03/14/2005
Pension Trustees Agenda
Date: 03/14/2005
Location: Council Chambers - City Hall
Call to Order
Approval of Minutes
02-14-2005
Pension Trustee Items
1. Accept the employees listed below into membership in the City of Clearwater's
Employees' Pension Plan.
2. Mark Petrie, Public Services Department, and Linda Burch, Police Department, be
allowed to vest their pensions under Section(s) 2.397 and 2.398 of the Employees'
Pension Plan as approved by the Pension Advisory Committee.
3. Janice King, Development & Neighborhood Services Department; Lois Klein, Library
Department; Sam Brown, Jr., Solid Waste Department; and Terry O'Neill, Fire
Department, be granted regular pensions under Section(s) 2.393 and 2.397 of the
Employees' Pension Plan as approved by the Pension Advisory Committee.
4. Accept the recommendations of the Asset Allocation Study performed by Kalson &
Associates and authorize staff to take the necessary actions to implement the
recommendations.
5. Approve a contract with Callan Associates Inc. for investment consulting services with
an annual fee of $72,000 and authorize appropriate officials to execute same.
Other Business
Adjourn
Pension Trustees Agenda 3/14/2005
Page 1 of 1
f
:~.
Pension Trustee Cover Memorandum
Trackinq Number: 1,166
Actual Date: 03/14/2005
Subiect / Recommendation:
Accept the employees listed below into membership in the City of Clearwater's Employees'
Pension Plan.
Summary:
Pension Elig.
Name, Job. Class, & Dept.jDiv.
Hire Date Date
Ekatrini Gerakious, Americorps Clearwater Coordinator/Police 1/3/05 1/3/05
Spiro Mannossos, Gas Technician I/Gas 12/13/04 12/13/04
Andrea Wells, Senior Staff Assistant/Legal 9/30/02 12/25/04 *
Kathleen Coakley, Customer Service Rep.jCustomer Service 1/10/05 1/10/05
Erin Keefe, Management Intern/City Manager 1/10/05 1/10/05
Matthew Anderson, Police Officer/Police 1/10/05 1/10/05
David Hassett, Police Officer/Police 1/10/05 1/10/05
Joseph Lespier, Police Officer/Police 1/10/05 1/10/05
Daniel Cote, Custodial Worker/Parks & Recreation 1/10/05 1/10/05
Suzanne Jackson, Systems Programmer/Information Technology 1/10/05 1/10/05
Mark Pellechio, Public Utilities Technician II/Public Utilities 1/10/05 1/10/05
Richard Hannigan, Gas Technician I/Gas 1/10/05 1/10/05
Christopher Lang, Senior Marine Lifeguard/Marine & Aviation 1/9/05 1/9/05
Vanessa Rogers, Police Cadet/Police 1/10/05 1/10/05
Jeffrey Naiman, Police Officer/Police 1/10/05 1/10/05
Alexander West, Public Utilities Technician I/Public Utilities 1/10/05 1/10/05
Terry Rongey, WWTP Operator A/Public Utilities 1/24/05 1/24/05
Jeremy Peplow, TV Production Specialist/Public Communications 1/24/05 1/24/05
Kenneth Mobley, WWTP Operator Trainee/Public Utilities 1/24/05 1/24/05
Oriqinatinq: Human Resources
Review Approval
Cvndie Goudeau
03-03-2005
13:57:08
f
:~.
Pension Trustee Cover Memorandum
Trackinq Number: 1,167
Actual Date: 03/14/2005
Subiect / Recommendation:
Mark Petrie, Public Services Department, and Linda Burch, Police Department, be allowed to vest
their pensions under Section(s) 2.397 and 2.398 of the Employees' Pension Plan as approved by
the Pension Advisory Committee.
Summary:
Mark Petrie, Materials Tester, Public Services Department, was employed by the City on
November 24, 1986, and began participating in the Pension Plan on that date. Mr. Petrie
terminated from City employment on January 14, 2005.
Linda Burch, Police Service Technician, Police Department, was employed by the City on January
23, 1995, and began participating in the Pension Plan on that date. Ms. Burch terminated from
City employment on January 24, 2005.
The Employees' Pension Plan provides that should an employee cease to be an employee of the
City of Clearwater after completing ten or more years of creditable service (pension
participation), such employee shall acquire a vested interest in the retirement benefits. Vested
pension payments commence on the first of the month following the month in which the
employee normally would have been eligible for retirement.
Section 2.393 (p) provides for normal retirement eligibility when a participant has reached age
55 and completed twenty years of credited service, has completed 30 years of credited service,
or has reached age 65 and completed ten years of credited service. Mr. Petrie would have
completed at least 20 years of service and reached age 55 on November 24, 2006. His pension
will be effective December 1, 2006. Ms. Burch would have completed 20 years of service and
reached age 55 on January 24, 2015. Her pension will be effective February 1, 2015. These
pensions were approved by the Pension Advisory Committee on February 10, 2005.
Oriqinatinq: Human Resources
Review Approval
Cvndie Goudeau
03-03-2005
13:56:04
f
:~.
Pension Trustee Cover Memorandum
Trackinq Number: 1,168
Actual Date: 03/14/2005
Subiect / Recommendation:
Janice King, Development & Neighborhood Services Department; Lois Klein, Library Department;
Sam Brown, Jr., Solid Waste Department; and Terry O'Neill, Fire Department, be granted regular
pensions under Section(s) 2.393 and 2.397 of the Employees' Pension Plan as approved by the
Pension Advisory Committee.
Summary:
Janice King, Code Enforcement Inspector, Development and Neighborhood Services Department,
was employed by the City on January 6, 1975, and her pension service credit is effective on that
date. Her pension will be effective March 1, 2005.
Based on an average salary of approximately $41,976 per year over the past five years, the
formula for computing regular pensions, and Ms. King's selection of the Joint & Survivor Annuity,
this pension will approximate $34,633 annually.
Lois Klein, Librarian III, Library Department, was employed by the City on April 6, 1979, and her
pension service credit is effective on that date. Her pension will be effective February 1, 2005.
Based on an average salary of approximately $55,056 per year over the past five years, the
formula for computing regular pensions, and Ms. Klein's selection of the 100% Joint & Survivor
Annuity, this pension will approximate $37,787 annually.
Sam Brown, Jr., Solid Waste Supervisor I, Solid Waste Department, was employed by the City
on August 20, 1980, and his pension service credit is effective on that date. His pension will be
effective March 1, 2005.
Based on an average salary of approximately $45,860 per year over the past five years, the
formula for computing regular pensions, and Mr. Brown's selection of the Joint & Survivor
Annuity, this pension will approximate $30,821 annually.
Oriqinatinq: Human Resources
Review Approval
Cvndie Goudeau
03-07-2005
08:11:09
Regular Pensions
Page 2
March 14, 2005
Terry O'Neill, FirefighterlDriver Operator, Fire Department, was employed by the City on February
17,1975, and his pension service credit is effective on that date. His pension will be effective March 1,
2005.
Based on an average salary of approximately $53,656 per year over the past five years, the formula for
computing regular pensions, and Mr. O'Neill's selection of the 75% Joint & Survivor Annuity, this
pension will approximate $44,713 annually.
These pensions were approved by the Pension Advisory Committee on March 10, 2005. Section 2.393 (p)
provides for normal retirement eligibility when a participant has reached age 55 and completed twenty years of
credited service, has completed thirty years of credited service, or has reached age 65 and completed ten years
of credited service. Section 2.393 also provides for normal retirement eligibility when a participant has completed
twenty years of credited service in a type of employment described as "hazardous duty" and further defines
service as a Firefighter/Driver Operator as meeting the hazardous duty criteria. Ms. King qualifies under the 30
years of service criteria. Ms. Klein and Mr. Brown qualify under the age 55 and twenty years of service criteria.
Mr. O'Neill qualifies under the hazardous duty criteria.
f
:~.
Pension Trustee Cover Memorandum
Trackinq Number: 1,183
Actual Date: 03/14/2005
Subiect / Recommendation:
Accept the recommendations of the Asset Allocation Study performed by Kalson & Associates
and authorize staff to take the necessary actions to implement the recommendations.
Summary:
In October 2004, the Trustees authorized the hiring of Kalson and Associates to conduct an
asset allocation study of the pension plan investments. The goal of this study was to find the
asset mix that would generate the highest possible return with an acceptable amount of risk.
Some of the recommendations made will require going to referendum for voter approval and
others will simply require the reallocation of funds from one investment type to another
including the hiring and termination of investment managers.
The following is a summary of the recommendations of the Asset Allocation Study performed by
Kalson & Associates:
Changes requiring voter approval via a referendum:
1. Add Real Estate Investment Trusts (REITs) as an allowable investment up to 15% of the
portfolio. REITs are companies that primarily own buildings and land.
2. Increase the allowable amount of International Investment from 10% to 20% of the portfolio
to provide for the addition of emerging market equity investments.
3. Increase the maximum equity allocation from 65% to 70%.
4. Allow up to 10% of the portfolio to be invested in High Yield Bonds.
Changes not requiring voter approval via a referendum:
5. Add a Mid Capitalization Value Equity Manager.
6. Add a Small Capitalization Growth Equity Manager.
7. Add a REIT's Manager (if investment type approved via referendum)
8. Add an active International EAFE Manager to replace the current passive EAFE Manager.
Upon acceptance by the Trustees of the Asset Allocation Study recommendations, staff will work
with the Pension Plan's Attorney and City Attorney to draft the appropriate language for the
voter referendum. In addition, staff will start the process to select a consultant to assist with
the various investment manager searches.
Oriqinatinq: Finance
Review Approval
Maraie Simmons 03-03-2005 16:28:50
Cvndie Goudeau 03-10-2005 15:22: 14
Bill Horne 03-07-2005 12: 10:45
Tina Wilson 03-07-2005 08:35:29
Garrv Brumback 03-07-2005 09:58:49
f
:~.
Pension Trustee Cover Memorandum
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
1. EXECUTIVE SUMMARY
The main purpose of this study was to determine the allocation targets for the City of
Clearwater Pension Fund. While the Trustees currently have allocation targets in
place, it has been over 10 years since the last such formal study. Starting with the
current allocation levels, how should those levels change when considering future
capital markets? Whether to project today's market volatility into the future is
another related issue.
We expected the modeling to show the net benefit of higher equity allocation and
diversification within some asset classes that the City of Clearwater currently does not
invest in. However, we were also guided by common sense and a "first things first"
attitude regarding any recommended changes. These are critical retirement vehicles
for the plan participants. Therefore, these allocation decisions are vitally important.
Likewise, adjustments for the risk tolerance of Trustees are just as important. To fail
to make such adjustments would eradicate an otherwise useful recommendation.
The model had three key inputs:
1. Five-year expected returns of relevant asset classes;
2. Expected volatility (defined as standard deviation risk) around the
average expected returns; and
3. Expected relationships (correlation) between pairs of asset classes
as they fluctuate in the marketplace over time. (See correlation chart
at the end of section 1)
Since model outputs are only as good as the inputs, we tapped several reputable
sources for these inputs. Return and risk forecasts were gathered from several
investment managers we respect. Kalson & Associates (K&A) made successive, but
minor adjustments to these forecasts to achieve a more rational set of outcomes.
Regarding cross-correlations, these relationships change slowly. Keeping the slow-
changing nature of correlations in mind, we examined 10-year historical relationships
among the various asset classes. We also gathered data from investment
professionals regarding their opinion of future correlations. As markets become
more global and information-transfer increases, we believe that asset class
correlations will change more substantially over time. Overall, we fused historical
and forecasted correlations with slight K&A biases to arrive at our assumptions.
The model then takes these inputs and processes them to develop an efficient set of
outputs. Given a certain risk level, what is the highest possible combined return that
could be obtained? Alternatively, given a specific return, what is the lowest risk
outcome? Each run of the model tries approximately 1,000 combinations of asset
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
classes. Either way, the model blends those asset classes having the lowest
correlation to each other. One output chart plots the combinations on a graph
labeled "Feasible Region". The Y axis is expected return. The X axis is the risk
level. The plot looks like a scatter diagram with the top points forming an elliptical
line. When you are on that line, called the efficient frontier, you have reached the
optimum return for a given risk or lowest risk for a given return.
Our most important recommendations are:
. invest the assets a bit more aggressively with approximately 70% equity
from the current 60%-65% target
. create accounts in the following "new" asset classes that Clearwater is
currently not invested in:
. REIT (Real Estate Investment Trust) Portfolio
. Small Cap Growth Portfolio
. Emerging Market Equity Portfolio
. Mid-Cap Value Portfolio
The remammg pages explain the process of the study in more detail. They also
specify the sources for the model inputs. In the end, the input decisions are only
judgment calls. However, the forecasts appear reasonable and were not designed to
materially influence the outputs and recommendations. At the end of the report,
there is a discussion of the limitations and overriding assumptions applicable to the
study.
This modeling effort was performed at a moment in time. One year, two years, and
certainly three years from now, capital markets will have changed. We strongly
encourage another informal look along the way.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
2. BACKGROUND
It has been over 10 years since the Trustees of the City of Clearwater Pension Fund
have taken a serious look at the future of capital markets, utilizing an asset allocation
study. At some point, we can expect interest rates to rise, oil prices to stabilize, the
equity markets to find some stability, etc. For all these reasons, as well as fiduciary
responsibility, the Trustees authorized an asset allocation study.
We grappled with capital market expectations and the relaxing of constraints.
Further, we layered in our biases, including:
. higher equity, if risk is contained
. projected outperformance of non-S&P stocks
. value-added and increased liquidity of REITs vs. privately-held real
estate
. rising price movement correlation between U.S. and non-U.S.
equities
. relatively high REIT price movement correlation with small-cap
stocks
. continuing belief that alternative asset classes (e.g. venture capital)
were inappropriate to consider at this time.
Kalson & Associates introduced the following asset classes into the model:
1. Large Cap Growth Domestic Equity
2. Large Cap Value Domestic Equity
3. Mid Cap Growth Domestic Equity
4. Mid Cap Value Domestic Equity
5. Small Cap Growth Domestic Equity
6. Small Cap Value Domestic Equity
7. REITs
8. International Equity
9. 90-day Treasury Bills
10. Aggregate Bonds
11. Emerging Market Equity
12. International Soverign
13. Emerging Market Debt
14. Hedge Funds
15. High Yield Bonds
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
Omitted were venture capital, private equity, commodities, and other alternative
investments. Moreover, in almost all of the runs, the allocation software did not
allocate to 90-Day T-Bills, so we tended to zero-out the allocation to this asset class.
Starting with the current fund allocations, we introduced several alternative scenarios.
The best way to judge whether an alternative scenario "works" is to look for a
combination of 1) higher expected returns; 2) lower expected risk; 3) Trustee comfort
with scenario constraints (e.g. no more than x% REITs); and 4) common sense
allocations.
Run 1 was a most interesting scenario because it was "unconstrained"; we did not
constrain any asset class under consideration. Arguably, the outcome was highly
suspect. The 8.2% expected return is higher than the actual portfolio's current 7.0%;
but the offset is a 12.4% standard deviation vs. 9.5%. Moreover, the 42% allocation
to REITs is far too concentrated, as is the 34% allocation to mid-cap growth stocks.
In reality, a negative performance year for REITs (such as 1998) could easily occur
again. Such an allocation, no matter how enticing from a return perspective, is just
impractical.
We have modeled a succession of allocation runs. The process involved placing mild,
then strong constraints on several asset classes, then determining if the return/risk
scenarios and asset class allocations made sense. In all, we ran approximately 10 runs,
but we would like to focus on Runs 3 and 4, as we believe these runs give us the
best return/risk tradeoff, with allocations that make "common sense".
In Run 3, we implemented constraints on many of the asset classes. We reduced our
maximum allowable REIT exposure to 15% and EAFE (international developed
equity markets) to 10% and also required at least a modicum of bonds (30%
minimum with a 45% maximum). Further, we looked at the current City of
Clearwater allocation and tried to work that layout into the scenario so as to avoid a
complete overhaul of asset classes. For instance, we placed minimums of 10% in
large-cap growth and large-cap value, and placed 5% minimums on small-cap value,
and both mid-cap growth and value.
The results of this run were good, in our opinion. Of the 11 portfolios generated by
the allocation software, we chose to focus on portfolio #9, which provided a 7.3%
expected return, and a 10.2% risk (standard deviation). For reference, the current
Clearwater allocation projects to a 7.0% expected return with a 9.5% risk (standard
deviation). This portfolio suggests adding four assets classes that are currently not in
the City of Clearwater allocation: REITs, Small Cap Domestic Growth stocks, Mid-
Cap Domestic Value Stocks, and Emerging Market Equity.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
In Run 4, we implemented many of the same constraints as Run 3, however we fixed
the allocation to domestic fixed income to 25% (no more, no less), and reduced our
maximum allowable REIT exposure to 10%. All other constraints were the same as
Run 3.
The results of this run were interesting. Of the 11 portfolios generated by the
allocation software, we chose to focus on portfolio #8, which provided a 7.4%
expected return, and a 10.8% risk (standard deviation). Similar to Run 3, this
portfolio suggests adding the four aforementioned assets classes that are currently not
in the City of Clearwater allocation, and also adding a small allocation (2.4%) to
emerging market debt.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
3. SUPPORTING INFORMATION
A five-year time horizon was chosen as the projection period. Because it is difficult
to predict returns far into the future, managers who undertake capital markets
forecasting typically limit the projection to five years.
The expected returns and volatility were a combination of 1) the average of the
forecasts of reputable and well-respected managers and 2) our own collective
thinking. In most cases, we closely agreed with the averages. We were arguably
more optimistic regarding smaller-cap and mid-cap issues. But in both cases, we
believe these classes have the potential to outperform the larger-cap names in the
future.
Cross correlations between pairs of asset classes were based on historical data Oatest
10 years) and forecasted relationships. This data changes slowly over time and
should not impede realistic expectations. Cross correlation statistics range between
-1.0 and +1.0. A correlation of +1.0 suggests the highest degree of co-movement
between two asset classes. For example, it is intuitive that intermediate-term bonds
correlate highly with longer-term bonds. In fact, the historical correlation is +0.98,
or almost +1.0. Correlations close to 0.0 indicate no co-movement between asset
classes. Small-cap stocks vs. T -bills is such an example. A correlation of -1.0
indicates two asset classes moving in opposite return directions over time. We have
forecasted that REITs and T-bills have a negative correlation (-0.06).
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
4. EXPLANATION OF OUTPUT PAGES FROM THE THREE MODEL
RUNS
Eleven-page outputs for each of the three model runs appear in Tabs 4-6. Each
output page is explained below.
The chart titled Asset Class Return and Risk Assumption and Constraints
simply shows the initial weighting in each asset class (actual Clearwater allocation as
of December, 2004); the constraints that were put in place (e.g. a minimum of 30%
and a maximum of 45% in domestic aggregate bonds); the return and risk
assumptions that were utilized; and the actual 10-year historical returns of each asset
class.
The Portfolio Optimization Summary chart illustrates the small but important
increments of each asset class as one moves more aggressively from Portfolio 1 to
Portfolio 11. One way to use the chart is to contemplate the asset allocation in a
portfolio one level higher (Portfolio 7) or lower (Portfolio 5) in risk than Portfolio 6.
Sometimes, there are dramatic changes when the risk level is modified by only one or
two degrees!
The Portfolio Downside Risk chart illustrates the one standard deviation return and
the two standard deviation return for both five-years and one-year. For instance, in
Scenario 3, there is a 50% probability of generating 7.28% over 5 years. This is the
expected return. However one standard deviation shows us that the return can range
from 13.12% to 1.44%. The two standard deviation returns would range from
14.77% to -0.21 %.
The "mountain" chart is entitled Asset Weights. This chart shows the continual ebb
and flow of each asset class as risk increases from Portfolio 1 to Portfolio 11. Some
colored sections of the chart become steady state, reflecting a maximum constraint.
Other colors start large and decay to zero; this phenomenon is depicting the decline
in optimization of the low-risk asset classes as the portfolio assumes more risk.
The Optimal Selection Summary page shows the optimal weights of each Portfolio
(we selected Portfolio #9) asset class in pie-chart format. Expected return is also
listed, together with risk. Note that five-year risk is much lower than single-year risk;
with the passage of time, risk hypothetically declines by the square root of the time
period.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
The Optimal vs. Initial Weights page shows the optimal weights of each Portfolio
(we selected Portfolio #9) asset class in pie-chart format versus the current, actual
exposure of the plan as of December 2004.
The next chart is Annual Return Distribution, 5 Years. This bar chart shows the
five-year annualized expected return in each ascending risk portfolio. In addition, the
68% confidence level (dark blue) is shown. This is equivalent to a one standard
deviation risk in either direction from the mean. The 95% confidence level Oight
blue) depicts an almost best/worst case scenario. This is a two standard deviation
(rather unlikely) outcome. The minimum target return was set at 7.5% for the overall
plan. This represents the actuarially-assumed assets earnings rate, net of fees.
The chart labeled Return Distribution, Single Year shows the one-year expected
return in each ascending risk portfolio. In addition, the 68% confidence level is
shown in dark blue. This is equivalent to a one standard deviation risk in either
direction from the mean. The 95% confidence level Oight blue) depicts an almost
best/worst case scenario. This is a two standard deviation outcome (rather unlikely).
The minimum target return was set at 7.5% for the overall plan. This represents the
actuarially-assumed assets earnings rate, net-of-fees. Notice that compared to the
chart labeled "Annual Return Distribution, 5 Years", the expected returns vary much
more in the one-year period.
The next chart is Annual Return Distribution to 5 Years (Portfolio specific).
This bar chart shows the five-year annualized expected return of a specific portfolio
portfolio. In addition, the 68% confidence level is shown in dark blue. This is
equivalent to a one standard deviation risk in either direction from the mean. The
95% confidence level Oight blue) depicts an almost best/worst case scenario. This is
a two standard deviation outcome (rather unlikely). The minimum target return was
set at 7.5% for the overall plan. This represents the actuarially-assumed assets
earnings rate, net-of-fees.
The Feasible region graph depicts return on the X axis and risk on the Y axis for
1,000 randomly generated plots. Each plot uses a slightly different set of asset
allocations. Eleven of the plots, numbered 1 through 11, represent varying levels of
risk. Those are the best possible positions for a portfolio, given the risk levels and
input constraints. We will always refer to Portfolio 6 of a given run as the middle risk
level to consider. Taking more risk would mean substituting Portfolio 7 or higher.
The reverse applies to Portfolios 5 down to Portfolio 1, the most risk-averse plot.
The letter i plot refers to the current portfolio allocation. You should expect an
improvement in the more efficient portfolios vs. the current (i) portfolio as of
December, 2004. The various F numbered plots refer to a single asset class. To
illustrate, plot F1 specifies T -Bills.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
The chart titled Probability of Achieving Minimum Return shows the likelihood
of achieving at least 6.0% per year. That likelihood naturally increases as the
investment period increases. This is a confidence-building chart, as well. In every
run for the overall pension plans, the performance news is expected to be better and
better by year 5. On the other hand, the expected probability of achieving at least
6.0% is still well below 100% by year 2010.
A similar chart using a 7.5% target return is a much tougher hurdle. Since that target
return is slightly higher than any recommended scenario, it becomes less and less
likely to achieve. Our response is that these are index-oriented forecasts.
Confidently, we believe active managers would provide a combined 7.5% or better 5-
year outcome. Charts in the attached handouts are targeted to 7.5%.
We will also discuss each of these charts in more depth when we meet with the
Trustees. Each Trustee may have a new perspective to bring to the table regarding
the optimal allocation for the plan.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
5. KEY FINDINGS
. Increasing equity from 60%-65% (current) to 70% increases expected
return at an acceptable cost of 1 % higher expected risk
. Adding more and more constraints does not cause return or risk problems
and offers more diversified equity style exposure
. Due to the fact that all forecasted returns are in the single-digits, it is
extremely difficult to generate a projected return of the actuarial assumed
7.5%. To get even close to the 7.5%, we had to increase equity exposure
to approximately 70%.
. Due to REITs relatively low forecasted risk (standard deviation), the
model is clearly partial to this asset class and would recommend a higher
allocation to the asset class if we had not constrained it to 15% in Run 3
and to 10% in Run 4.
. Although Emerging Market Equities have a relatively high forecasted risk
(25% standard deviation), this asset class has the highest forecasted return
on the page (9.5%). Further, due to the low correlation of emerging
market equities with many of the other asset classes, the model, in many
portfolios, recommends an allocation to this asset class.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
6. RECOMMENDATIONS
We offer several straightforward recommendations to the City of Clearwater
Trustees. In our view, there is no rush to make the investment style changes, but we
urge consideration by year-end 2005.
. Shift to 70% equity exposure from the current 60%-65% range, as the
forecasts for fixed income asset classes are not particularly high, and with
rates expected to increase, fixed income may not be the "safe haven" that
it has previously been.
. Introduce an allocation to REITs as a portfolio diversifier.
. Use a portion of the current large-cap value exposure to fund an allocation
to an active small-cap growth manager
. Use a portion of the current mid-cap growth exposure to fund an
allocation to an active mid-cap value manager
. Introduce exposure to emerging market equities. This asset class is
projected to provide solid returns, with higher risk. However, the asset
class also further diversifies the City of Clearwater plan due to its relatively
low correlation with other asset classes
At the end of the day, the views of the City of Clearwater Trustees are paramount.
KALSON & ASSOCIATES
CITY OF CLEARWATER PENSION PLAN
ASSET ALLOCATION STUDY
FEBRUARY, 2005
7. INHERENT LIMITATIONS TO ASSET ALLOCATION MODELING
Listed below are three limitations applicable to the Clearwater Study:
. The five-year estimates of return and risk are just that, estimates.
. There are inherent K&A biases in the return estimates. For example, we believe
that small-cap and mid-cap equities will do better than their large-cap brethren.
More importantly, we are assuming that future returns are much, much lower than
the high double-digit performance earned for most of the 1990s.
. This was an assets-only study, independent of all but one (earnings rate) actuarial
assumption. In theory, a "better" approach would be to forecast assets growth in
tandem with liability growth, i.e. the present value of future benefit payments.
However, the actuarial data represents estimates, as well. Also, the plan is close to
fully-funded status on an accrued benefit basis. As a result, asset allocation
planning by the Trustees is more discretionary.
KALSON & ASSOCIATES
e:
o
',;:;
Cll
(.)
e:.2
.!!!<t
...ll._
Se:51
Cll 0 ell
~ 'iij <t
Clle:"C
Qj Qj Qj
U- ll. ell
ell 0
'0 "gj g-
>->-...
:!:::Oll.
U-"C
c..e:
E Cll
W_
e:
~
...
:::l
U
.....
0)
0)
ell
e:
ell
~
~
:::l
0-
W
o
-
x
0)
-0
e:
..c:
t)
.....
ell
0)
CJ)
.....
0)
0)
ell
e:
ell
E
t5
:::l
-0
e:
o
()
e: -'
ell CJ) e:
ell E ..... 0)
e: ..c: ~ E
o 0) ell 0)
',;:; ...J e: 0)
(.) _ ..... ..... ell ell
<t ~ 0) 0) E ffi
O)~ e:e:~~~ E
E e: 0 0 e: e: ~ 0)
~ w ~ ~ eIlE eIlE ,CJ)x- >
..c: e:e: 1:5
~ ~ 'E"E $ $ 0) ell ell ell
-z "ww~~~e.85
10 c)+""'+"'" ~+::; Q..Q.) c..
-g ~ "B w w .E .E '0 g-"~ g-
-0'::: e: ~~..c:..c: 0)..... CJ)
0) O)..s!l e: e: ~ ~ .;; -00) ~ -00)
0) e: ell ell ell ell ell '-' .....
e: ~ ~ E E ~ ~ 3 > E >
53~~~t5t5~CJ)e~
1:5'+-w.c.c:::l:::lt)~'+-"'"
ell ~ E '~ "~ -g -g -5 5- ~ 5-
o 0 0 0 0 0 0 0) 0) 0 0)
Z~ZIl..Il..()()O:::O:::~O:::
Qj
(.)
~~~~
-('")<<,0')
~ 0 '0
c
~
~oo
::000
co
"C
Qj
e: e:
e: "C 0
oQj',;:;
ell E Cll ~
iiiE(.)o
~0.2
:rl<t
e::::
o 0 ~
d'd'~~oo~
00::0::000::0
co
~ ~
I'-~::o
"<I:&'~
o I'-
~ ~
('o...~u,
"<I:oco
o I'-
~
o
o
~
o
o
o
e:
- 0 ~ ~ ~
e: ',;:; ~ ~ ~ ~ ~
~ 0 0 0 0
Cll 0 0 L() ~ ('") 0
('") 0 ..- C'! 0
... (.) ~ co cri 00 co 0 0
:::l .2 ('") 0')
U ..-
<t
S..c: ..c:Qj
..c:Cllj~:l2Qjj..:!
ell E'oiii:>..:! 0 Cll
CllQj"'>OCll"'>
U g: C) c.. (5) C)
"C <t c.. Cll<c..<c..
..9:! Cll U c..Cll/Cll
oe:uQj(3uu=
OCllQjCl "C=Cll
E Cl ... "C.-Cll E
ll. ..c: ... Cll '-:2:E
~ j ...J :2:JIJ: tJ)
>-
t::
:::l
C'"
W
w_
LL Qj
<t-c
W Cll
-:2:
uCl
tJ) e:
:2: ,-
Cl
..
Qj
E
W
Cll
-
o
I-
f
:~.
Pension Trustee Cover Memorandum
Trackinq Number: 1,184
Actual Date: 03/14/2005
Subiect / Recommendation:
Approve a contract with Callan Associates Inc. for investment consulting services with an annual
fee of $72,000 and authorize appropriate officials to execute same.
Summary:
At the Trustee's direction, staff issued a Request for Proposal(RFP) for Investment Performance
Measurement. The criteria used to evaluate the proposals included qualifications and resources
of the firm and staff, fees, and the quality of their reports.
A total of 24 firms responded to the RFP. The RFP committee ranked the responses and
requested additional information from the top eleven ranked firms. Those firms are:
Callan Associates
Consulting Services Group LLC
Frank Russell Company
GRS Asset Consultants
Kalson & Associates
Marquette Associates, Inc.
Mercer Investment Consulting
Merrill Lynch Consulting
New England Pension Consulting
Palmer & Cay
PFM Advisors
The RFP committee reviewed the second round of submittals and selected three finalists for
presentations. After those presentations the RFP committee ranked the firms as follows:
Callan Associates
Palmer & Cay
Kalson & Associates
The annual fee for the performance measurement service will be $72,000. Any additional
services (such as manager searches) will require additional fees and will need the approval of
the trustees.
Oriqinatinq: Finance
Review Approval
Maraie Simmons 02-28-2005 16: 19:23
Cvndie Goudeau 03-11-2005 08:32:45
Bill Horne 03-10-2005 18:08: 10
Tina Wilson 03-07-2005 09:04:21
Garrv Brumback 03-07-2005 14: 54: 52