Professional Documents
Culture Documents
1 | S U M M E R 2008
The root cause of asset bubbles is not too-lax credit but too-gullible human beings.
Behavioral finance sheds light on the origins of bubbles—and points the way for investors
to capitalize on the busts that invariably follow.
FOR CENTURIES, INVESTORS HAVE FALLEN > Investors rationalize the soaring prices
prey to the lure of a heady rise in stocks or by telling themselves that some unique or
commodities or houses, and these overzealous revolutionary aspect of the inflated asset has
buyers have suffered losses in the bust that has rendered the old economic rules obsolete.
inevitably ensued (Display 1, following page).
With each bubble/bust cycle comes investors’ > At last (though it’s always hard to predict
refrain: “This time it’s different.” But while when), some failure—a scandal, an earnings
the object of affection-then-disdain may vary, shock, or a sign of weakness in the support
bubbles and busts exhibit a predictable pattern: for the boom—signals the old rules do in
fact still apply, rattling investors and precipi-
> A sharp rise in the price of stocks or some tating a crisis of confidence. Prices start to
other object of speculation sparks the interest fall, gathering momentum as investors’ panic
of investors. contaminates even good investments.
> Hunger for the coveted investment fuels > Eventually the fears subside and “normal”
borrowing, which fuels demand that drives market drivers reassert themselves, creating
prices up further. With no end to the outperformance opportunities for those inves-
increases apparently in sight, still more inves- tors who remained true to fact, not euphoria.
tors are lured into the marketplace, thereby
pushing up prices even higher. All the while, The US Housing Bubble
capital flows freely, and often some financial The US housing bubble and its painful unravel-
innovation facilitates the leveraging process. ing are a vivid reminder that markets have a
disturbing habit of becoming grossly overin-
> As a bubble builds, valuations become more flated—and then collapsing, inflicting damage
and more distorted; fundamental consider- on investors’ wealth. Home prices started to
ations, such as realistic profit potential, are rise in the late 1990s, as low interest rates
disregarded amid the prevailing irrational and easy credit facilitated by fi nancial innova-
exuberance. tions—notably, subprime mortgages pitched to
The specific securities identified and described in this article do not represent all of the securities
purchased, sold, or recommended for the strategy or the portfolio, and it should not be assumed
that investments in the securities identified were or will be profitable. Please note that the specific
securities discussed here may no longer be held within the strategy or portfolio.
SUMMER 2008 | 1
Display 1
Bubbles are as old as investing itself, but no matter what the chosen asset is, the pattern of investor
self-delusion persists
Bubbles Across the Centuries
borrowers with limited or bad credit history— 50 million homes with mortgages, as many
made home ownership widely accessible. as 7.5 million could have seriously negative
equity—that is, be worth significantly less than
In a classic pattern, the favorable environment the principal remaining on their mortgages—if
drove increasing demand for homes, propelling the average house price were to fall 20% from
home prices higher. The escalating prices the 2006 peak (Display 2).
attracted speculators, causing still more money
Display 2
to flow into housing as buying and selling
Many homes could soon be worth less than the
became fervid. By 2005, buyers and lenders alike principal remaining on their mortgages
reassured themselves that prices could only go
up, and buyers used their houses’ values to fund Houses with Negative Equity ≥10%
Depending on Price Declines*
a spending spree. Credit standards eased further. (Millions)
The June 13, 2005, cover of Time magazine, 10.0
Display 3
SUMMER 2008 | 3
out-of-favor stocks despite their longer-term Display 4
promise, or to avoid risk even at the cost of a A loss hurts twice as much as a gain pleases
reduction in yield or return? Loss Aversion
The average investor’s results lag the market— Distinguishing a permanent problem from a temporary
because of flawed, emotional decision making setback is critical to value investing
Annualized Returns
1988–2007
11.8%
Long-Term
Earnings Power
Investment
Controversy
4.5% Which?
3.0%
Profits & Stock
Price Decline
S&P 500 Inflation Average Stock Research
Fund Investor Conclusion
“Value Trap”
Past performance is no guarantee of future results.
Source: Dalbar, Inc. Source: AllianceBernstein
Risk-aversion features here, too, as investors The emotional challenge, though, in value
tend to shy away from stocks when times get investing is that the greatest buying opportuni-
difficult, thus locking in losses and missing ties are created when negative sentiment is at its
out on surges. most intense—the very time when buying into
controversy seems entirely irrational.
From Destructive Bias to Advantaged Insight For example, think back to 1990, when big
When faced with such faulty decision-making American banks were in a not-unfamiliar
frameworks, can an investor come out ahead? position: mired in a mortgage crisis, amid the
The answer is yes, although what it takes to worst downturn in commercial real estate since
succeed does not always feel good. the 1930s. On top of that, banks were still
coping with soured developing-country and
Perhaps the greatest example is value-stock leveraged merger loans. As most investors saw
investing. As Bernstein practices it, value it, the “troubled” big banks would go on hem-
investing seeks out stocks that intensive orrhaging loan losses, and, what’s more, their
research reveals are exceptionally inexpen- basic business model was becoming obsolete.
sive versus their long-term earnings potential. People assumed a number of big banks would
During booms, these are often stocks of collapse, along with many S&Ls.
companies not associated with the fad of
the moment; during busts, when widespread Our research uncovered a different picture:
anxiety fuels high levels of risk aversion and We felt certain out-of-favor banks had the
often indiscriminate avoidance of stocks, the profit potential to more than cover their bad
opportunity set for value investors widens. As loans and that industry consolidation plus
Display 6 illustrates, the pertinent question for cost-cutting and fee-raising would in fact
a value investor is: Is the stock price depressed spark an earnings boom. Convinced that
because of some problem from which the loss-averse investors had vastly overreacted,
company is unlikely to recover, or is it in fact we sought out beleaguered but basically sound
a setback that ultimately will be corrected, banks, like Citicorp (whose total return was
restoring the company’s long-term earnings down 52% in 1990); Chemical Bank, (59)%;
power—and its stock price?
SUMMER 2008 | 5
and Bank of Boston, (64)%. More stringent overhaul, IBM stock embarked on an impres-
lending policies soon brought loan losses under sive recovery (Display 8).
control, and merger synergies helped restore
Display 8
profits. Banks and bank stocks rebounded
dramatically (Display 7). From 1990 through Investors’ emotional overreaction to the computer
maker’s travails created a value opportunity
1996, Citicorp rewarded investors with 338%
in total return; Chemical, 321%; and Bank of IBM Stock Price
Boston, 318%. Profits
Recover
Display 7
$180 Massive Restructuring,
Layoffs, Cost-Cutting
Earnings
Bank stocks suffered in 1990 but ultimately 120 Plummet New
rebounded, outperforming the market
Management
60
Bank Stocks vs. S&P 500
199% 0
156% 92 93 94 95 96 97
Note: IBM went on to split two-for-one in May 1997, and again in
May 1999.
S&P Bank Source: FactSet and AllianceBernstein
500 Stocks
(3)%
(38)% Value investing thrives on controversy, which
1990 Cumulative Return invariably has an emotional component.
1990–1996 Although dominant in computer printers,
Past performance is no guarantee of future results. Hewlett-Packard in 2001 was subscale
Source: Standard & Poor’s and AllianceBernstein in personal computers and hardware and
services. By the end of 2002, the market valued
In a fast-changing world, investor opinion often HP more like an appliance-maker than a high-
veers rapidly from one extreme to another, as tech company. And when HP announced its
behavioral finance has documented. A shining acquisition of Compaq in 2002, the merger
star of high tech for decades, IBM fell on hard appeared merely to combine two weak PC
times in the late 1980s, as demand for main- makers, thereby diluting the high-value printer
frames shrank and its share of the personal business. Compounding the investment con-
computer market fell from 100% in 1981 to troversy was a formidable rival in Dell and a
a mere 16% 10 years later. By 1993, IBM was high-profile CEO at HP who served as a light-
earning less than $1 a share, and its stock price ning rod for investor anxiety. Many doubted
fell to $41 in August of that year, down from HP management could make the merger work,
$100 in July 1992. and the stock continued to slide through 2002.
Nevertheless, under new management IBM However, our objective research persuaded
began to reboot itself, although investors us that HP’s prospects were underrated. The
remained skeptical. Our research indicated merger fundamentally improved the company’s
the company’s strategy was sound and that competitive position in both personal comput-
the stock was significantly undervalued. IBM ers and enterprise hardware. Greater leverage
revamped its product mix to increase revenues; over suppliers and a tougher negotiating
sharply cut costs (expenses went from 43% of stance allowed HP to boost gross margins,
sales in 1991 to 28% in 1996); and boosted and it decisively lowered operating costs by
earnings per share more than tenfold. When reducing head count. Operating earnings on
investors finally began to catch on to the the PC business went from a loss of nearly
SUMMER 2008 | 7
US Strategic Value (All Accounts)
1. General Notes: These monthly and quarterly performance figures are geo-
metrically linked to calculate cumulative and/or annualized
a. Performance Statistics Are Not Financial Statements—
“time-weighted” rates of return for various time periods.
There are various methods of compiling or reporting
Closed accounts are included in the composite for each full
performance statistics. The standards of performance
quarter prior to their closing.
measurement used in compiling these data are in accor-
dance with the methods set forth below. Past performance d. Benchmark—The benchmark for the composite is the
does not guarantee future results. A portfolio could suffer S&P 500 Index. The S&P 500 Index is widely regarded
losses as well as achieve gains. as the standard for measuring large-cap US stock market
performance.
b. Composite Structure—Beginning in 1993, the Bernstein
US Strategic Value (all accounts) composite (the “com- 2. Net-of-fee performance figures for the composite have
posite”) includes only fee-paying private and institutional been calculated as follows:
discretionary accounts not subject to significant investment
restrictions imposed by clients. From 1974 through 1992, a. Prior to 1983, management fees were not charged;
the composite includes all private and institutional discre- instead, the accounts incurred transaction costs.
tionary US Strategic Value accounts. b. From 1983 through 1992, the composite’s net-of-fee
c. Rate of Return—Performance returns for each account return is the equal-weighted average of the actual after-
are calculated monthly using trade-date accounting. fee returns of each account in the composite. From 1993
Performance results are reported on a total-return basis, forward, the composite’s net-of-fee return is the asset-
which includes all income from dividends and interest, weighted average of the actual after-fee returns of each
and realized and unrealized gains or losses. Prior to July account in the composite.
1993, all cash flows were assumed to have occurred on the c. Net-of-fee returns for the past 10 years are as follows:
last day of the month. From July 1993 through 2000, if an 1998: 10.1%; 1999: (0.2)%; 2000: 10.0%; 2001: 9.3%;
account’s net monthly cash flows were equal to or exceeded 2002: (17.6)%; 2003: 32.0%; 2004: 13.5%; 2005: 8.6%;
10% of its beginning market value, the Modified Dietz 2006: 20.1%; 2007: (1.2)%.
Method was used to daily-weight the cash flows. When an
account’s net monthly cash flows were less than 10% of its 3. Dispersion—Dispersion is calculated on the gross-of-fee
beginning market value, the cash flows were assumed to annual returns of the accounts included in the composite
have occurred on the last day of the month. Beginning in for all 12 months of the calendar year; it is the asset-
2001, all cash flows are daily-weighted using the Modified weighted standard deviation of these returns. Dispersion of
Dietz Method. Beginning in 1993, the monthly composite returns for the composite is as follows: 1998: 2.0%; 1999:
returns are calculated by weighting each account’s monthly 2.0%; 2000: 1.6%; 2001: 1.7%; 2002: 1.6%; 2003: 1.4%;
return by its beginning market value as a percent of the 2004: 1.2%; 2005: 1.1%; 2006: 0.8%; 2007: 1.1%.
total composite’s beginning market value. Prior to 1993, the
composite results are equal-weighted on a quarterly basis.
At least one silver lining has emerged from the recent dark clouds over financial markets:
Low interest rates and depressed asset valuations make it easier to escape gift and estate
taxes. But be careful not to rush into long-term plans based on short-term economic trends.
Seemingly attractive market opportunities can be misleading, although any opportunity that
triggers a review of your estate plan is useful.
WHEN THE FEDERAL RESERVE SLASHED Nonetheless, given the recent upheaval in the
interest rates early this year, a spate of head- markets and the decline in interest rates, many
lines declared the time was right for certain individuals and their professional advisors are
tax-minimization strategies. As the Wall Street being tempted to establish long-term GRATs
Journal wrote: “Low interest rates and the in order to “lock in” a low rate. For a fresh
slumping market make this the best time in at perspective on the effect of interest rates and
least five years to use so-called estate-freeze market conditions on wealth transfer, we con-
strategies, which can help shield a fortune ducted new research on GRAT strategies. The
from estate taxes, say wealth advisors.”1 results were eye-opening:
While it is true that low interest rates and a > Contrary to common wisdom, locking in
slumping market may facilitate a few such a low interest rate on a long-term GRAT is
strategies, trying to “time” estate planning almost certainly not the best choice among
according to market conditions is shortsighted. GRAT strategies for transferring liquid
Smart estate planning is timeless: You can assets. A series of rolling short-term GRATs
shield wealth from gift and estate taxes in will most likely outperform a long-term
any interest rate or market environment. For GRAT regardless of interest rates.
example, as Bernstein research has shown, a
series of “rolling” short-term grantor-retained > Further, a rolling GRAT strategy will most
annuity trusts (GRATs) funded with publicly likely outperform a long-term GRAT regard-
traded stocks is a simple and effective wealth less of the stock market environment at the
transfer strategy, regardless of market condi- strategy’s inception.
tions at its inception.2
1
“Market Slump Means Time Is Right for Strategies to Curtail Estate Taxes,” Wall Street Journal, April 1, 2008, page D4.
2
See Keeping It in the Family: Planning for Effi cient Wealth Transfer, Bernstein, May 2006.
Bernstein does not provide tax, legal, or accounting advice. In considering this material, you should
discuss your individual circumstances with professionals in those areas before making any decisions.
SUMMER 2008 | 9
The Case for Wealth Transfer Today Display 2
When interest rates drop, you can expect to The hurdle rate set by the IRS is near an all-time low
hear that “it’s a great time to GRAT,” because
7520 Rate
a key rate set by the IRS, the so-called Section
12%
7520 rate, is integral to the GRAT structure. May 1989 11.6%
10
As shown in Display 1, a GRAT is a trust that
Average 6.8%
allows you to move a portion of the future 8
return on assets to beneficiaries without paying
6
gift tax or using your gift-tax exemption. You
make an irrevocable gift to the GRAT and 4
retain the right to receive an annuity payment
from it each year during the trust’s term. 2 June 2008 3.8%
89 91 93 95 97 99 01 03 05 07 08
Anything left in the GRAT at the end of its
term passes to your beneficiaries. If you struc- Source: US Treasury Department
ture the GRAT so that it is “zeroed-out”—in
other words, so that the present value of the A low 7520 rate sets a lower “hurdle” for a
annuity payments equals the value of the assets GRAT to succeed in transferring wealth. If the
you transfer to the GRAT—then you have assets in the GRAT grow at a rate in excess
made no gift under tax law. of the 7520 rate, all that excess growth (the
“remainder” of your GRAT) will transfer to
Display 1
your beneficiaries free of gift tax. This is the
A simple GRAT structure allure of locking in a low 7520 rate for a long
GRAT You (the Grantor) term—it should be easier to beat—but it may
Contribution
not be the wisest choice. For a more nuanced
GRAT Personal Assets perspective, let’s take a look at an alternative
GRAT strategy—rolling short-term GRATs.
Annuity
Remainder Payments Taxes
In a rolling GRAT strategy (Display 3), you
Government create a short-term GRAT (say, two years) and
use each year’s annuity to create a new GRAT.
Your Income Taxes on
Beneficiaries All Trust Income You can keep doing this for as many years
as you want, but for the sake of comparison,
Source: AllianceBernstein suppose you continue this process for 10 years.
During this time you will have created nine
Here’s why the 7520 rate matters: It is the two-year GRATs (the final one expires at the
interest rate for the present value calculation. end of year 10).
The IRS publishes the 7520 rate monthly. In
simple terms, it is set at a modest premium Previous Bernstein research has shown that a
to prevailing intermediate-term interest rates. rolling short-term GRAT strategy will most
As Display 2 shows, the 7520 rate is near its likely outperform a single, long-term GRAT
historical low. for two main reasons:
Second, the shorter, two-year time horizon Just How Important Are Low 7520 Rates?
minimizes the chance that good investment One might argue that in the example above,
performance in one year will be offset by the 7520 rate was much higher than it is today:
poor investment performance in another year. In 1998 it was 7.2%; in June of this year it
Even if the compound return during a 10-year was 3.8%. Shouldn’t such a low rate provide
period is poor, there may be good two-year a powerful tailwind to the long-term GRAT
periods along the way that will successfully structure? Since no one can predict what the
transfer wealth. markets will do in the future, we approached
the question in two ways: 1) by forecast-
As an example, consider the decade that just ing potential future market returns, using
ended: 1998 through 2007. This dramatic Bernstein’s proprietary Wealth Forecasting
period saw the rise of the tech stock bubble, the System,3 and 2) by looking at history—which in
depths of the ensuing bear market, and a strong some periods (the 1940s and early 1950s) had
recovery. From start to finish of this roller- interest rates even lower than today’s.
coaster ride, stocks turned in a 5.9% annualized
return. Yet, as Display 4, following page, shows, First, using our Wealth Forecasting System, we
a 10-year GRAT funded with $5 million in 1998 modeled 10,000 future market environments
would have failed to transfer any wealth. A series to show their effect on two GRAT strategies:
of rolling two-year GRATs, however, would a single 10-year term GRAT versus a series of
have succeeded in six of the 10 years, transfer- two-year rolling GRATs for 10 years. Next, we
ring $3.4 million out of your estate free of gift looked at the GRATs beginning in low interest
tax. This shows how the rolling GRAT strategy rate periods—defined as the lowest quartile of
capitalizes on the volatility of the stock market. 7520 rates modeled. The rolling GRAT strategy
3
See Notes on Wealth Forecasting Analysis, page 34.
SUMMER 2008 | 11
Display 4
structure that uses its first annuity to create a
1998–2007: How rolling short-term GRATs capitalized nine-year GRAT, and its next annuity (plus the
on volatility nine-year GRAT’s first annuity) to create an
10-Year Term GRAT
eight-year GRAT, and so on, keeping all the
Initial Contribution: $5 Mil. money at work and ending with a two-year
GRAT. Yet, as Display 5 shows, this strategy
40%
30 29% 29 will likely underperform the simple rolling
20 21
16 short-term GRAT strategy—both in terms of
10 11
5 5 success rate and the median amount transferred.
0
(10) Display 5
(20) (9) (12)
(30) Even with low interest rates, rolling short-term GRATs
(22)
beat other GRAT strategies
Initial Section 7520 Rate: 7.2%
10-Year Term
10-Year S&P Compound Return: 5.9% 10-Year GRAT with Annual Rolling
Term Decreasing Term Short-Term
Remainder: $0.0 GRAT* GRATs* GRATs
1941–1998: Rolling short-term GRATs beat 1941–1998: Rolling GRATs beat term GRATs in both bull
long-term GRATs and bear markets
All strategies are funded with $10 million and invested in a portfolio
The results were striking: The rolling short- representative of the S&P 500. Wealth to beneficiaries is reinvested and
term GRAT strategy beat the 10-year term adjusted for inflation. See Notes on Wealth Forecasting Analysis, page 34.
Term GRATs assume 20% increasing annuities, while rolling GRATs
GRAT in every period, and succeeded 100% of assume constant annuities.
Source: AllianceBernstein
the time at transferring wealth to the next gen-
eration, while the long-term GRAT succeeded
only 80% of the time, as Display 6 shows. long-term GRAT might outperform rolling
short-term GRATs. But as Display 7 shows,
Not only was the rate of success higher, but the rolling GRATs outperformed term GRATs
amount of wealth transferred was much greater. by wide margins, no matter what the stock
Even when the 10-year term GRATs succeeded, market had done in the year prior to the
the rolling GRAT strategy transferred nearly strategies’ inceptions. This reflects the fact that
twice as much wealth: a median transfer of historically, stocks have tended to rise over any
$11.0 million compared with $6.1 million. 10-year period, with market downturns being
relatively short in duration.
We also compared shorter-term GRATs,
because a 10-year term might not fit every In summary, our research provided an over-
individual’s needs. The results were compa- whelming case against trying to “time”
rable: Running the same historical analysis your wealth transfer strategy with GRATs.
using four-year term GRATs versus four years Regardless of interest rates or the current
of rolling two-year GRATs, the rolling strategy market environment, a strategy of rolling short-
succeeded 98% of the time, compared with term GRATs funded with publicly traded stocks
79% for the term GRATs. And out of 684 appears very likely to provide better results than
trials, the four-year GRAT transferred more a single, longer-term GRAT.
wealth than the rolling strategy only 18 times,
or in just 2.6% of the trials. When Timing Does Pay Off
A few wealth transfer strategies will benefit
An All-Market Strategy from low interest rates. For example, our
What about the stock market environment? research showed that charitable lead annuity
One might argue that if the stocks in a GRAT trusts (CLATs) should perform better when
were poised for an upsurge (even though launched in low interest rate environments.
no one can predict future performance), a CLATs resemble GRATs, with the main dif-
ference being that the annuities go to charity,
SUMMER 2008 | 13
rather than back to the donor. As a result, However, CLATs are not right for everyone,
a rolling short-term structure is impossible or every situation. First, success is by no
because each year’s annuity goes to charity and means guaranteed. Even in the low interest
is out of your hands. This favors a longer-term rate scenarios, only 88% of the CLATs in
structure, so that the assets have plenty of time our simulation had money left at the end of
to grow. Consequently, CLATs are typically their term. So if your primary goal is to pass
created with long terms—10 to 20 years. money on to the next generation, better wealth
transfer vehicles exist. Further, if your primary
We used our Wealth Forecasting System to goal is giving to charity, there may be better
model hypothetical 20-year CLATs funded ways to do so.
with a $10 million initial contribution. We
compared how CLATs would fare if started in
It’s Always a Good Time for Estate Planning
a low interest rate environment (defined as the
lowest quartile of interest rate scenarios) versus One of the well-worn maxims of investment
high interest rates (the top quartile). Display 8 management is that it doesn’t pay to time the
shows the results: CLATs perform much better market. This wisdom applies to wealth transfer
with a low 7520 rate. strategies as well. While a few specialized
strategies, such as CLATs, may benefit from
Display 8
low interest rates, it is better to start early
CLATs benefit from a low hurdle rate with a comprehensive estate plan designed to
weather any market environment, rather than
20-Year CLAT
Median Wealth Transfer waiting for the market to turn your way.
$6.0
$5.2 Mil. The reasons are simple: First, as our research
5.0
above shows, a rolling short-term GRAT
4.0 strategy can effectively transfer wealth regard-
3.0 less of interest rates or market environment.
2.0 Second, by trying to time the market you run
$1.5 Mil. the risk of using up your “time capital,” that
1.0
is, the amount of time you have to transfer
0.0
wealth during your lifetime. In other words,
Low Rates High Rates
for every year you procrastinate, you lose the
Probability of ability to transfer assets and have them grow
passing wealth
to beneficiaries 88% 59% outside of your estate, and you increase the
risk that upon your death you will leave an
Low rates are defined as the bottom quartile of potential interest rate
scenarios; high rates defined as the top quartile. Wealth to beneficiaries is overly large estate for the government to tax.
adjusted for inflation.
CLAT strategies are funded with $10 million, with assets invested in a
diversified portfolio composed of 60% equities (35% US value/ 35% US If, however, the current market environ-
growth/25% developed international/5% emerging markets stocks), 30%
taxable bonds, and 10% REITs. See Notes on Wealth Forecasting Analysis, ment induces you to “use lemons to make
page 34. lemonade,” today is as good a time as any
Source: AllianceBernstein
to consider estate planning. Just be sure to
consult your tax advisor and use a thoughtful
If a CLAT strategy suits your estate planning approach grounded in research, rather than
goals, today’s market conditions are ideal. relying on common wisdom. ■
Amid the unusual developments in the bond market during the past year, an actively managed
bond portfolio guided by rigorous research and a disciplined investment process avoided the
pitfalls of a greedy market while taking advantage of its opportunities—earning significant
premiums along the way.
Display 1
IN BERNSTEIN’S VIEW, BONDS ARE AN
essential complement to stocks in an inves- The shift from great calm to extreme fear caused the
tor’s overall portfolio. They provide income, municipal yield curve to steepen dramatically
and they stabilize the total portfolio against
AAA Municipal Yield Curves
stocks’ volatility. In addition, a bond portfolio
can generate incremental return when actively 5.5%
managed using a sound research framework. March 31, 2008
4.5
The investing landscape of the past 12 months
provides the backdrop for a compelling narra- 3.5 March 31, 2007
tive of how active bond management can add
value across all three strategic imperatives. 2.5
1.5
Setting the Stage: The Hunt for Yield 1-Year 6-Year 20-Year 30-Year
For several years prior to the summer of 2007, Estimates for AAA-rated insured municipal bonds
the stock and bond markets were unusually Source: Municipal Market Data Corp. and AllianceBernstein
calm. With risk seemingly absent, many Municipal bonds were not immune to the
municipal bond managers were grabbing yield contagion: Buyers rushed to purchase safer
anywhere they could find it, purchasing both (that is, higher credit-quality and shorter
very long-term and lower-rated bonds. As a maturity) bonds, causing prices to rise in
result, the shape of the municipal yield curve response to demand. And in its efforts to quell
had become flatter than it had been in decades: the subprime crisis, the US Federal Reserve
With so many eager buyers, long bonds offered aggressively cut interest rates several times,
little additional yield versus short-term bonds of helping to further lift bond prices. As prices
the same credit quality. Plus, the additional yield and yields are inversely correlated, by the
for riskier bonds fell to historically low levels. end of March 2008, the yield curve had gone
from flat to steep: Yields on short-term bonds
But the reign of calm came to an abrupt end in had fallen, while yields on riskier longer-term
July of last year, as the effects of the subprime bonds had risen (Display 1). Managers holding
crisis paralyzed the global markets.
SUMMER 2008 | 15
the lower-rated or longer-term bonds saw their Display 2
prices plummet. The dramatic change in the In early 2007 we concentrated the maturity structure of
yield curve and the increased demand for high- our portfolios
quality bonds reflected the switch in investor Our Maturity Selection
sentiment from greed to fear. 5%
88% 2%
4 10%
The Calm Before the Storm
Back in the spring of 2007, it was clear that
3
market risk was absent from many investors’
minds—they weren’t demanding much more 2 Duration Target
(4.1 Years)
yield for holding long bonds. We determined
that the modest additional yield being offered 1
by long-term bonds was not sufficient compen- 1-Year 6-Year 20-Year 25-Year
sation for the mounting risk that these bonds’ As of March 31, 2007
Source: Municipal Market Data Corp. and AllianceBernstein
prices could fall; when income and price
return were both considered, our calculations
indicated our investors would be best served Our analysis of credit quality revealed a
if we concentrated holdings in the intermedi- similar story. By March 2007, credit spreads
ate-maturity range, resulting in a four-year had become extremely compressed, meaning
duration overall (Display 2). the premium being offered to take on the extra
SUMMER 2008 | 17
Display 3 Assessing Insured Bonds’ Risks
Bond insurers’ ratings have declined Insured bonds are structured assets with two
parts—the insurance and the underlying credit
Ratings of Bond Insurers*
of the issuer. In evaluating an insured bond for
Mar 07 Outlook Mar 08 Outlook
inclusion in a portfolio, Bernstein has always
Ambac AAA stable AA negative
used our credit research to separately evaluate
FGIC AAA stable BB negative
the insurer and the underlying bond.
FSA AAA stable AAA stable
MBIA AAA stable AAA negative
In 2007, our analysis indicated that buying
XLCA AAA stable BB negative
insured bonds whose underlying issuer was
*As of March 2007, all three rating agencies gave the same ratings. March poorly rated didn’t offer enough of a yield
2008 data (as of March 31) reflect the lowest rating assigned by one of
the three rating agencies. benefit versus those where the underlying
Source: Fitch Ratings, Moody’s Investor Services, and Standard & Poor’s
issuer was highly rated, so we bought bonds
of municipalities with solid underlying credit;
By March 2008, due to the subprime crisis insurance was not a material part of the instru-
and exacerbated by the downgrading of bond ment. As a result, the underlying credit rating
insurers, the difference in yield offered by of our insured holdings averaged a very high
high-quality investment-grade municipal bonds AA-/A+. When the bond-insurer storm hit, we
versus those of lower quality had widened knew that even if the insurance were completely
sharply (Display 4). stripped away, the value of our insured-bond
holdings would not change materially. Thus,
Display 4
the downgrades did not significantly affect our
The subprime crisis caused municipal bond credit portfolios. We further insulated our portfolios
spreads to widen sharply
from exposure to any single insurer by virtue
Yield Advantage of BBB Debt over AAA Debt of strict portfolio construction guidelines that
assured our portfolios were well diversified by
1.5% From June 30, 2007, to March 31, 2008,
spreads widened by 69 basis points
bond insurer.
1.0
Where Risk Does Bring Reward
0.5 Unusual dislocations in the municipal bond
market continued through the first quarter
0.0
of 2008, including problems in auction-rate
securities, troubles with variable-rate demand
88 91 94 97 00 03 06
notes, and margin calls on leveraged municipal
Through March 31, 2008; 10-year municipal securities portfolios. Again, navigating these develop-
Source: Municipal Market Data Corp.
ments required intensive analysis, which
uncovered some real opportunities.
SUMMER 2008 | 19
The Power of Active Bond Management Bernstein commits substantial resources to
This period of extreme volatility in the usually active bond management: tracking the con-
earthbound municipal bond market provides stantly shifting mix of risk and opportunity in
a useful lens for examining how our research the marketplace; assessing and monitoring the
and the application of our long-term invest- soundness of every bond we buy; and securing
ing paradigm work to protect and build client the best prices for the bonds we buy and sell.
portfolios. In this fast-moving environment, At the core of our operation is an established
returns on our intermediate-duration munici- Bernstein strength: rigorous research. Our goal
pal bond portfolios were among the best in as active bond managers is to use research to
their peer groups for the one-year period identify values in the marketplace created by
ending March 31, 2008 (Display 6). changes in price.
Display 6
The advantages of active management are
Our active management delivered in a turbulent time
clear: It took only six months to move from
One-Year Returns* one of the flattest yield curves in history to one
Bernstein Municipals vs. Peer Groups† with a normal shape and for credit spreads to
5%
move from an all-time low to more attractive
4.69%
4.31 4.46 levels. Thus, our concentrated maturity struc-
4
ture and high-credit-quality stance paid off. By
3
2.60
2.83 optimizing our portfolios to avoid the pitfalls
2 1.95 of this environment while taking advantage of
its opportunities, we were able to create value
1
for our clients. ■
0
Diversified California New York
Capital gains taxes are scheduled to rise from 15% to 20% in 2011, but many think that the
increase could be greater and could come two years sooner, when the next president takes
office and a new Congress convenes. Should a prospective change in tax policy influence your
investment strategy?
would do, J.P. Morgan famously asserted, “It Tax rates and gains realizations: A contrary
will fluctuate.” The same is true of tax rates, relationship
and, as with the markets, their fluctuations can Realized Capital Gains as Percent of GDP vs.
have a sizable impact on individual wealth. Tax Rate on Long-Term Capital Gains
1960–2011*
7 40
expected tax increases. For example, investors *Realized gains as % of GDP in 2006 and 2007 reflect Bernstein estimates;
the capital gains tax rate from 2008–2011 reflects the current schedule.
realized unusually large gains in the last quarter Source: Department of the Treasury—Office of Tax Analysis and
of 1986, in advance of a scheduled tax rate AllianceBernstein
Bernstein does not provide tax, legal, or accounting advice. In considering this material, you should
discuss your individual circumstances with professionals in those areas before making any decisions.
SUMMER 2008 | 21
A Taxing Matter: No Easy Answers Finally, if your expected holding period is
At first blush, realizing taxable gains in the forever (or at least until death parts you from
face of higher future tax rates may seem like your investment, assuming the law still allows
the obvious thing to do. But trying to sidestep a step-up in basis), the question whether to
any presumed or even scheduled shift in tax harvest or defer is moot—capital gains taxes
rates before a comprehensive assessment of all are not an issue. But the situation is more com-
the investment variables involved could prove plicated if you anticipate selling the position
both untimely and imprudent. After all, the in the future, either as part of your natural
benefit of deferring taxes is well established— turnover or to raise cash for a particular need.
when we defer taking gains, money that would To help investors take a more clearheaded,
otherwise have gone to the government instead objective view of this difficult and emotional
can continue to grow, leading to greater future topic, let’s explore the drivers that determine
wealth. And, of course, the investment objec- whether or not accelerating gains ahead of an
tive is to maximize the likelihood of better anticipated tax increase makes sense.
after-tax returns, not simply to minimize taxes.
To Harvest Early or Defer Gains: A Framework
To complicate matters further, there’s another
important (and often overlooked) factor in At the most basic level, the decision to tax-trade
examining the value of skirting any prospec- ahead of an expected increase in rates hinges on
tive rate increase: the very real possibility that the interplay of two key variables: the inves-
tax rates at the end of an investor’s expected tor’s expected horizon, or holding period before
holding period may be very different from the liquidating the portfolio, and the expected
rate expected in the near term. In our 1986 increase in tax rates (Display 2).1 Generally
example, rates fell a decade later by as much as speaking, the shorter the investment horizon,
they rose, and then dropped another five per- the more valuable it is to harvest gains in the
centage points a few years after that. In fact, lower-rate regime—the benefits of deferral are
rates have varied considerably over the years: offset by larger future taxes paid under the
The maximum tax rate on long-term capital higher-rate regime. And, of course, the higher
gains has ranged from a high of almost 40% in the future tax rate, the more incentive there is to
1978 to the current, historic low of 15%. capitalize on the current lower rate.
Display 2
Given that tax rates have fluctuated and likely Harvest or defer? The decision hinges on two
will continue to do so over the coming years, key factors
there may be real danger in focusing myopi-
Defer Harvest Early
cally on the next rate regime, which, if history
is any guide, is not likely to persist for long.
Even more dismaying, changes in capital Longer Investment Horizon Shorter
gains rates have, on occasion, been enforced
retroactively, so that almost any strategy of
anticipating a change may have an outcome
Lower Future Tax Rate Higher
other than the one expected.
Source: AllianceBernstein
1
This leaves out the important question of existing cost basis. Realizing gains on a lower-cost-basis holding means a larger tax upon
liquidation, so a higher future tax rate might appear to incline investors to accelerate their gains realizations. But ultimately, while
cost basis will magnify the difference in wealth between a strategy of harvesting or deferring, it does not significantly impact the
likelihood that a harvesting strategy is preferable.
SUMMER 2008 | 23
Looking for Breakeven Display 4
But our example extends out only one year. Time and taxes: Break-even time horizons for different
rate scenarios
What if Earl had a longer investment horizon,
as most investors typically do? A more practi- Median Increase in Post-Liquidation Wealth
cal way of framing the decision, therefore, is to Sell vs. Hold (%)
than average, then tax deferral has higher Gauging the likelihood of a successful early harvest
value. Should future markets be disappointing, Probability That Harvesting Early Delivers
then harvesting early becomes more attractive. Higher Post-Liquidation Wealth (%)
100%
Of course, we cannot divine the future. But to
help our clients make thoughtful decisions 75
based on a realistic assessment of potential
future outcomes, we’ve developed a quan- 50
titative capital markets model. Our Wealth
25
Forecasting System3 takes the known facts—
our clients’ assets, income needs, risk toler-
0
ance, tax brackets, and time horizon—and 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
runs various investment scenarios through our Liquidation Horizon (Yrs.)
model to project 10,000 possible return paths. 28% Tax Rate 20% Tax Rate 15% Tax Rate
With it, we can analyze the impact of changes
The initial position is a $1 million diversified equity portfolio (35% US
in tax rates across the spectrum of possible value/35% US growth/25% developed international/5% emerging
future capital market outcomes, thus helping markets) with an existing cost basis of $500,000. The current tax rate
on capital gains is 15% and either 20% or 28% in all future years, while
clients make informed decisions regarding the the ordinary income tax rate is 39.6% in all years. The analysis reflects
post-liquidation proceeds and 15.25% annual gains realizations and
probability that accelerating gains realizations assumes future capital losses are used to offset outside gains. See Notes
will result in greater wealth. on Wealth Forecasting Analysis, page 34.
Source: AllianceBernstein
3
See Notes on Wealth Forecasting Analysis, page 34.
SUMMER 2008 | 25
While the allure of building massive wealth by When we add the future capital gains rate
concentrating your portfolio in one or several increases into the mix, the “harvest or defer”
holdings is great, the likelihood of achieving decision is easier: In all of our prospective tax
returns superior to the markets’ is low, and regimes (a static 15% rate, as well as increases
the volatility one can expect to encounter is to 20% and 28%), the benefits of harvesting
high. This higher volatility has a real cost. Our and diversifying the concentrated position, in
research has shown that historically, a typical the median case, appear significant: With an
single stock has compounded at a growth rate increase in rates to 28%, the magnitude of the
almost three percentage points below that of benefit of harvesting early amounts to about
the S&P 500. And for single stocks with the 18% in the first year (Display 7). And, unlike
highest volatility, the results were even worse, a diversified portfolio, for which the advantage
compounding at a rate of only half that of the of front-running the tax increase declines over
market.4 In fact, if we look at a similar liqui- time, with a single stock portfolio the advan-
dation analysis but without any change in tax tage of selling and diversifying never sets on
rates, simply exchanging a single stock position this particular horizon, rising to over 30% in
for a diversified portfolio of global equities, the year 15.
median wealth outcome is superior over every
Display 7
time period (Display 6).
A magnitude of difference: Harvesting a single stock
Display 6
versus diversified portfolios
Exchanging a single stock for a diversified portfolio Median Increase in Post-Liquidation Wealth
Sell vs. Hold (%)
Median Wealth Advantage
Diversified Stock Portfolio vs. Single Stock (%)
30%
25%
20 20
15
10
10
5 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
4
See our comprehensive research study on single stock positions, The Enviable Dilemma: Concentrated Stock—Hold, Sell, or Hedge?
in which the future returns are known with Positively probable: Tax-trading out of a single stock
certainty is not enough. We must consider Probability That a Diversified Portfolio Will Deliver
the full spectrum of portfolio outcomes to Higher Post-Liquidation Wealth (%)
judge the likelihood of any strategy’s success. 80%
When harvesting gains and diversifying the
single stock position, we found that the prob- 60
ability of success remains above 50% across
the entire 15-year time frame of our analysis, 40
The initial position is a $1 million single stock position with 30% volatility
and 1.0 beta to US equities; the analysis assumes a zero cost basis, a cur-
Decision Time? rent tax rate of 15% on capital gains and 15% in all future years, and an
ordinary income tax rate of 39.6% in all years. The portfolio’s allocation is
The moral of the story: Tax rates change over 35% US value/35% US growth/25% developed international/5% emerg-
ing markets, and the portfolio’s results reflect post-liquidation proceeds
time—and they change often and fairly unpre- and 15.25% annual gains realizations; the analysis assumes future capital
dictably. For investors whose expected holding losses are used to offset outside gains. See Notes on Wealth Forecasting
Analysis, page 34.
period is not forever, it’s a difficult decision Source: AllianceBernstein
whether to forgo tax deferral in favor of locking
in today’s low tax rate. For holders of sub- with time. But in the event that a 28% rate
stantial single stock positions, the appeal of looms, there is a higher probability of success
harvesting gains on those stocks now and diver- in accelerating gains, and a superior average
sifying their portfolio is high, in terms of both wealth outcome, and this may bias some toward
the magnitude and the probability of success, taking preemptive action.
regardless of the level of any future tax rate. But
for investors who already hold diversified port- Regardless of the future tax regime, our
folios, the decision is less clear: It will depend Wealth Forecasting System can help clients
both on the investor’s time horizon and the quantify the likelihood and magnitude of
size of the anticipated change in tax rates. At a success of different investment strategies
future capital gains rate of 20%, the magnitude so that they can make the most informed
of the benefit from harvesting early is modest, decision. ■
and the probability of success declines quickly
SUMMER 2008 | 27
China: The Impact of Domestic Policy on the
Future of Global Trade
by Anthony Chan, Asian Sovereign Strategist—Global Economic Research
In its push for industrial expansion, China has released a flood of cheap manufactured goods—
and challenged the world economic order.
OVER THE PAST DECADE, CHINA’S new to them, but also huge—China’s popula-
rapid industrialization has had a profound tion crossed the one billion threshold during
impact on world trade. In its drive to supply the 1980s—and growing fast. China appeared
the world, China has helped push down the to be an El Dorado, a golden opportunity
price of finished goods, while its enormous for massively expanding their global sales.
appetite for raw materials has helped push According to popular lore, shoe manufacturers
up and sustain the prices of energy and com- aimed to sell an additional two billion
modities. This double threat—economists shoes simply by selling one pair to each
call it “downstream deflation” combined person in China.
with “upstream inflation”—has squeezed the
operating margins of foreign manufacturers, In reality, the policy played out very differ-
forcing many of them to either relocate or out- ently from what anyone had anticipated. While
source production to China in order to remain Deng’s reforms opened the economy to foreign
competitive. It has also helped build profits for capital, they also liberated China’s homegrown
corporations and others that have invested in capitalist talent. Chinese-made home appli-
China. Will the next 10 years be anything like ances and other electronic products—likely
the past 10 years? Is China likely to continue to be followed in the not-too-distant future
its current economic strategy? by Chinese cars—are building market share
not only in the developed world but, increas-
ingly, in developing markets as well. And today
When It Started: Deng Xiaoping’s Economic Reforms
China’s textile and garment exports are the
When Deng Xiaoping launched China’s Open targets of a protectionist backlash within the
Door Policy in 1979, his aim was simple: to US and European markets.
raise China’s standard of living in the after-
math of the Cultural Revolution by making Statistics illustrate the scale of China’s impact.
the economy more attractive to foreign inves- In 1997, Chinese finished-goods exports
tors. Deng probably never dreamt that in little averaged about US$10 billion a month. In
more than two decades it would help make his 2007, the figure was close to US$80 billion—
country the factory to the world. an eightfold increase. Imports of raw materials
have grown nearly sevenfold during the same
Foreign companies saw Deng’s economic period, from US$3 billion a month to US$20
reforms as an opportunity to profit from a billion (Display 1).
consumer market that was not only totally
SUMMER 2008 | 29
Display 2
Display 3
Although rural migration is unprecedented… . …40% of the population will still be rural in 12 years
60
% Share
0 50
40
(5) 30
20 Urban Population
(10)
10
0 Open Door Policy
(15)
79 82 85 88 91 94 97 00 03 06 50 55 60 65 70 75 80 85 90 95 00 05 10 15 20
encouraging the urban migration. China’s China Lags the World in Energy Efficiency
leaders must create enough jobs in urban China’s expansion will doubtless keep its
areas to absorb this continuing influx—or appetite for energy and commodities enormous.
risk social and political instability. In the next Its energy consumption has accelerated sharply,
10 years, in fact, China needs to create 69 with its share of world consumption jumping
million new jobs just to support its working- to about 15.5%, far higher than Japan’s and
age population. Therefore, China must—and close to that of the European Union (Display 4).
will—expand its industrial capacity rapidly at Unfortunately, China isn’t using the energy it
all costs. consumes very efficiently. Although the country
made solid energy-efficiency improvements
20
trouble, China’s leadership would face a tough
European Union
choice between righting banks’ balance sheets
15 China
and preserving or creating jobs. On the whole,
10
Japan
we think that job creation and social stability
5 would take precedence and that any shortfall
0 on banks’ balance sheets would be remedied
79 82 85 88 91 94 97 00 03 06 by an injection of public funds. The pressure to
Source: BP Statistical Review of World Energy 2006, CEIC Data, create 69 million new jobs in the next 10 years
and AllianceBernstein
will remain intense, and the incentive to reduce
production capacity in labor-intensive indus-
Corporate Strategy: Expansion, Not Profit, Is the Key tries will be small.
The sharp increase in investment and indus-
trial growth in China highlights the country’s
unique perspective on corporate profitability: Policy Choices and the Protectionist Backlash
Growing the business and increasing market The main threat to China’s ability to maintain
share are the highest priorities; profits come its high-growth, low-margin business model
second. This stands in marked contrast to is the potential for a protectionist backlash
a market economy, where fiscal prudence by other countries. The US and other
eventually dictates that business lending to Organisation for Economic Co-operation and
companies with low profits be scaled back. Development countries have seen their jobs
However, China’s business model is unlikely to and manufacturing plants move to China—
change for one simple reason: Its future under and they’ve seen their domestic markets
that model is promising. China’s potential for flooded with cheaper-priced Chinese goods.
industrial development is enormous. In fact, As China’s industrial revolution continues, this
the recent surge in wages and land prices in a situation will likely become more pronounced.
few highly developed eastern regions is sending If China maintains its current policy, it risks a
just the right market signals to Chinese fac- trade war, a deceleration of the globalization
tories: to maintain their competitive edge by process, and a decline in world trade, all of
starting to relocate to lower-cost regions, which could pose a serious threat to China’s
where the labor supply is abundant and the export-led economy.
vast rural countryside underdeveloped.
SUMMER 2008 | 31
To confront this threat, China could decide Korea, and Taiwan, the correlation between
to bow to external pressure and “revalue” per capita GDP and consumption to GDP
its currency—the renminbi—more quickly, tends to be high and positive—that is, the
strengthening it versus other currencies. In more people earn, the more they tend to spend
theory, if this were to happen rapidly, it would (Display 5, top). The correlation between
increase China’s purchasing power overseas, consumption-to-GDP ratios and savings rates
thereby encouraging domestic consumption, tends to be inverse—that is, the more people
decreasing a massive balance-of-trade surplus, spend, the less they tend to save (Display 5,
and boosting GDP. A stronger currency would bottom). The high consumption-to-GDP ratio
also encourage Chinese manufacturers to start (70%) in the US, for example, is the result not
focusing on producing higher-quality, high-end only of its superior income per capita, but also
goods, and would help enlarge its growing its extremely low savings rate (about 13.5%).
services sector.
Display 5
A stronger Chinese currency would also make The development of the country’s interior will
China’s goods more expensive in foreign increase. Some domestic Chinese companies
markets, which would pose a huge threat to the are already leading the production shift inland,
country’s low-profit manufacturing businesses. and a small but growing number of foreign
While the price of the energy and commodities firms are following them. In our view, China’s
that these businesses import from abroad would industrial sector will become more diverse
drop, that would not be enough to make up the as a result: The developing hinterland will
difference. On balance, we think it unlikely that become the focus of the country’s low-margin,
China will opt for aggressive revaluation of the volume-oriented manufacturing base, allowing
renminbi. Instead, we expect it to continue to production capabilities in the industrialized
pursue its high energy/consumption, low-margin east—where skilled labor and quality infra-
business development model, even at the risk of structure and logistics are readily available—to
retaliatory action from other countries. be upgraded.
SUMMER 2008 | 33
Notes on Wealth Forecasting Analysis
The Bernstein Wealth Forecasting Analysis SM (WFA) are represented by the S&P/Barra Value Index, with an
is designed to assist investors in making a range of key assumed 20-year compounding rate of 8.2%, based on
decisions, including setting their long-term allocation of simulations with capital market conditions as of December
fi nancial assets. The WFA consists of a four-step process: 31, 2007; US growth stocks by the S&P/Barra Growth
(1) Client Profi le Input: the client’s asset allocation, income, Index (compounding rate of 8.1%); developed international
expenses, cash withdrawals, tax rate, risk-tolerance goals, stocks by the Morgan Stanley Capital International (MSCI)
and other factors; (2) Client Scenarios: in effect, questions EAFE Index of major markets in Europe, Australasia, and
the client would like our guidance on, which may touch the Far East, with countries weighted by market capitaliza-
on issues such as which vehicles are best for intergenera- tion and currency positions unhedged (compounding rate
tional and philanthropic giving, what his/her cash-flow of 8.0%); emerging markets stocks by the MSCI Emerging
stream is likely to be, whether his/her portfolio can beat Markets Index (compounding rate of 6.6%); taxable
infl ation long-term, when to retire, and how different bonds by diversified securities with seven-year maturi-
asset allocations might impact his/her long-term security; ties (compounding rate of 5.4%); real estate investment
(3) The Capital Markets Engine: our proprietary model, trusts (REITs) by the NAREIT Index (compounding rate
which uses our research and historical data to create a vast of 5.3%); a single stock with a beta of 1.0, volatility of
range of market returns, taking into account the linkages 30%, and a dividend yield of 0% (compounding rate of
within and among the capital markets (not Bernstein 5.3%); and inflation by the Consumer Price Index (com-
portfolios), as well as their unpredictability; and (4) A pounding rate of 2.5%). Expected market returns on bonds
Probability Distribution of Outcomes: based on the assets are derived taking into account yield and other criteria.
invested pursuant to the stated asset allocation, 90% of the An important assumption is that stocks will, over time,
estimated returns and asset values the client could expect outperform long-term bonds by a reasonable amount,
to experience, represented within a range established by although this is by no means a certainty. Moreover, actual
the 5th and 95th percentiles of probability. However, future results may not be consonant with Bernstein’s esti-
outcomes outside this range are expected to occur 10% of mates of the range of market returns, as these returns are
the time; thus, the range does not establish the boundar- subject to a variety of economic, market, and other vari-
ies for all outcomes. Further, we often focus on the 10th, ables. Accordingly, this analysis should not be construed
50th, and 90th percentiles to represent the upside, median, as a promise of actual future results, the actual range of
and downside cases. Asset-class projections used in this future results, or the actual probability that these results
publication are derived from the following: US value stocks will be realized.
Bernstein was founded in 1967 to manage investments for individuals and families and is dedicated
solely to investment research and management. Today, as a unit of AllianceBernstein L.P., we
oversee some $99 billion* in private capital. Research is the basis of our ability to prudently
manage the assets under our care; it is also the foundation of the full array of investment products,
both global and local, that we offer.
BER–5293–0708 www.bernstein.com