Based Style Analysis See Style Analysis.

Sortino Ratio A measure of risk-adjusted performance, indicating how many units of return in excess of the investor’s Target Return are provided per unit of Downside Risk (where Downside Risk is Downside Deviation). Mathematically, Sortino Ratio = (R - TR)/DD Where, R = Expected Return TR = Target Return DD = Downside Deviation

PAGE 24 • DTR®

Standard Deviation A statistical measure of risk reflecting the extent to which rates of return for a portfolio may vary from period to period and gauges the dispersion of monthly returns around the average return. The larger the standard deviation, the greater the range of possible returns and, therefore, the more risky the portfolio. If the distribution is normal, the Standard Deviation gives a good estimate of its dispersion around the average. If the distribution is non-symmetric, the standard deviation measure can give misleading information.

Glossary

Style Analysis Style Analysis is used to determine the investment style (e.g., growth versus value, large cap versus small cap) of a portfolio. Returns-Based Style Analysis (RBSA), developed by Nobel Prize winner William Sharpe, uses the manager’s historical monthly returns to find a blend of passive indices that replicates the manager’s performance. Holdings-Based Style Analysis (HBSA) classifies a manager’s style orientation based on the characteristics of the portfolio’s underlying securities. Sortino Investment Advisors (SIA) uses RBSA to determine a manager’s style benchmark and to measure riskadjusted performance and value added. HBSA is used to validate this assessment, and where the two approaches differ considerably, provides the basis for further investigation.

Style Benchmark A set of passive indices developed with Returns-Based Style Analysis (RBSA) that replicates the style and performance of an active manager. SIA uses the following style indices: U.S. Equities: Large Value, Large Centric, Large Growth, Mid Value, Mid Centric, Mid Growth, Small Value, Small Centric, Small Growth International Equities: Europe ex U.K., Pacific ex Japan, Japan Fixed Income: 7-10 Year Bonds, 3-5 Year Bonds, 1-3 Year Bonds, and Treasury Bills

Style Beta The ratio of the manager’s Downside Risk to the Style-Blended Benchmark’s Downside Risk for a given TR. Style Beta indicates whether the manager takes more or less risk than is inherent in their style where risk is measured by Downside Deviation. The higher the ratio, the greater the manager’s risk per unit of style risk. Values greater than 1.0 indicate proportionately more risk than the Style Benchmark, and values less than 1.0 indicate proportionately less risk. Mathematically, Style Beta = DDMgr/DDBenchmark Where, DD = Downside Deviation Mgr = Manager Benchmark = Style Benchmark

PAGE 25 • DTR®

Three-Parameter Lognormal Distribution A method, developed by Aitchison and Brown of Cambridge University, for fitting a continuous curve over a discrete return distribution. The three parameters are the mean, the standard deviation, and the extreme value of the annual returns. It allows for both normal and skewed return distributions. Sortino Investment Advisors (SIA) uses the Three-Parameter Lognormal Distribution to fit a curve to the bootstrapped distribution of possible annual returns so that a variety of statistical characteristics — including average, standard deviation, downside risk, upside potential, etc. — can be estimated.

Glossary

Surz Style Pure® (SSP®) Indices Style groupings are based on data provided by Compustat. Two security databases are used. The U.S. database covers more than 6000 firms, with total capitalization exceeding $18 trillion. The non-U.S. database coverage exceeds 15,000 firms, 20 countries, and $31 trillion – substantially broader than EAFE. To construct style groupings, Surz first break the Compustat database for the region into size groups based on market capitalization, calculated by multiplying shares outstanding by price per share. There are 3 regions maintained in our system: U.S., Foreign and Global. Beginning with the largest capitalization company, Surz adds companies until 65% of the entire capitalization of the region is covered. This group of stocks is then categorized as “large cap” (capitalization). There are generally about 200 companies in this group for the U.S., 800 for Foreign, and 1000 for Global. The second size group represents the next 25% of market capitalization and is called “mid cap”. There are generally about 1000 companies in this group for U.S., 2700 for Foreign, and 3500 for Global. Finally, the bottom 10% is called “small cap”. There are generally 5000 U.S. securities in this group, 10,000 Foreign, and 15,000 Global. Then within each size group, a further breakout is made on the basis of orientation. Value, centric, and growth stock groupings within each size category are defined by establishing an aggressive measure. Aggressiveness is a proprietary measure that combines dividend yield and price/earnings ratio. The top 40% (by count) of stocks in aggressiveness are designated as “growth,” while the bottom 40% are called, “value,” with the 20% in the middle of falling into “centric.”

Upside Potential The average return above the Desired Target Return®, measuring how often and how far above the Desired Target Return® a portfolio’s returns are likely to occur. Upside Potential, a term coined by Nobel Prize winner Daniel Kahneman, captures investors’ perception of risks concerning gains as opposed to risk concerning losses.

U-P Ratio (Upside Potential Ratio) The ratio of Upside Potential to Downside Deviation at a given DTR® Desired Target Return®, measuring how much upside potential is provided by a manager at a given level of downside risk. Mathematically, U-P Ratio = UP/DD

PAGE 26 • DTR®

Where, UP = Upside Potential DD = downside deviation

D E S I R E D TA R G ET R ET U R N

®

Disclaimer The material presented in this brochure is for illustration purposes only and is an attempt to communicate a complex subject in a simplified fashion. For more information regarding Sortino’s Upside Potential/Downside Risk optimization methodology you can visit www.sortino.com or www.sortinoim.com. Sortino’s methodology strives to take a cautious approach to analyze past performance in order to predict future results in a world of uncertainty. Many calculations are performed as it related to protecting the investor on the downside. From Style Beta, Downside Deviation, Risk Aversion and DTR®-Alpha, in essence each statistic measurement is calculated by design to penalize each active manager solution more severely for performing below than giving them credit for meeting or exceeding the Target Rate of Return. Passive management is integrated where this scrutinized satisfactory active manager is absent and to budget active manager risk. The culmination of over 40 years of work by Dr. Sortino and the work of highly acclaimed academics and Noble Prize winners has brought us to a new paradigm of investing — one that focuses on the return needed to accomplish an investor’s goal while at the same time providing the downside protection every investor seeks. The Desired Target Return® Program does not take into consideration the investor’s tax situation, debt or future changes in assets or income needs. The investor should evaluate their level of risk tolerance based on their own investment knowledge, experience, demographics and net worth and consider either adjusting their portfolio risk, investigating alternative investment options, or consulting with an investment advisor to consider their specific situation and needs. The program assumes that the DTR® is the underlying goal of the investor, and the possibility of the investor failing to reach the Desired Target Return® accordingly is a primary investor risk.

PAGE 27 • DTR®

Investing in financial securities involves risk. The higher the Desired Target Return® the greater the potential for significant loss of principle in your retirement account balance. Ignoring or under estimating the Desired Target Return® increases the risk of not replacing a sufficient amount of pre-retirement salary. Sortino Investment Advisors, LLC (SIA) makes no assurances, implied or implicit, that the Desired Target Return® assigned to each participant will be achieved nor if achieved will result in replacing a significant amount of pre-retirement salary. Past performance is no guarantee of future results. SIA, its affiliates, and its employees are not in the business of providing tax or legal advice. These materials and any tax-related statements are not intended or written to be used, and cannot be used or relied upon, by any such taxpayer for the purpose of avoiding tax penalties. Tax-related statements, if any, may have been written in connection with the “promotion or marketing” of the transaction(s) or matters(s) addressed by these materials, to the extent allowed by applicable law. Any such taxpayer should seek advice based on the taxpayer’s particular circumstances from an independent tax advisor.