VOLUME VI, NO.

1 | S U M M E R 2008

> The Anatomy of Bubbles and Busts—

and the Opportunities They Create

> Wealth Transfer Strategies:

The Timely and the Timeless

> Using Research to Navigate a

Dynamic Bond Market

> Can Early Tax Harvesting Reap

Greater Gains?

> China: The Impact of Domestic Policy

on the Future of Global Trade

Global Wealth Management

 PERMISSION TO REPRINT MATERIALS FROM THE BERNSTEIN JOURNAL
With the exception of fair dealing for the purposes of research or private study, or criticism or review, no part
of the Bernstein Journal may be reproduced, stored, or transmitted in any form or by any means without prior
permission. To obtain permission to reprint any material from this or prior issues, please call 212.969.6724.

© Copyright July 2008, AllianceBernstein L.P. All rights reserved.

This publication is for use only with the private client business of Bernstein Global Wealth Management, a unit of
AllianceBernstein L.P.
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.
Table of Contents

> The Anatomy of Bubbles and Busts—

and the Opportunities They Create 1

> Wealth Transfer Strategies:

The Timely and the Timeless 9

> Using Research to Navigate a

Dynamic Bond Market 15

> Can Early Tax Harvesting Reap

Greater Gains? 21

> China: The Impact of Domestic Policy

on the Future of Global Trade 28
The Anatomy of Bubbles and Busts—and the
Opportunities They Create
by Dianne Lob, Chairman—Private Client Investment Policy Group
and Beata Kirr, Senior Portfolio Manager, Wealth Management Group

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
prey to the lure of a heady rise in stocks or
commodities or houses, and these overzealous
buyers have suffered losses in the bust that has
inevitably ensued (Display 1, following page).
With each bubble/bust cycle comes investors’
refrain: “This time it’s different.” But while
the object of affection-then-disdain may vary,
bubbles and busts exhibit a predictable pattern:
> A sharp rise in the price of stocks or some
other object of speculation sparks the interest
of investors.
> Investors rationalize the soaring prices
by telling themselves that some unique or
revolutionary aspect of the inflated asset has
rendered the old economic rules obsolete.
> At last (though it’s always hard to predict
when), some failure—a scandal, an earnings
shock, or a sign of weakness in the support
for the boom—signals the old rules do in
fact still apply, rattling investors and precipi-
tating a crisis of confidence. Prices start to
fall, gathering momentum as investors’ panic
contaminates even good investments.

> Hunger for the coveted investment fuels
borrowing, which fuels demand that drives
prices up further. With no end to the
increases apparently in sight, still more inves-
tors are lured into the marketplace, thereby
pushing up prices even higher. All the while,
capital flows freely, and often some financial
innovation facilitates the leveraging process.
> As a bubble builds, valuations become more
and more distorted; fundamental consider-
ations, such as realistic profit potential, are
disregarded amid the prevailing irrational
exuberance.
> Eventually the fears subside and “normal”
market drivers reassert themselves, creating
outperformance opportunities for those inves-
tors who remained true to fact, not euphoria.
The US Housing Bubble
The US housing bubble and its painful unravel-
ing are a vivid reminder that markets have a
disturbing habit of becoming grossly overin-
flated—and then collapsing, inflicting damage
on investors’ wealth. Home prices started to
rise in the late 1990s, as low interest rates
and easy credit facilitated by fi nancial innova-
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

Bubble When Coveted Asset Inspired Idea Telltale Warning

Tulip Mania 1630s Purple-streaked Foreign tulip fanciers would sustain Tulip bulbs cost five times a worker’s
Semper Augustus boom for Dutch speculators annual salary
South Sea 1720 Stock in South Sea Spain granted monopoly on all South Joint stock in company rose tenfold
Company American trade in one year
Mississippi 1729 Stock in Monopoly on trade in “French Shares shot from 500 to 10,000 livres
Mississippi Louisiana” on beaver pelts, gold, in a few months (and plunged as fast)
Company taxes, coinage
Roaring Late 1920s “High-tech” Purportedly transformational “New Massive public participation—and
Twenties stocks of the day Economics” revolving around auto and shares trading at dot-com-style P/E
airplane stocks ratios (until market fell 90%)
Nifty Fifty 1960s–70s Blue Chips like High-P/E, brand-name growth stocks Kodak soared to $148 at its peak,
Kodak, Polaroid, were deemed so reliable as to be only to plunge 60%
and Xerox “one-decision” purchases—buy and
hold forever
Japan Late 1980s Any big Japanese Japan was said to be the unstoppable The Nikkei index hit 38,915 in
stocks new global economic powerhouse ‘89—even today it is back only to
around 13,000
Tech and Late 1990s Internet and The Internet and communications Internet start-ups with no actual
Telecom telecom stocks revolution had supposedly transformed profits, just a business idea, traded
the rules of economics, permitting for triple-digit multiples (Nasdaq
permanent rapid growth index would plunge 80%)
Source: AllianceBernstein

borrowers with limited or bad credit history—
made home ownership widely accessible.
In a classic pattern, the favorable environment
drove increasing demand for homes, propelling
home prices higher. The escalating prices
attracted speculators, causing still more money
to flow into housing as buying and selling
became fervid. By 2005, buyers and lenders alike
reassured themselves that prices could only go
up, and buyers used their houses’ values to fund
a spending spree. Credit standards eased further.
The June 13, 2005, cover of Time magazine,
50 million homes with mortgages, as many
as 7.5 million could have seriously negative
equity—that is, be worth significantly less than
the principal remaining on their mortgages—if
the average house price were to fall 20% from
the 2006 peak (Display 2).
Display 2
Many homes could soon be worth less than the
principal remaining on their mortgages
Houses with Negative Equity ≥10%
Depending on Price Declines*
(Millions)
10.0

which depicted a man hugging a house, bore the 7.5

line: “Home $weet Home.”
5.0
3.0
That may well have marked the housing 1.9
1.1
bubble’s psychological high point. Soon after,
investors became alarmed when overextended 0% (5)% (10)% (15)% (20)% (25)%
subprime borrowers began falling seriously Depreciation
behind on mortgage payments. By 2007, the *Estimates based on what would occur if the average house depreciated
housing market entered an inevitable correc- up to 25% from the 2006 peak
Source: Federal Reserve, National Association of Realtors, and
tion, which is still under way. Of the nation’s AllianceBernstein

2 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

The US housing bubble has had far-reaching Key Concepts
ramifications for the global investing markets. > Asset bubbles and busts—such as today’s
Subprime and other asset-backed loans had
housing crisis—have occurred throughout
been bundled into packages of securities and
sold to financial institutions globally. As history, in remarkably similar guise
indications mounted that the underlying loans > Behavioral finance scientists have identified
were at risk, turmoil grew, and the market emotional traits of investors that ultimately lead
for such securities evaporated. Holders—
to bubbles and busts: among them, believing
predominantly financial institutions and other
that rising markets will keep rising; irrationally
leveraged investors—were forced to mark
down their holdings, often accompanied by fearing even small losses; and preferring instant
margin calls that prompted widespread selling gratification to greater long-term reward
of other liquid securities, especially stocks. > Given human nature, cycles of bubbles and
Stock markets around the globe have tumbled,
busts are inevitable
and stock prices of financial institutions, at the
heart of the maelstrom, have been hardest hit. > These cycles offer great opportunity for value
investing that’s driven by intensive research and
With rising home values no longer available a disciplined process
as the credit line for US consumers’ spending,
fears of a US economic recession have grown,
fanning the flames of investor anxiety. In a How Behavioral Finance Illuminates Bubbles
sign of how pronounced the tidal wave of fear
Why the investing markets are prone to such
had become, as of the end of March 2008,
cycles has been explained, in part, by experts
investors were willing to accept a real (after-
in the field of behavioral finance—in which
inflation) yield of (2.38)% on two-year US
investing, economics, and psychology intersect.
Treasuries—in other words, they were willing
Pioneered in the 1970s by Daniel Kahneman,
to accept a negative investment return for the
winner of the 2002 Nobel Prize in Economic
safety of holding Treasuries.
Sciences, and the late Amos Tversky, the study
of how human beings make decisions when
Although the resolution of the housing bubble
faced with unknown outcomes has shed light
has yet to occur, history may offer some hope
on investor behavior. What compels inves-
for the future: A study of modern stock market
tors to chase “hot” performers long after the
bubbles reminds us that recovery does come
opportunity has passed, or to steer clear of
eventually (Display 3).

Display 3

Stock bubbles burst…and markets recover

Subsequent Percent Percent Change First

Bubble Had to Own Peak Trough Change Year After Trough
Nifty Fifty Kodak $148 (1972) $59 (1974) (60)% 75%
Oil S&P Oil & Gas Index $51.23 (1980) $24.39 (1982) (52) 55
Japan Nikkei index 38,915 (1989) 14,309 (1992) (63) 45
Biotech Amex Biotech Index 257 (1992) 73 (1994) (71) 24
Technology Nasdaq 5,046 (2000) 1,114 (2002) (78) 72

Source: FactSet, Markit, and AllianceBernstein

SUMMER 2008 | 3
out-of-favor stocks despite their longer-term
promise, or to avoid risk even at the cost of a
reduction in yield or return?
In all cases, the answer is predicated on human
emotion or, more precisely, on a set of deci-
sion-making biases that kick in when risk is
involved. Nevertheless, there’s a positive side
to investors’ penchant for engaging in mass
psychological self-deception: It can create
opportunities to profit from emotion-driven
investors’ overreactions to good or bad news.
While the field of behavioral finance is rich
with evidence of a wide array of such biases,
here we focus on just a few of the critical ones
that come into play during boom/bust cycles.
Decision-Making Glitches
Behavioral finance suggests that human beings
are as poorly equipped to be investors as could
be imagined: Our own wiring works against
our very success. For example, through exten-
sive research, Kahneman and Tversky proved
that we feel pain from loss more keenly than
we feel pleasure from gain (Display 4). In
fact, based on their studies, if you won $100
one day and lost it the next, you’d be twice as
unhappy about your loss as you were happy
about your good fortune. Such extreme sen-
sitivity to pain makes investors generally risk
averse, and that trait becomes even more
pronounced when bubbles burst. That’s a
powerful headwind in investing, where greater
return opportunities typically entail more risk.
Another powerful bias that foils good decision
making is “anchoring.” When an investment
has done especially well, investors become
convinced that such performance will continue
into the future, fueling frenetic demand that,
for some period of time, becomes self-fulfilling:
Prices continue to rise. The same holds true on
the bust side: When something has faltered, or
some bad news besets the investment, investors
become adamant that it is permanently imper-
4 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
Display 4
A loss hurts twice as much as a gain pleases
Loss Aversion
Pleasure
Small
Pleasure
Loss Gain
Big Pain
Pain
Source: AllianceBernstein
iled and they can’t get rid of it fast enough,
which hastens the downward spiral of the
investment’s price.
The bait for another thinking trap is “avail-
ability,” or people’s tendency to be influenced
by what they hear most frequently. In today’s
always connected world, headlines can become
stand-ins for “the truth,” and decision making
reflects that.
The Cost of Emotions
These biases are so very natural that fighting
them seems unthinkable. Yet, such emotion-
driven investing can be extremely detrimental,
as Display 5 illustrates. Over the past 20 years,
the average stock mutual fund investor reaped
an annualized return of about 4.5%, barely
outpacing inflation, while the US stock market
grew at close to 12% annualized over the same
period. The marked discrepancy is a function of
flawed choices driven by these biases. Anchoring
and availability tend to cause investors to move
out of an investment that may be lagging and
into one that has been prospering. But doing
so typically means they’ve missed most of the
opportunity in the hot idea and forgone the
developing opportunity in the disappointment.
Display 5
The average investor’s results lag the market—
because of flawed, emotional decision making
Annualized Returns
1988–2007
11.8%
4.5%
3.0%
S&P 500 Inflation Average Stock
Fund Investor
Past performance is no guarantee of future results.
Source: Dalbar, Inc.
Risk-aversion features here, too, as investors
tend to shy away from stocks when times get
difficult, thus locking in losses and missing
out on surges.
From Destructive Bias to Advantaged Insight
When faced with such faulty decision-making
frameworks, can an investor come out ahead?
The answer is yes, although what it takes to
succeed does not always feel good.
Perhaps the greatest example is value-stock
investing. As Bernstein practices it, value
investing seeks out stocks that intensive
research reveals are exceptionally inexpen-
sive versus their long-term earnings potential.
During booms, these are often stocks of
companies not associated with the fad of
the moment; during busts, when widespread
anxiety fuels high levels of risk aversion and
often indiscriminate avoidance of stocks, the
opportunity set for value investors widens. As
Display 6 illustrates, the pertinent question for
a value investor is: Is the stock price depressed
because of some problem from which the
company is unlikely to recover, or is it in fact
a setback that ultimately will be corrected,
restoring the company’s long-term earnings
power—and its stock price?
Display 6
Distinguishing a permanent problem from a temporary
setback is critical to value investing
Long-Term
Earnings Power
Investment
Controversy
Which?
Profits & Stock
Price Decline
Research
Conclusion
“Value Trap”
Source: AllianceBernstein
The emotional challenge, though, in value
investing is that the greatest buying opportuni-
ties are created when negative sentiment is at its
most intense—the very time when buying into
controversy seems entirely irrational.
For example, think back to 1990, when big
American banks were in a not-unfamiliar
position: mired in a mortgage crisis, amid the
worst downturn in commercial real estate since
the 1930s. On top of that, banks were still
coping with soured developing-country and
leveraged merger loans. As most investors saw
it, the “troubled” big banks would go on hem-
orrhaging loan losses, and, what’s more, their
basic business model was becoming obsolete.
People assumed a number of big banks would
collapse, along with many S&Ls.
Our research uncovered a different picture:
We felt certain out-of-favor banks had the
profit potential to more than cover their bad
loans and that industry consolidation plus
cost-cutting and fee-raising would in fact
spark an earnings boom. Convinced that
loss-averse investors had vastly overreacted,
we sought out beleaguered but basically sound
banks, like Citicorp (whose total return was
down 52% in 1990); Chemical Bank, (59)%;
SUMMER 2008 | 5
and Bank of Boston, (64)%. More stringent
lending policies soon brought loan losses under
control, and merger synergies helped restore
profits. Banks and bank stocks rebounded
dramatically (Display 7). From 1990 through
1996, Citicorp rewarded investors with 338%
in total return; Chemical, 321%; and Bank of
Boston, 318%.
Display 7
Bank stocks suffered in 1990 but ultimately
rebounded, outperforming the market
Bank Stocks vs. S&P 500
199%
156%
S&P Bank
500 Stocks
(3)%
(38)%
1990 Cumulative Return
1990–1996
Past performance is no guarantee of future results.
Source: Standard & Poor’s and AllianceBernstein
In a fast-changing world, investor opinion often
veers rapidly from one extreme to another, as
behavioral finance has documented. A shining
star of high tech for decades, IBM fell on hard
times in the late 1980s, as demand for main-
frames shrank and its share of the personal
computer market fell from 100% in 1981 to
a mere 16% 10 years later. By 1993, IBM was
earning less than $1 a share, and its stock price
fell to $41 in August of that year, down from
$100 in July 1992.
Nevertheless, under new management IBM
began to reboot itself, although investors
remained skeptical. Our research indicated
the company’s strategy was sound and that
the stock was significantly undervalued. IBM
revamped its product mix to increase revenues;
sharply cut costs (expenses went from 43% of
sales in 1991 to 28% in 1996); and boosted
earnings per share more than tenfold. When
investors finally began to catch on to the
6 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
overhaul, IBM stock embarked on an impres-
sive recovery (Display 8).
Display 8
Investors’ emotional overreaction to the computer
maker’s travails created a value opportunity
IBM Stock Price
Profits
Recover
$180 Massive Restructuring,
Layoffs, Cost-Cutting
Earnings
120 Plummet New
Management
60
0
92 93 94 95 96 97
Note: IBM went on to split two-for-one in May 1997, and again in
May 1999.
Source: FactSet and AllianceBernstein
Value investing thrives on controversy, which
invariably has an emotional component.
Although dominant in computer printers,
Hewlett-Packard in 2001 was subscale
in personal computers and hardware and
services. By the end of 2002, the market valued
HP more like an appliance-maker than a high-
tech company. And when HP announced its
acquisition of Compaq in 2002, the merger
appeared merely to combine two weak PC
makers, thereby diluting the high-value printer
business. Compounding the investment con-
troversy was a formidable rival in Dell and a
high-profile CEO at HP who served as a light-
ning rod for investor anxiety. Many doubted
HP management could make the merger work,
and the stock continued to slide through 2002.
However, our objective research persuaded
us that HP’s prospects were underrated. The
merger fundamentally improved the company’s
competitive position in both personal comput-
ers and enterprise hardware. Greater leverage
over suppliers and a tougher negotiating
stance allowed HP to boost gross margins,
and it decisively lowered operating costs by
reducing head count. Operating earnings on
the PC business went from a loss of nearly
$1 billion in fiscal 2001 to a profit of nearly significantly outperformed the broad stock
$800 million in fiscal 2007. The Dell threat market since its inception, after fees (Display
abated, as that company’s supply chain was 10, bottom).
replicated by Taiwanese assemblers. And HP
Display 10
countered a competitive challenge to its printer
business by introducing a low-cost unit capable Disciplined value investing involves a rational
approach that has been successful over time
of producing high-quality images. New man-
agement in 2005 added momentum to HP’s US Strategic Value
recovery (Display 9). But again, being able to Growth of $1 Million (after fees)
Jan 1990–June 2007
spot the underlying opportunity when things Bernstein US
Strategic Value:
looked bleakest was key. $7.7 Mil.
$8 Mil.
Display 9 7
6
The market was skeptical of HP’s merger with Compaq,
but our research foresaw that in time it would pay off 5
S&P 500
4 $6.2 Mil.
Hewlett-Packard Stock Price 3
$45 2
40 1
0
35
90 92 94 96 98 00 02 04 06
30 Announced Merger
with Compaq
25
US Strategic Value Since Inception* vs. S&P
Proxy
20 Battle
Growth of $1 Million (after fees)
Jan 1974–June 2008
15 New Management
Appointed $71.4 Mil.
10
$39.6 Mil.
01 02 03 04 05 06 07

Source: Company reports and AllianceBernstein

Strategic Value S&P 500

Bernstein’s history as a value investor suggests Past performance is no guarantee of future results. See notes on
that a disciplined process, driven by intensive performance, page 8.
*January 1, 1974
research, can effectively gauge the oppor- Source: Standard & Poor’s and AllianceBernstein
tunities created by controversy to generate
long-term success. Over the period from 1990
Cycles of booms and busts seem to be unavoid-
through June 2007, which saw the pronounced
able features of investing, and the process of
declines and recoveries referenced above, $1
equilibrium being restored—whether at the
million invested in Bernstein’s US Strategic
individual stock level or, as today, across a
Value portfolio would have grown to $7.7
wide swath of the capital markets—is painful.
million after fees, while a million dollars
But the emotional biases that, in part, underlie
invested in the S&P 500 would have amounted
them can translate into unduly pronounced
to $6.2 million (Display 10, top). And by
security mispricing. For those who rely on deep
adhering to this discipline throughout our
research as a guide, exploiting the discrepancies
history, our US Strategic Value portfolio has
can create significant opportunity for gain. ■

SUMMER 2008 | 7
US Strategic Value (All Accounts)
1. General Notes:
a. Performance Statistics Are Not Financial Statements—
There are various methods of compiling or reporting
performance statistics. The standards of performance
measurement used in compiling these data are in accor-
dance with the methods set forth below. Past performance
does not guarantee future results. A portfolio could suffer
losses as well as achieve gains.
b. Composite Structure—Beginning in 1993, the Bernstein
US Strategic Value (all accounts) composite (the “com-
posite”) includes only fee-paying private and institutional
discretionary accounts not subject to significant investment
restrictions imposed by clients. From 1974 through 1992,
the composite includes all private and institutional discre-
tionary US Strategic Value accounts.
c. Rate of Return—Performance returns for each account
are calculated monthly using trade-date accounting.
Performance results are reported on a total-return basis,
which includes all income from dividends and interest,
and realized and unrealized gains or losses. Prior to July
1993, all cash flows were assumed to have occurred on the
last day of the month. From July 1993 through 2000, if an
account’s net monthly cash flows were equal to or exceeded
10% of its beginning market value, the Modified Dietz
Method was used to daily-weight the cash flows. When an
account’s net monthly cash flows were less than 10% of its
beginning market value, the cash flows were assumed to
have occurred on the last day of the month. Beginning in
2001, all cash flows are daily-weighted using the Modified
Dietz Method. Beginning in 1993, the monthly composite
returns are calculated by weighting each account’s monthly
return by its beginning market value as a percent of the
total composite’s beginning market value. Prior to 1993, the
composite results are equal-weighted on a quarterly basis.
8 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
These monthly and quarterly performance figures are geo-
metrically linked to calculate cumulative and/or annualized
“time-weighted” rates of return for various time periods.
Closed accounts are included in the composite for each full
quarter prior to their closing.
d. Benchmark—The benchmark for the composite is the
S&P 500 Index. The S&P 500 Index is widely regarded
as the standard for measuring large-cap US stock market
performance.
2. Net-of-fee performance figures for the composite have
been calculated as follows:
a. Prior to 1983, management fees were not charged;
instead, the accounts incurred transaction costs.
b. From 1983 through 1992, the composite’s net-of-fee
return is the equal-weighted average of the actual after-
fee returns of each account in the composite. From 1993
forward, the composite’s net-of-fee return is the asset-
weighted average of the actual after-fee returns of each
account in the composite.
c. Net-of-fee returns for the past 10 years are as follows:
1998: 10.1%; 1999: (0.2)%; 2000: 10.0%; 2001: 9.3%;
2002: (17.6)%; 2003: 32.0%; 2004: 13.5%; 2005: 8.6%;
2006: 20.1%; 2007: (1.2)%.
3. Dispersion—Dispersion is calculated on the gross-of-fee
annual returns of the accounts included in the composite
for all 12 months of the calendar year; it is the asset-
weighted standard deviation of these returns. Dispersion of
returns for the composite is as follows: 1998: 2.0%; 1999:
2.0%; 2000: 1.6%; 2001: 1.7%; 2002: 1.6%; 2003: 1.4%;
2004: 1.2%; 2005: 1.1%; 2006: 0.8%; 2007: 1.1%.
Wealth Transfer Strategies:
The Timely and the Timeless
by David Weinreb, Director—Wealth Management Group
and Ted Mann, Analyst—Wealth Management Group

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
interest rates early this year, a spate of head-
lines declared the time was right for certain
tax-minimization strategies. As the Wall Street
Journal wrote: “Low interest rates and the
slumping market make this the best time in at
least five years to use so-called estate-freeze
strategies, which can help shield a fortune
from estate taxes, say wealth advisors.”1
Nonetheless, given the recent upheaval in the
markets and the decline in interest rates, many
individuals and their professional advisors are
being tempted to establish long-term GRATs
in order to “lock in” a low rate. For a fresh
perspective on the effect of interest rates and
market conditions on wealth transfer, we con-
ducted new research on GRAT strategies. The
results were eye-opening:

While it is true that low interest rates and a
slumping market may facilitate a few such
strategies, trying to “time” estate planning
according to market conditions is shortsighted.
Smart estate planning is timeless: You can
shield wealth from gift and estate taxes in
any interest rate or market environment. For
example, as Bernstein research has shown, a
series of “rolling” short-term grantor-retained
annuity trusts (GRATs) funded with publicly
traded stocks is a simple and effective wealth
transfer strategy, regardless of market condi-
tions at its inception.2
> Contrary to common wisdom, locking in
a low interest rate on a long-term GRAT is
almost certainly not the best choice among
GRAT strategies for transferring liquid
assets. A series of rolling short-term GRATs
will most likely outperform a long-term
GRAT regardless of interest rates.
> Further, a rolling GRAT strategy will most
likely outperform a long-term GRAT regard-
less of the stock market environment at the
strategy’s inception.

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
When interest rates drop, you can expect to
hear that “it’s a great time to GRAT,” because
a key rate set by the IRS, the so-called Section
7520 rate, is integral to the GRAT structure.
As shown in Display 1, a GRAT is a trust that
allows you to move a portion of the future
return on assets to beneficiaries without paying
gift tax or using your gift-tax exemption. You
make an irrevocable gift to the GRAT and
retain the right to receive an annuity payment
from it each year during the trust’s term.
Anything left in the GRAT at the end of its
term passes to your beneficiaries. If you struc-
ture the GRAT so that it is “zeroed-out”—in
other words, so that the present value of the
annuity payments equals the value of the assets
you transfer to the GRAT—then you have
made no gift under tax law.
Display 1
A simple GRAT structure
GRAT You (the Grantor)
Contribution
GRAT Personal Assets
Annuity
Remainder Payments Taxes
Government
Your Income Taxes on
Beneficiaries All Trust Income
Source: AllianceBernstein
Here’s why the 7520 rate matters: It is the
interest rate for the present value calculation.
The IRS publishes the 7520 rate monthly. In
simple terms, it is set at a modest premium
to prevailing intermediate-term interest rates.
As Display 2 shows, the 7520 rate is near its
historical low.
10 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
Display 2
The hurdle rate set by the IRS is near an all-time low
7520 Rate
12%
May 1989 11.6%
10
Average 6.8%
8
6
4
2 June 2008 3.8%
89 91 93 95 97 99 01 03 05 07 08
Source: US Treasury Department
A low 7520 rate sets a lower “hurdle” for a
GRAT to succeed in transferring wealth. If the
assets in the GRAT grow at a rate in excess
of the 7520 rate, all that excess growth (the
“remainder” of your GRAT) will transfer to
your beneficiaries free of gift tax. This is the
allure of locking in a low 7520 rate for a long
term—it should be easier to beat—but it may
not be the wisest choice. For a more nuanced
perspective, let’s take a look at an alternative
GRAT strategy—rolling short-term GRATs.
In a rolling GRAT strategy (Display 3), you
create a short-term GRAT (say, two years) and
use each year’s annuity to create a new GRAT.
You can keep doing this for as many years
as you want, but for the sake of comparison,
suppose you continue this process for 10 years.
During this time you will have created nine
two-year GRATs (the final one expires at the
end of year 10).
Previous Bernstein research has shown that a
rolling short-term GRAT strategy will most
likely outperform a single, long-term GRAT
for two main reasons:
Display 3
Rolling GRATs can lock in gains
Annuities Annuities Annuities Annuities
GRAT 1 GRAT 2 GRAT 3 GRAT 4
2 Years 2 Years 2 Years 2 Years
Your Beneficiaries
Source: AllianceBernstein
First, it keeps more of the capital committed
to the strategy. In a single long-term GRAT,
the capital declines each year, as the annuity
payments return assets in the GRAT to you, the
grantor. By contrast, the rolling GRAT strategy
maintains all of the original capital committed
to the strategy in the GRATs.
Second, the shorter, two-year time horizon
minimizes the chance that good investment
performance in one year will be offset by
poor investment performance in another year.
Even if the compound return during a 10-year
period is poor, there may be good two-year
periods along the way that will successfully
transfer wealth.
As an example, consider the decade that just
ended: 1998 through 2007. This dramatic
period saw the rise of the tech stock bubble, the
depths of the ensuing bear market, and a strong
recovery. From start to finish of this roller-
coaster ride, stocks turned in a 5.9% annualized
return. Yet, as Display 4, following page, shows,
a 10-year GRAT funded with $5 million in 1998
would have failed to transfer any wealth. A series
of rolling two-year GRATs, however, would
have succeeded in six of the 10 years, transfer-
ring $3.4 million out of your estate free of gift
tax. This shows how the rolling GRAT strategy
capitalizes on the volatility of the stock market.
3
See Notes on Wealth Forecasting Analysis, page 34.
Key Concepts
> The common wisdom that low interest rates
recommend creation of a long-term grantor-
retained annuity trust (GRAT) may cause
individuals and wealth planners to overlook a
better GRAT strategy
> Building on previous Bernstein research, new
analysis and a study of historical data show
that rolling short-term GRATs are more likely
to transfer more wealth, free of gift tax, than a
long-term GRAT, regardless of interest rates or
market environment
> “Don’t try to time the market” is good advice in
investment management, and our new research
indicates it is equally apt for estate planning
Just How Important Are Low 7520 Rates?
One might argue that in the example above,
the 7520 rate was much higher than it is today:
In 1998 it was 7.2%; in June of this year it
was 3.8%. Shouldn’t such a low rate provide
a powerful tailwind to the long-term GRAT
structure? Since no one can predict what the
markets will do in the future, we approached
the question in two ways: 1) by forecast-
ing potential future market returns, using
Bernstein’s proprietary Wealth Forecasting
System,3 and 2) by looking at history—which in
some periods (the 1940s and early 1950s) had
interest rates even lower than today’s.
First, using our Wealth Forecasting System, we
modeled 10,000 future market environments
to show their effect on two GRAT strategies:
a single 10-year term GRAT versus a series of
two-year rolling GRATs for 10 years. Next, we
looked at the GRATs beginning in low interest
rate periods—defined as the lowest quartile of
7520 rates modeled. The rolling GRAT strategy
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

Two-Year Rolling GRATs Probability of

Success 76% 96% >98%
Initial Contribution: $5 Mil.
Median Wealth
S&P S&P Compound Transfer $3.6 Mil. $5.3 Mil. $7.4 Mil.
7520 Annual Two-Year Wealth
Rate Return Forward Return Transferred All strategies are funded with $10 million, beginning in the lowest-
quartile interest rate environments. All assets are invested in a globally
1998 7.2% 28.6% 24.8% $0 diversified equity portfolio composed of 35% US value/35% US
1999 5.6 21.0 4.9 $1.65 Mil. growth/25% developed international/5% emerging markets stocks.
Wealth to beneficiaries is reinvested and adjusted for inflation. See Notes
2000 7.4 (9.1) (10.5) $179,756 on Wealth Forecasting Analysis, page 34.
2001 6.8 (11.9) (17.1) $0 *The 10-year term GRAT in each case uses 20% increasing payouts to keep
money at work, as do the nine-year through three-year term GRATs in the
2002 5.4 (22.1) 0.1 $0 decreasing term GRAT strategy. Rolling GRATs have constant annuities.
2003 4.2 28.7 19.4 $0 Source: AllianceBernstein

2004 4.2 10.9 7.8 $877,809

2005 4.6 4.9 10.2 $190,887 A Historical Analysis Drives the Case Home
2006 5.4 15.8 10.5 $176,527
While the simulated results were compelling,
2007 NA 5.5 NA $322,580
we were curious to see how the term GRAT
Total Remainder: $3.4 Mil.
and rolling GRAT strategies would have
Term GRATs assume 20% increasing annuities, while rolling GRATs assume
constant annuities. See Notes on Wealth Forecasting Analysis, page 34.
fared historically. We compared the results
Source: Standard & Poor’s, US Treasury Department, and AllianceBernstein of the two strategies—rolling short-term
versus 10-year term—assuming they had been
won handily, succeeding more than 98% of launched every month from May 1941 through
the time, compared with a 10-year term GRAT April 1998, for 684 trials in all. We assumed
success rate of only 76%. each GRAT began with $10 million invested
in the S&P 500. And, since the 7520 rate has
As noted above, one reason a rolling GRAT existed only since 1989, we created a proxy
strategy will likely outperform a single long- for earlier periods based on the IRS methodol-
term GRAT is that it keeps more of the capital ogy. This 57-year span covers a wide range of
committed to the strategy at work. Accordingly, interest rates and stock market returns, with
we also tested the benefit of locking in a low the 7520 rate averaging 6.7%, but dipping as
interest rate by modeling a 10-year GRAT low as 1.2%.

12 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

Display 6 Display 7

1941–1998: Rolling short-term GRATs beat 1941–1998: Rolling GRATs beat term GRATs in both bull
long-term GRATs and bear markets

10-Year Term 10-Year Rolling Short-

Rolling GRATs GRATs Term GRAT Term GRAT
S&P 500 Trailing 12-Month Median Median
Frequency of Return Remainder Remainder
Success 100% 80%
Lowest (38.9)% $6.6 Mil. $11.0 Mil.
Median Wealth Quartile to 2.3%
Transfer $11.0 Mil. $6.1 Mil.
Second 2.5% $6.5 $11.4
All strategies are funded with $10 million and invested in a portfolio Quartile to 14.4%
representative of the S&P 500. Wealth to beneficiaries is reinvested and
adjusted for inflation. See Notes on Wealth Forecasting Analysis, page 34. Third 14.4% $4.3 $10.4
Term GRATs assume 20% increasing annuities, while rolling GRATs Quartile to 26.1%
assume constant annuities.
Source: AllianceBernstein
Highest 26.2% $5.6 $10.9
Quartile to 61.2%

The results were striking: The rolling short-
term GRAT strategy beat the 10-year term
GRAT in every period, and succeeded 100% of
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.
Not only was the rate of success higher, but the
amount of wealth transferred was much greater.
Even when the 10-year term GRATs succeeded,
the rolling GRAT strategy transferred nearly
twice as much wealth: a median transfer of
$11.0 million compared with $6.1 million.
We also compared shorter-term GRATs,
because a 10-year term might not fit every
individual’s needs. The results were compa-
rable: Running the same historical analysis
using four-year term GRATs versus four years
of rolling two-year GRATs, the rolling strategy
succeeded 98% of the time, compared with
79% for the term GRATs. And out of 684
trials, the four-year GRAT transferred more
wealth than the rolling strategy only 18 times,
or in just 2.6% of the trials.
An All-Market Strategy
What about the stock market environment?
One might argue that if the stocks in a GRAT
were poised for an upsurge (even though
no one can predict future performance), a
All strategies are funded with $10 million and invested in a portfolio
representative of the S&P 500. Wealth to beneficiaries is reinvested and
adjusted for inflation. See Notes on Wealth Forecasting Analysis, page 34.
Term GRATs assume 20% increasing annuities, while rolling GRATs
assume constant annuities.
Source: AllianceBernstein
long-term GRAT might outperform rolling
short-term GRATs. But as Display 7 shows,
rolling GRATs outperformed term GRATs
by wide margins, no matter what the stock
market had done in the year prior to the
strategies’ inceptions. This reflects the fact that
historically, stocks have tended to rise over any
10-year period, with market downturns being
relatively short in duration.
In summary, our research provided an over-
whelming case against trying to “time”
your wealth transfer strategy with GRATs.
Regardless of interest rates or the current
market environment, a strategy of rolling short-
term GRATs funded with publicly traded stocks
appears very likely to provide better results than
a single, longer-term GRAT.
When Timing Does Pay Off
A few wealth transfer strategies will benefit
from low interest rates. For example, our
research showed that charitable lead annuity
trusts (CLATs) should perform better when
launched in low interest rate environments.
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. ■

14 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

Using Research to Navigate a Dynamic
Bond Market
by Guy Davidson, Director—Municipal Investments
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.
IN BERNSTEIN’S VIEW, BONDS ARE AN
essential complement to stocks in an inves-
tor’s overall portfolio. They provide income,
and they stabilize the total portfolio against
stocks’ volatility. In addition, a bond portfolio
can generate incremental return when actively
managed using a sound research framework.
The investing landscape of the past 12 months
provides the backdrop for a compelling narra-
tive of how active bond management can add
value across all three strategic imperatives.
Setting the Stage: The Hunt for Yield
For several years prior to the summer of 2007,
the stock and bond markets were unusually
calm. With risk seemingly absent, many
municipal bond managers were grabbing yield
anywhere they could find it, purchasing both
very long-term and lower-rated bonds. As a
result, the shape of the municipal yield curve
had become flatter than it had been in decades:
With so many eager buyers, long bonds offered
little additional yield versus short-term bonds of
the same credit quality. Plus, the additional yield
for riskier bonds fell to historically low levels.
But the reign of calm came to an abrupt end in
July of last year, as the effects of the subprime
crisis paralyzed the global markets.
Display 1
The shift from great calm to extreme fear caused the
municipal yield curve to steepen dramatically
AAA Municipal Yield Curves
5.5%
March 31, 2008
4.5
3.5 March 31, 2007
2.5
1.5
1-Year 6-Year 20-Year 30-Year
Estimates for AAA-rated insured municipal bonds
Source: Municipal Market Data Corp. and AllianceBernstein
Municipal bonds were not immune to the
contagion: Buyers rushed to purchase safer
(that is, higher credit-quality and shorter
maturity) bonds, causing prices to rise in
response to demand. And in its efforts to quell
the subprime crisis, the US Federal Reserve
aggressively cut interest rates several times,
helping to further lift bond prices. As prices
and yields are inversely correlated, by the
end of March 2008, the yield curve had gone
from flat to steep: Yields on short-term bonds
had fallen, while yields on riskier longer-term
bonds had risen (Display 1). Managers holding
SUMMER 2008 | 15
the lower-rated or longer-term bonds saw their
prices plummet. The dramatic change in the
yield curve and the increased demand for high-
quality bonds reflected the switch in investor
sentiment from greed to fear.
The Calm Before the Storm
Back in the spring of 2007, it was clear that
market risk was absent from many investors’
minds—they weren’t demanding much more
yield for holding long bonds. We determined
that the modest additional yield being offered
by long-term bonds was not sufficient compen-
sation for the mounting risk that these bonds’
prices could fall; when income and price
return were both considered, our calculations
indicated our investors would be best served
if we concentrated holdings in the intermedi-
ate-maturity range, resulting in a four-year
duration overall (Display 2).
Bonds in Brief
Bonds can present a complex landscape for inves-
tors, as their returns are influenced by so many
factors. Here, we recap the key drivers that affect
bonds’ performance.
Interest rates: When interest rates rise, the prices
of bonds that are outstanding fall, as the income
they generate will be less than that from a new
bond issued at the higher interest rate. The bond’s
price needs to fall as a way of compensating
buyers for its lower income versus the new bond.
Duration: A measure of how much a bond’s price
will change for every 1% change in the interest
rate. Expressed in years and mathematically
computed, the longer a bond’s duration, the greater
its price change will be when interest rates change.
Yield: There are several measures of “yield” in
bonds. We’ll focus for simplicity on “yield-to-
maturity”: This is the internal rate of return of
16 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
Display 2
In early 2007 we concentrated the maturity structure of
our portfolios
Our Maturity Selection
5%
88% 2%
4 10%
3
2 Duration Target
(4.1 Years)
1
1-Year 6-Year 20-Year 25-Year
As of March 31, 2007
Source: Municipal Market Data Corp. and AllianceBernstein
Our analysis of credit quality revealed a
similar story. By March 2007, credit spreads
had become extremely compressed, meaning
the premium being offered to take on the extra
the bond investment, assuming the bond is held
until maturity and that income is reinvested at a
constant rate.
Credit quality: Third-party agencies rate the
creditworthiness of each bond; bonds with higher
credit quality are deemed safer in terms of the
issuer’s ability to make its interest payments and
return the full value of the bond to the buyer upon
the bond’s maturity. Bond prices move up or down
in line with credit-rating changes up or down.
Maturity: The length of time until a bond matures
and the issuer pays back the face value of the bond.
Long maturity bonds, typically those with maturi-
ties of 10 or more years, are riskier than bonds
with lesser maturities in that their price moves are
greater, owing to interest rate changes or credit
downgrades. As compensation for this added risk,
long bonds typically offer higher yields than bonds
with shorter maturities. ■
risk of lower-quality BBB bonds versus the
highest-quality AAA bonds was almost non-
existent. In fact, an increase of only 0.08 in
the yield of BBB bonds translated into a drop
in prices that more than offset the benefit of
higher income over one year.
With virtually no incentive to move down in
credit quality, we chose to hold primarily very
high-credit-quality bonds in our portfolio. The
only BBB bonds we bought were those that
our credit team determined were likely to be
upgraded by the rating agencies or were stable
and scheduled to mature soon.
We didn’t make these portfolio decisions in
response to a premonition about the subprime
crisis: We made them based on our ongoing
research and disciplined approach to invest-
ing. An intensive assessment of the after-tax
total expected return (including yield and
likely price movements) versus the risks led to
our deep conviction about how to design the
portfolio to ensure it played its three roles well:
income, stability, and incremental return.
The Market Awakens to Risk
In July 2007, the scope of the subprime
mortgage crisis began to sink in. As investors
became increasingly aware of the potential
risks, both the yield curve and credit spreads
began moving. The markets convulsed as
nervous investors sold risky assets and bought
the safest investments they could find. US
Treasuries were the prime beneficiary of inves-
tors’ flight to safety, especially short-term
bonds, whose yields fell significantly more
than those of long bonds.
Key Concepts
> Fallout from the subprime crisis caused
significant dislocations in the municipal bond
market over the last year
> As risk entered the market, the shape of the
yield curve changed dramatically—moving from
flat to normal
> Thanks to the high average credit quality of
our portfolios’ holdings and their concentration
around a four-year duration, our portfolios
outperformed
> As the market shifted, we took advantage of
opportunities to increase yield by adding high-
yielding auction-rate securities and lower-rated
but solid bonds
In the first quarter of 2008, a new series of
shocks emanating from the subprime crisis
hit the municipal bond market head-on. Bond
insurers had long been a part of the munici-
pal bond market landscape. By guaranteeing
timely interest and principal payments from a
bond whose underlying credit rating may have
been only A or BBB, they assured the bond’s
higher AAA rating. For this insurance to
have value, the investors counted on insurers
remaining AAA rated. But insurers ran into
problems when they expanded beyond the
municipal bond market to subprime debt and
that debt went sour. Suddenly the insurers
were being downgraded—along with many
of the bonds they’d insured (Display 3,
following page).
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.

18 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

Consider the municipal auction-rate market.
Historically, these securities have been viewed
as near-cash equivalents with interest rates
that reset at some predetermined frequency,
usually seven, 28, or 35 days, when an auction
is held. Investors in these securities place bids
indicating how much they want to buy and the
yield they require. Once all the bids are in, the
auction agent fills the lowest-rate bids first, then
the next highest, and so on, until all the bids
are filled. The rate at which the last bid is filled
is called the clearing rate, and all the bidders
receive that rate until the next auction. If there
are insufficient bids for all the securities up for
auction, the auction “fails” and the holders
retain all the bonds at a maximum interest rate
specified in each bond’s indenture—the written
agreement between a bond issuer and bond-
holders—until the next auction. Each bond
indenture is different, and the maximum rate is
based either on a formula typically tied to some
short-term index (which results in a relatively
low maximum rate) or on a specific fixed rate,
which is usually quite high, 12% or 15%.
Capital-constrained investment banks aban-
doned the municipal auction-rate market in
mid-February of 2008, causing these auctions
to fail at record levels. As a result, tax-exempt
yields on these securities were as high as 15%.
We carefully evaluated the underlying credit
quality of each issuer and the terms of each
issue’s auction, seeking to determine whether
the yields were a function of true risk or of
undue risk aversion.
Based on our analysis, we determined that
those securities with high fixed maximum rates
were attractive, while those with formula-based
rates were not. When many of the auction-rate
securities we first bought failed, they did so
at very high rates, such as 15%. And, because
their rates could jump to 15%, these securities
attracted new investors, significantly reducing
the risk of future failed auctions. Thus, we took
advantage of the risk aversion of money market
investors to add high-yielding auction-rate secu-
rities to our portfolios.
A second opportunity was in bonds with lower
credit quality. Having avoided the negative
impact of the downgrades in bond quality
on our portfolios, we worked to exploit the
downgrades to our advantage. With spreads on
investment-grade municipals having widened
sharply since July 2007, significant extra
yield was available for going down in credit
quality from AAA to BBB. Plus, the number
of bonds being insured had fallen from 50%
of new issuance over the last few years to only
25%—so the supply of lower-rated bonds on
the market was greater. Many of these bonds
were quite solid, and we took advantage of
this singular opportunity to earn extra yield
in our portfolios. A comparison of our buys
in the first quarter of 2008 versus 2007 shows
a significant increase—more than double—in
bonds with credit ratings of less than AAA
(Display 5).
Display 5
We found opportunities in downgraded bonds
Ratings Distributions of Buys
2007 Q1 2008 Q1
BBB & Other
A 6%
3% BBB & Other
A 5%
AA 16%
AA
19% 46%
AAA AAA
72% 33%
Source: AllianceBernstein
SUMMER 2008 | 19
 The Power of Active Bond Management
This period of extreme volatility in the usually
earthbound municipal bond market provides
a useful lens for examining how our research
and the application of our long-term invest-
ing paradigm work to protect and build client
portfolios. In this fast-moving environment,
returns on our intermediate-duration munici-
pal bond portfolios were among the best in
their peer groups for the one-year period
ending March 31, 2008 (Display 6).
Display 6
Our active management delivered in a turbulent time
One-Year Returns*
Bernstein Municipals vs. Peer Groups†
5%
4.69%
4.31
4
3
2.60
2 1.95
1
0
Diversified California
Bernstein Lipper Average
Past performance is no guarantee of future results.
*Through March 31, 2008
†Bernstein Intermediate portfolios versus their Lipper peer group averages
Source: Lipper and AllianceBernstein
20 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
Bernstein commits substantial resources to
active bond management: tracking the con-
stantly shifting mix of risk and opportunity in
the marketplace; assessing and monitoring the
soundness of every bond we buy; and securing
the best prices for the bonds we buy and sell.
At the core of our operation is an established
Bernstein strength: rigorous research. Our goal
as active bond managers is to use research to
identify values in the marketplace created by
changes in price.
The advantages of active management are
clear: It took only six months to move from
one of the flattest yield curves in history to one
with a normal shape and for credit spreads to
move from an all-time low to more attractive
4.46 levels. Thus, our concentrated maturity struc-
ture and high-credit-quality stance paid off. By
2.83 optimizing our portfolios to avoid the pitfalls
of this environment while taking advantage of
its opportunities, we were able to create value
for our clients. ■
New York
Can Early Tax Harvesting Reap Greater Gains?
by Greg Singer, Director of Research—Wealth Management Group
and Vincent Childers, Research Analyst—Wealth Management Group

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?

WHEN ASKED WHAT THE STOCK MARKET Display 1

would do, J.P. Morgan famously asserted, “It
will fluctuate.” The same is true of tax rates,
and, as with the markets, their fluctuations can
have a sizable impact on individual wealth.
Tax rates and gains realizations: A contrary
relationship
Realized Capital Gains as Percent of GDP vs.
Tax Rate on Long-Term Capital Gains
1960–2011*

(%) Tax Reform Act (%)

Timing the Taxman 8 Capital Gains of 1986 45
Throughout history, both the level of the capital Tax Rate
Capital Gains Realizations

7 40

Capital Gains Tax Rate

gains tax rate and expected increases in rates 6 35
have influenced investor behavior. Specifically, 5
30

investors tend to defer more gains during 25

4
20
periods of higher taxes and become more 3
15
willing to harvest gains in low-rate regimes. 2
Capital Gains 10
Investors have also been tempted to realize 1 Tax Reform Act
Realized, % of GDP 5
of 1969
embedded gains in anticipation of higher tax 0 0
rates, reflecting their desire to get a jump on any 1960 1970 1980 1990 2000 2010

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

increase of eight percentage points (from 20%

to 28%) on January 1, 1987, as a result of the
Tax Reform Act of 1986 (Display 1).
The increase in capital gains rates in 1987 must
have seemed like a locomotive coming down the
line: Many investors simply leapt off the tracks
to avoid it. Under current tax law, the long-term
capital gains tax rate increases five percentage
points, from 15% to 20%, in January 2011.
But there’s widespread belief that the rate
increase could be higher and that it could come
as early as mid-2009, after the next president
takes office and a new Congress convenes. The
question this raises is: Should investors acceler-
ate their harvesting of embedded capital gains
before higher rates become a reality?

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
At first blush, realizing taxable gains in the
face of higher future tax rates may seem like
the obvious thing to do. But trying to sidestep
any presumed or even scheduled shift in tax
rates before a comprehensive assessment of all
the investment variables involved could prove
both untimely and imprudent. After all, the
benefit of deferring taxes is well established—
when we defer taking gains, money that would
otherwise have gone to the government instead
can continue to grow, leading to greater future
wealth. And, of course, the investment objec-
tive is to maximize the likelihood of better
after-tax returns, not simply to minimize taxes.
To complicate matters further, there’s another
important (and often overlooked) factor in
examining the value of skirting any prospec-
tive rate increase: the very real possibility that
tax rates at the end of an investor’s expected
holding period may be very different from the
rate expected in the near term. In our 1986
example, rates fell a decade later by as much as
they rose, and then dropped another five per-
centage points a few years after that. In fact,
rates have varied considerably over the years:
The maximum tax rate on long-term capital
gains has ranged from a high of almost 40% in
1978 to the current, historic low of 15%.
Given that tax rates have fluctuated and likely
will continue to do so over the coming years,
there may be real danger in focusing myopi-
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
gains rates have, on occasion, been enforced
retroactively, so that almost any strategy of
anticipating a change may have an outcome
other than the one expected.
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.
22 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
Finally, if your expected holding period is
forever (or at least until death parts you from
your investment, assuming the law still allows
a step-up in basis), the question whether to
harvest or defer is moot—capital gains taxes
are not an issue. But the situation is more com-
plicated if you anticipate selling the position
in the future, either as part of your natural
turnover or to raise cash for a particular need.
To help investors take a more clearheaded,
objective view of this difficult and emotional
topic, let’s explore the drivers that determine
whether or not accelerating gains ahead of an
anticipated tax increase makes sense.
To Harvest Early or Defer Gains: A Framework
At the most basic level, the decision to tax-trade
ahead of an expected increase in rates hinges on
the interplay of two key variables: the inves-
tor’s expected horizon, or holding period before
liquidating the portfolio, and the expected
increase in tax rates (Display 2).1 Generally
speaking, the shorter the investment horizon,
the more valuable it is to harvest gains in the
lower-rate regime—the benefits of deferral are
offset by larger future taxes paid under the
higher-rate regime. And, of course, the higher
the future tax rate, the more incentive there is to
capitalize on the current lower rate.
Display 2
Harvest or defer? The decision hinges on two
key factors
Defer Harvest Early
Longer Investment Horizon Shorter
Lower Future Tax Rate Higher
Source: AllianceBernstein
To bring this framework to life and illus- Key Concepts
trate the trade-offs at a very simplistic level, > Shifting capital gains tax rates have influenced
imagine an investor, Earl E. Gaines, who has
investor behavior in the past
a $1 million portfolio of diversified equities
with a cost basis of $500,000. Let’s say he > Capital gains taxes are scheduled to rise from
plans on selling the portfolio a year from now, 15% to 20% in 2011, but many think that the
and expects it to earn 8% during that period. increase could be even greater, and come sooner
Convinced that rates will increase in the fol-
lowing year from 15% to 20%, Earl wants to > For holders of diversified equity portfolios, an
look at two scenarios. In the fi rst, he sells his increase in rates to 20% would not present a
entire portfolio now, pays taxes at the current particularly strong case to actively tax-trade
15% capital gains rate, reinvests the proceeds ahead of the policy change; but an increase in
in an identical equity portfolio, and then rates to 28% may suggest a clearer rationale to
cashes that out a year hence, paying tax again,
take action
but now at the prevailing 20% long-term
capital gains rate. In the second scenario, Earl > For investors holding concentrated stock
simply holds on to his current portfolio and positions, any rise in capital gains rates increases
liquidates the entirety a year later, paying all of the appeal of harvesting embedded gains in a
the capital gains taxes once, at the going 20% lower-rate regime and diversifying the proceeds
rate. Display 3 shows how the two portfolios
into a lower-risk portfolio
fare over that period, the capital gains taxes
they incurred, and their final wealth values.
Display 3 The “deferral” strategy shows higher pretax
The harvest or defer trade-off: growth (ending at $1,080,000), as the original
Greater growth or lower taxes? sum, undiminished by any up-front tax hit,
continues to compound for an additional
Harvest Strategy Deferral Strategy
year. But it also has the highest tax bill: The
Initial Value $1 Mil. $1 Mil.
larger total gain (of $580,000) was all taxed
Cost Basis $500,000 $500,000
at the higher 20% rate. The “harvest” strategy
Tax Rate 15% N/A
saves on taxes, but reinvests less princi-
Tax on Gain $75,000
pal: It ends the year with a pretax value of
Proceeds $925,000
$999,000. However, the initial embedded gain
Tax Rate Next
Year 20% 20% (of $500,000) was taxed at the lower 15%
Value Next Year $999,000 $1.08 Mil. rate, while only the following year’s gain (of
Gain Next Year $74,000 $80,000 $74,000) was taxed at the higher 20% rate, for
Tax on Next a combined capital gains tax of 16%. In this
Year’s Gain $14,800 $116,000* case, harvesting early (at the end of one year)
Combined provides an advantage of about 2% in overall
Tax Rate 16% 20%
wealth versus the deferral scenario.
Total Taxes $89,800 $116,000
Liquidation
Proceeds $984,200 $964,000
Advantage 2.1%

*Includes tax on the embedded gain of $500,000

Source: AllianceBernstein

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 (%)

examine what we call the “break-even” horizon, 7%

28% Future Tax Rate
the expected holding period that would result 6

in the same wealth outcome whether one defers 5 Sell and

4 Repurchase
the tax hit for some time or realizes gains, Strategy Results
pays tax at today’s rate, and then reinvests the 3 in Higher
2 20% Future Tax Rate Average Wealth
proceeds before realizing a gain at the end of
1 Break-Even
the investment horizon. Liquidation
0 Horizon
(1) Hold Strategy
Display 4 shows the break-even horizon for Results in Higher
(2) Average Wealth
the same portfolio calculated with two differ-
ent future tax scenarios in mind: a moderate 1 3 5 7 9 11 13 15
Liquidation Horizon (Yrs.)
increase in the capital gains tax rate from 15%
to 20% in one analysis, and a more aggressive 28% Rate 20% Rate
move from 15% to 28% in the other. The area
The initial position is a $1 million diversified equity portfolio (35% US
below the zero line indicates those outcomes value/35% US growth/25% developed international/5% emerging
markets) with an existing cost basis of $500,000. The current tax rate on
where the deferral strategy is optimal; above capital gains is 15% and either 20% or 28% in all future years, while the
the line the harvest strategy wins out. You can ordinary income tax rate is 39.6% in all years. The analysis reflects post-
liquidation proceeds and 15.25% annual gains realizations and assumes
see that at a 20% future capital gains rate, the future capital losses are used to offset outside gains. See Notes on Wealth
Forecasting Analysis, page 34.
advantage of harvesting versus deferring over Source: AllianceBernstein
a one-year horizon is Earl’s 2%, as detailed
in the example above. But that modest advan- early is more than 6% at year one, and while
tage declines over time and turns negative at that advantage too declines over time, there is
about year seven, as the benefits of tax deferral no liquidation horizon within which the port-
win out. folio’s growth can overcome the drag of the
higher future tax.2
For investors with a liquidation horizon of less
than seven years, there is some benefit to har-
vesting early, though it is modest. The average From the Possible to the Probable
increase in wealth over this horizon is less than But the analysis is still too simplistic, as
1%. For investors with longer holding periods, we’ve considered only a scenario in which
deferring appears to make sense. the future return was assured. We haven’t yet
taken account of the uncertainty surrounding
However, when we look at a potential increase the level and path of future returns, and the
in rates from 15% to 28%, the story changes success of early harvesting depends critically
dramatically: Here the benefit of harvesting on the return environment over the course of
2
The break-even horizon for a diversified portfolio actually depends not only on portfolio performance and the liquidation horizon
but also on what percentage of the portfolio’s embedded capital gains are realized annually—what we refer to as the gains realiza-
tion rate. The gains realization rate is often proxied by the more common metric of portfolio turnover—although, strictly speaking,
efficiently tax-managed portfolios can endure higher turnover without excessive gains realizations. High rates of gains realization,
which typically characterize a less tax-efficient portfolio, will extend the break-even horizon, and in extreme cases can overcome any
advantage to tax deferral over any time horizon. For example, an investor with a gains realization rate approaching 100% (he pays
taxes on all of his gains every year) leaves no opportunity for the compounding benefit of tax deferral to work to his advantage, and
as such is almost always better off realizing embedded gains at today’s lower rate than at “tomorrow’s” higher rate.

24 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

the holding period. If future markets are better
than average, then tax deferral has higher
value. Should future markets be disappointing,
then harvesting early becomes more attractive.
Of course, we cannot divine the future. But to
help our clients make thoughtful decisions
based on a realistic assessment of potential
future outcomes, we’ve developed a quan-
titative capital markets model. Our Wealth
Forecasting System3 takes the known facts—
our clients’ assets, income needs, risk toler-
ance, tax brackets, and time horizon—and
runs various investment scenarios through our
model to project 10,000 possible return paths.
With it, we can analyze the impact of changes
in tax rates across the spectrum of possible
future capital market outcomes, thus helping
clients make informed decisions regarding the
probability that accelerating gains realizations
will result in greater wealth.
In Display 5 you can see that should the
capital gains rate stay constant at 15%, accel-
erating gains will typically be a poor decision,
although there still would be scenarios—that
is, when markets are particularly hostile—
in which the client is better off. But if rates
increase to 20%, the appeal of harvesting
embedded gains early rises, and the likelihood
of being better off remains above 50% for
investors with a horizon of up to seven years
(our break-even holding period).
But is a 50/50 shot at success enough entice-
ment to take gains early? If an investor with
a five-year horizon wants to be 75% certain
that realizing gains early is in his best interest,
he will not want to take action if rates rise to
20% (given there is only a 60% probability
of success). However, the same investor can
be very comfortable he is making the right
decision if rates are scheduled to rise to 28%,
in which case the probability of success is high
3
See Notes on Wealth Forecasting Analysis, page 34.
Display 5
Gauging the likelihood of a successful early harvest
Probability That Harvesting Early Delivers
Higher Post-Liquidation Wealth (%)
100%
75
50
25
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Liquidation Horizon (Yrs.)
28% Tax Rate 20% Tax Rate 15% Tax Rate
The initial position is a $1 million diversified equity portfolio (35% US
value/35% US growth/25% developed international/5% emerging
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
the ordinary income tax rate is 39.6% in all years. The analysis reflects
post-liquidation proceeds and 15.25% annual gains realizations and
assumes future capital losses are used to offset outside gains. See Notes
on Wealth Forecasting Analysis, page 34.
Source: AllianceBernstein
across all investment horizons, and the mag-
nitude of success, in terms of greater after-tax
wealth, is higher too.
But considering the inherent uncertainty as to
what tax rates will be at the end of the pro-
spective holding period (itself not known with
certainty), as well as whether tax rates will
in fact be higher next year (who knows what
Congress will approve?), one should be cautious
when contemplating harvesting gains as part of
a strategy of “front-running” the rate increase.
The Perils of the Concentrated Position
There is one situation, however, in which a
prospective shift in capital gains tax rates has
powerful relevance to the “harvest or defer”
decision: concentrated positions of low-cost-basis
stock. In this case, in addition to taking advan-
tage of a lower tax rate on your sale, you also
reinvest the proceeds in a lower-risk portfolio.
SUMMER 2008 | 25
While the allure of building massive wealth by
concentrating your portfolio in one or several
holdings is great, the likelihood of achieving
returns superior to the markets’ is low, and
the volatility one can expect to encounter is
high. This higher volatility has a real cost. Our
research has shown that historically, a typical
single stock has compounded at a growth rate
almost three percentage points below that of
the S&P 500. And for single stocks with the
highest volatility, the results were even worse,
compounding at a rate of only half that of the
market.4 In fact, if we look at a similar liqui-
dation analysis but without any change in tax
rates, simply exchanging a single stock position
for a diversified portfolio of global equities, the
median wealth outcome is superior over every
time period (Display 6).
Display 6
Exchanging a single stock for a diversified portfolio
Median Wealth Advantage
Diversified Stock Portfolio vs. Single Stock (%)
25%
20
15
10
5 0
0 1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Liquidation Horizon (Yrs.)
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-
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
35% US value/35% US growth/25% developed international/5% emerg-
ing markets, and the portfolio’s results reflect post-liquidation proceeds
and 15.25% annual gains realizations; the analysis assumes future capital
losses are used to offset outside gains. See Notes on Wealth Forecasting
Analysis, page 34.
Source: AllianceBernstein
4
See our comprehensive research study on single stock positions, The Enviable Dilemma: Concentrated Stock—Hold, Sell, or Hedge?
26 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
When we add the future capital gains rate
increases into the mix, the “harvest or defer”
decision is easier: In all of our prospective tax
regimes (a static 15% rate, as well as increases
to 20% and 28%), the benefits of harvesting
and diversifying the concentrated position, in
the median case, appear significant: With an
increase in rates to 28%, the magnitude of the
benefit of harvesting early amounts to about
18% in the first year (Display 7). And, unlike
a diversified portfolio, for which the advantage
of front-running the tax increase declines over
time, with a single stock portfolio the advan-
tage of selling and diversifying never sets on
this particular horizon, rising to over 30% in
year 15.
Display 7
A magnitude of difference: Harvesting a single stock
versus diversified portfolios
Median Increase in Post-Liquidation Wealth
Sell vs. Hold (%)
30%
20
10
2 3 4 5 6 7 8 9 10 11 12 13 14 15
Liquidation Horizon (Yrs.)
28% Tax Rate 20% Tax Rate 15% Tax Rate
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-
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
35% US value/35% US growth/25% developed international/5% emerg-
ing markets, and the portfolio’s results reflect post-liquidation proceeds
and 15.25% annual gains realizations; the analysis assumes future capital
losses are used to offset outside gains. See Notes on Wealth Forecasting
Analysis, page 34.
Source: AllianceBernstein
But once again, looking only at scenarios
in which the future returns are known with
certainty is not enough. We must consider
the full spectrum of portfolio outcomes to
judge the likelihood of any strategy’s success.
When harvesting gains and diversifying the
single stock position, we found that the prob-
ability of success remains above 50% across
the entire 15-year time frame of our analysis,
and does so in both of our prospective rate
regimes (Display 8). Yet even here, of course,
the questions of how much to sell—and how
quickly—are matters that must be tailored to
each investor’s risk tolerance, spending needs,
and overall financial profile.
Decision Time?
The moral of the story: Tax rates change over
time—and they change often and fairly unpre-
dictably. For investors whose expected holding
period is not forever, it’s a difficult decision
whether to forgo tax deferral in favor of locking
in today’s low tax rate. For holders of sub-
stantial single stock positions, the appeal of
harvesting gains on those stocks now and diver-
sifying their portfolio is high, in terms of both
the magnitude and the probability of success,
regardless of the level of any future tax rate. But
for investors who already hold diversified port-
folios, the decision is less clear: It will depend
both on the investor’s time horizon and the
size of the anticipated change in tax rates. At a
future capital gains rate of 20%, the magnitude
of the benefit from harvesting early is modest,
and the probability of success declines quickly
Display 8
Positively probable: Tax-trading out of a single stock
Probability That a Diversified Portfolio Will Deliver
Higher Post-Liquidation Wealth (%)
80%
60
40
20
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Liquidation Horizon (Yrs.)
28% Tax Rate 20% Tax Rate
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-
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
35% US value/35% US growth/25% developed international/5% emerg-
ing markets, and the portfolio’s results reflect post-liquidation proceeds
and 15.25% annual gains realizations; the analysis assumes future capital
losses are used to offset outside gains. See Notes on Wealth Forecasting
Analysis, page 34.
Source: AllianceBernstein
with time. But in the event that a 28% rate
looms, there is a higher probability of success
in accelerating gains, and a superior average
wealth outcome, and this may bias some toward
taking preemptive action.
Regardless of the future tax regime, our
Wealth Forecasting System can help clients
quantify the likelihood and magnitude of
success of different investment strategies
so that they can make the most informed
decision. ■
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
rapid industrialization has had a profound
impact on world trade. In its drive to supply
the world, China has helped push down the
price of finished goods, while its enormous
appetite for raw materials has helped push
up and sustain the prices of energy and com-
modities. This double threat—economists
call it “downstream deflation” combined
with “upstream inflation”—has squeezed the
operating margins of foreign manufacturers,
forcing many of them to either relocate or out-
source production to China in order to remain
competitive. It has also helped build profits for
corporations and others that have invested in
China. Will the next 10 years be anything like
the past 10 years? Is China likely to continue
its current economic strategy?
When It Started: Deng Xiaoping’s Economic Reforms
When Deng Xiaoping launched China’s Open
Door Policy in 1979, his aim was simple: to
raise China’s standard of living in the after-
math of the Cultural Revolution by making
the economy more attractive to foreign inves-
tors. Deng probably never dreamt that in little
more than two decades it would help make his
country the factory to the world.
Foreign companies saw Deng’s economic
reforms as an opportunity to profit from a
consumer market that was not only totally
28 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
new to them, but also huge—China’s popula-
tion crossed the one billion threshold during
the 1980s—and growing fast. China appeared
to be an El Dorado, a golden opportunity
for massively expanding their global sales.
According to popular lore, shoe manufacturers
aimed to sell an additional two billion
shoes simply by selling one pair to each
person in China.
In reality, the policy played out very differ-
ently from what anyone had anticipated. While
Deng’s reforms opened the economy to foreign
capital, they also liberated China’s homegrown
capitalist talent. Chinese-made home appli-
ances and other electronic products—likely
to be followed in the not-too-distant future
by Chinese cars—are building market share
not only in the developed world but, increas-
ingly, in developing markets as well. And today
China’s textile and garment exports are the
targets of a protectionist backlash within the
US and European markets.
Statistics illustrate the scale of China’s impact.
In 1997, Chinese finished-goods exports
averaged about US$10 billion a month. In
2007, the figure was close to US$80 billion—
an eightfold increase. Imports of raw materials
have grown nearly sevenfold during the same
period, from US$3 billion a month to US$20
billion (Display 1).
Display 1
Key Concepts
Exports of finished goods and imports of commodities
have risen together > China is experiencing a massive population
migration from rural areas to urban centers with
China’s Foreign Trade: 1995–2007
80
higher-paying jobs
Finished-Goods > China will continue to push down the prices of
60 Exports
(US$ Billions)

finished goods while sustaining the high prices of

40 energy and commodities
20 > Despite the threat of a protectionist backlash
Raw-Materials from the rest of the world, China is likely to
0 Imports
95 96 97 98 99 00 01 02 03 04 05 06 07 08
continue to expand its industrial capacity
Source: CEIC Data and AllianceBernstein > We expect the decline in world prices and
manufacturers’ operating margins to persist for
High Consumption, Low Prices
another decade at least
For everyone—from the consumers and com-
modity producers benefiting from China’s
emergence to the foreign manufacturing fi rms
The Long March to Urbanization
without access to its production base—the
After thousands of years as a predominantly
critical question is: How sustainable is China’s
agrarian society, China is rapidly becoming
business model? In our view, China’s economic
more urbanized. The combination of low
fundamentals are such that, in the absence
agrarian wages and the lure of better-paying
of external pressure to change, its current
jobs in the cities is causing a massive redistri-
business model will be maintained.
bution of China’s population of 1.3 billion.
From 1996 to 2006, 122 million rural
China’s model, in essence, consists of high
inhabitants—an average of 11 million a year—
energy and commodity consumption and low-
moved to the cities. Indeed, no other country
priced goods. For example, today the country
in history has witnessed migration from the
consumes about 25% of global commodi-
country to the city on such a scale.
ties (such as steel, copper, and iron ore), but
accounts for only about 6% of global nominal
But the rural population still accounts for 55%
GDP, as shown in Display 2, following page. In
of the total—a far higher proportion than in
the absence of exogenous shocks, there appear
such neighboring countries as South Korea,
to be few incentives for China to consume
Taiwan, and Japan, where less than 10% live
energy and commodities more efficiently and
on the land. This relative imbalance is unlikely
increase production value-added more quickly.
to change soon: Even if the country’s rural
If China continues on this path, its impact on
exodus were to accelerate to 15 million a year,
the world economy could become even more
40% of the people would remain on the land
dramatic over the next 10 to 20 years.
in 2020 (Display 3, following page).
Indeed, China has powerful incentives to
The urban migration has led to great income
maintain its current model, as we shall see.
disparities between farm and city families.
The main risk, in our view, is a potential
The average household income in the indus-
economic backlash that could even escalate
trialized east is three times that of the more
to a trade war between China and major
rural west—and the gap is widening, further
Western economies.

SUMMER 2008 | 29
Display 2

China consumes roughly one-fourth of global

commodities… …yet produces only 6% of global GDP

Share of Global Consumption Share of World Nominal GDP

40% 5.9%
5.5%
35
5.1%
30 4.7%
4.4% 4.5%
4.2%
25 3.8%
20
15
10
5
0
Aluminum Copper Zinc Iron Crude Nickel Stainless 2000 2001 2002 2003 2004 2005 2006 2007F
Ore Steel Steel
2001 2004 2006E
Source: Deutsche Bank Global Commodity Research, International Monetary Fund (IMF), and AllianceBernstein

Display 3

Although rural migration is unprecedented… . …40% of the population will still be rural in 12 years

Net Change in Rural Population Share of Total Population

15 100
90
10 80
70 Rural Population
5
Million Persons

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

Source: CEIC Data and AllianceBernstein

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

30 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

during the 1990s, the investment boom has
eroded them. According to our estimates, China
brought up the rear in terms of world energy
efficiency in 2006, being 3.5 times less efficient
than the US. Even if it managed to improve
efficiency dramatically, it would still rank near
the bottom.
Display 4
China consumes almost as much energy as the
European Union
Percent Share of World Consumption
30%
US
25
20
European Union
15
10
Japan
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
Source: BP Statistical Review of World Energy 2006, CEIC Data,
and AllianceBernstein
Corporate Strategy: Expansion, Not Profit, Is the Key
The sharp increase in investment and indus-
trial growth in China highlights the country’s
unique perspective on corporate profitability:
Growing the business and increasing market
share are the highest priorities; profits come
second. This stands in marked contrast to
a market economy, where fiscal prudence
eventually dictates that business lending to
companies with low profits be scaled back.
However, China’s business model is unlikely to
change for one simple reason: Its future under
that model is promising. China’s potential for
industrial development is enormous. In fact,
the recent surge in wages and land prices in a
few highly developed eastern regions is sending
just the right market signals to Chinese fac-
tories: to maintain their competitive edge by
starting to relocate to lower-cost regions,
where the labor supply is abundant and the
vast rural countryside underdeveloped.
Why Banks Won’t Pull the Plug
Thus far, China’s government-controlled banks
have supported the expansion of low-profit
companies, with no apparent attempt to scale
back expansion in favor of profits. Beyond a
certain threshold, of course, this low-margin
bubble may burst. But a fully market-oriented
banking system has yet to be developed in
China, and Chinese state banks—which
account for about 70% of banking system
assets—remain heavily influenced by govern-
ment directives.
If the bank loan market were to run into
trouble, China’s leadership would face a tough
choice between righting banks’ balance sheets
China
and preserving or creating jobs. On the whole,
we think that job creation and social stability
06 by an injection of public funds. The pressure to
create 69 million new jobs in the next 10 years
will remain intense, and the incentive to reduce
production capacity in labor-intensive indus-
tries will be small.
Policy Choices and the Protectionist Backlash
The main threat to China’s ability to maintain
its high-growth, low-margin business model
is the potential for a protectionist backlash
by other countries. The US and other
Organisation for Economic Co-operation and
Development countries have seen their jobs
and manufacturing plants move to China—
and they’ve seen their domestic markets
flooded with cheaper-priced Chinese goods.
As China’s industrial revolution continues, this
situation will likely become more pronounced.
If China maintains its current policy, it risks a
trade war, a deceleration of the globalization
process, and a decline in world trade, all of
which could pose a serious threat to China’s
export-led economy.
SUMMER 2008 | 31
To confront this threat, China could decide
to bow to external pressure and “revalue”
its currency—the renminbi—more quickly,
strengthening it versus other currencies. In
theory, if this were to happen rapidly, it would
increase China’s purchasing power overseas,
thereby encouraging domestic consumption,
decreasing a massive balance-of-trade surplus,
and boosting GDP. A stronger currency would
also encourage Chinese manufacturers to start
focusing on producing higher-quality, high-end
goods, and would help enlarge its growing
services sector.
Korea, and Taiwan, the correlation between
per capita GDP and consumption to GDP
tends to be high and positive—that is, the
more people earn, the more they tend to spend
(Display 5, top). The correlation between
consumption-to-GDP ratios and savings rates
tends to be inverse—that is, the more people
spend, the less they tend to save (Display 5,
bottom). The high consumption-to-GDP ratio
(70%) in the US, for example, is the result not
only of its superior income per capita, but also
its extremely low savings rate (about 13.5%).
Display 5

China’s consumption is constrained by its low

Obstacles to Revaluation per capita income…
Although speeding up the pace of currency Per Capita GDP vs. Consumption/GDP Ratio
revaluation and turning itself into a consump- 2006
50,000
tion-driven economy makes sense for China in US
theory, a number of huge obstacles remain— 40,000 UK
Per Capita GDP US$

making this course of action unlikely. While Japan Germany

China’s nominal GDP (estimated at US$3.3 30,000 Singapore France

trillion in 2007 prices) ranks among the Hong Kong

20,000 South Korea
world’s highest, the country’s per capita GDP
Taiwan
remains among the lowest—about US$2,500
10,000
at current prices. This is 17 times lower than China Thailand Indonesia Philippines
Malaysia
the per capita GDP of the US, 14 times lower 0 India Vietnam
than that of Germany and Japan, and six to 35 40 45 50 55 60 65 70 75
seven times lower than that of South Korea Consumption-to-GDP Ratios (%)
and Taiwan. Even if a currency revaluation …and exceptionally high savings rate
made the renminbi twice as valuable in relation
to the US dollar, the Chinese consumer’s Gross Savings Rates vs. Consumption/GDP Ratio
2006
purchasing power would still remain small.
Gross Domestic Savings Rates (% of GDP)

Further, not only are the Chinese low wage 55

Singapore
earners, they are low spenders. China’s con- 50
45 China Malaysia
sumption-to-GDP ratio, which was about 36%
40
in 2006, is low by international standards. South Korea
35 Hong Kong
Vietnam
30 India
Low income and high savings are seen in few Thailand
25 Japan Indonesia
other countries, and the fact that they coincide
20 France Taiwan Philippines
in China helps explain the country’s low
15 Germany UK US
consumption-to-GDP ratio. In industrialized
10
countries such as the US, Germany, France, 35 40 45 50 55 60 65 70 75
and Japan, as well as in certain advanced Consumption-to-GDP Ratios (%)
Asian economies like Hong Kong, South
Source: Asian Development Bank, CEIC Data, IMF, and AllianceBernstein

32 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT

In order to boost consumption, China needs to
both lower the savings rate and boost income.
The real challenge will be to lower the savings
rate while accepting the harsh reality that
raising the income-per-capita levels of China’s
huge population will take several years.
A stronger Chinese currency would also make
China’s goods more expensive in foreign
markets, which would pose a huge threat to the
country’s low-profit manufacturing businesses.
While the price of the energy and commodities
that these businesses import from abroad would
drop, that would not be enough to make up the
difference. On balance, we think it unlikely that
China will opt for aggressive revaluation of the
renminbi. Instead, we expect it to continue to
pursue its high energy/consumption, low-margin
business development model, even at the risk of
retaliatory action from other countries.
Our Outlook:
China Will Continue to Push Down Prices
China’s role as factory to the world is unlikely
to diminish—in fact, it will become more pro-
nounced as the country’s economy develops,
and its far-reaching effects on the world
economy will become more entrenched. China
will continue to push up energy prices—and
push down prices across a wider range of
manufactured goods. As a result of the model’s
success, however, a number of developments
will arise. For example, the push now being
seen across the industrial base toward the
manufacture of higher-margin goods is likely
to raise China’s per capita income, helping
to create a consumer class large enough for
multinationals to seek to target. This would be
an important driver for the expansion of the
services sector, although such growth would
probably be concentrated in the major cities,
not evenly distributed nationwide.
The development of the country’s interior will
increase. Some domestic Chinese companies
are already leading the production shift inland,
and a small but growing number of foreign
firms are following them. In our view, China’s
industrial sector will become more diverse
as a result: The developing hinterland will
become the focus of the country’s low-margin,
volume-oriented manufacturing base, allowing
production capabilities in the industrialized
east—where skilled labor and quality infra-
structure and logistics are readily available—to
be upgraded.
The transformation will resemble that of
Japan and Southeast Asia during the 1980s.
It will also have similar consequences. For
example, whereas China’s growth as a global
production center over the past two decades
has been achieved entirely through borrowed
technology, the emergence of higher-margin
businesses will serve as a catalyst for research
and development as well as for original
product design, and the quality and sophisti-
cation of the goods it exports will rise. China
is ramping up its production capacity, and we
expect the decline in world prices and manu-
facturers’ operating margins will persist for
another decade at least. ■
SUMMER 2008 | 33
Notes on Wealth Forecasting Analysis
The Bernstein Wealth Forecasting Analysis SM (WFA)
is designed to assist investors in making a range of key
decisions, including setting their long-term allocation of
fi nancial assets. The WFA consists of a four-step process:
(1) Client Profi le Input: the client’s asset allocation, income,
expenses, cash withdrawals, tax rate, risk-tolerance goals,
and other factors; (2) Client Scenarios: in effect, questions
the client would like our guidance on, which may touch
on issues such as which vehicles are best for intergenera-
tional and philanthropic giving, what his/her cash-flow
stream is likely to be, whether his/her portfolio can beat
infl ation long-term, when to retire, and how different
asset allocations might impact his/her long-term security;
(3) The Capital Markets Engine: our proprietary model,
which uses our research and historical data to create a vast
range of market returns, taking into account the linkages
within and among the capital markets (not Bernstein
portfolios), as well as their unpredictability; and (4) A
Probability Distribution of Outcomes: based on the assets
invested pursuant to the stated asset allocation, 90% of the
estimated returns and asset values the client could expect
to experience, represented within a range established by
the 5th and 95th percentiles of probability. However,
outcomes outside this range are expected to occur 10% of
the time; thus, the range does not establish the boundar-
ies for all outcomes. Further, we often focus on the 10th,
50th, and 90th percentiles to represent the upside, median,
and downside cases. Asset-class projections used in this
publication are derived from the following: US value stocks
34 | BERNSTEIN JOURNAL: PERSPECTIVES ON INVESTING AND WEALTH MANAGEMENT
are represented by the S&P/Barra Value Index, with an
assumed 20-year compounding rate of 8.2%, based on
simulations with capital market conditions as of December
31, 2007; US growth stocks by the S&P/Barra Growth
Index (compounding rate of 8.1%); developed international
stocks by the Morgan Stanley Capital International (MSCI)
EAFE Index of major markets in Europe, Australasia, and
the Far East, with countries weighted by market capitaliza-
tion and currency positions unhedged (compounding rate
of 8.0%); emerging markets stocks by the MSCI Emerging
Markets Index (compounding rate of 6.6%); taxable
bonds by diversified securities with seven-year maturi-
ties (compounding rate of 5.4%); real estate investment
trusts (REITs) by the NAREIT Index (compounding rate
of 5.3%); a single stock with a beta of 1.0, volatility of
30%, and a dividend yield of 0% (compounding rate of
5.3%); and inflation by the Consumer Price Index (com-
pounding rate of 2.5%). Expected market returns on bonds
are derived taking into account yield and other criteria.
An important assumption is that stocks will, over time,
outperform long-term bonds by a reasonable amount,
although this is by no means a certainty. Moreover, actual
future results may not be consonant with Bernstein’s esti-
mates of the range of market returns, as these returns are
subject to a variety of economic, market, and other vari-
ables. Accordingly, this analysis should not be construed
as a promise of actual future results, the actual range of
future results, or the actual probability that these results
will be realized.
 Global Wealth Management

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.

OUR CLIENT-CENTERED MISSION

• To have more knowledge and to use knowledge better than any other investment firm in the world
• To use and share knowledge to help our clients achieve investment success and long-term security
• To place our clients’ interests first and foremost

Research Excellence Broad Array of Services

We believe that superior research is the ultimate source
of superior investment returns and requires both
knowing more and using knowledge better. Knowing
more—having an information advantage over other
market participants—requires doing deep fundamental
and economic research on a truly global scale. Using
knowledge better means identifying and exploiting
pricing anomalies that can provide incremental return
and employing portfolio-construction techniques to
manage risk and return efficiently.
With those goals in mind, we’ve built one of the largest
and broadest research footprints in the business: more
than 300 analysts operating in 12 countries and
covering thousands of securities in capital markets
around the world. Our research effort is organized
into separate groups dedicated to growth equities,
value equities, and fixed income, reflecting the unique
needs of each investment approach.
Disciplined Investment Processes
We leverage our research with systematic portfolio
management. Because our top investment management
professionals determine the policies and make the
decisions that underlie all our investment strategies,
each client, regardless of account size, gets the very best
our firm has to offer. These strategies and decisions are
then further customized in relation to each account’s
tax status and the client’s goals and circumstances.
We offer value, growth, and style-blended stock port-
folios across the global markets, real estate investment
trusts, hedge funds, and taxable and tax-exempt fixed
income portfolios—all actively managed.
Wealth Management Planning
Because we recognize that private clients of very
substantial means have complex needs, we’ve created
a team of people with expertise in a wide range of
disciplines to counsel clients on sophisticated financial
planning. Our wealth management professionals have
experience in areas such as estate planning, intergen-
erational wealth transfer, philanthropy, alternative
asset classes, liquidity events, and investment strategies
for corporate executives. Working together with our
clients’ other professional advisors and aided by a
quantitative state-of-the-art wealth forecasting tool,
we stress-test multiple solutions to complex investment
problems to help clients identify the strategies best
suited to them financially and emotionally.
Client Service and Communications
We recognize that client needs are varied. Our invest-
ment professionals seek to provide clients with the
investment approach that is best for them, and they
pride themselves on personalized and timely service.
Further, our content-rich communications explain the
research basis for our portfolio decisions, our analysis
of recent market developments, and our market
outlook, as well as other research findings.

*As of June 30, 2008

 Global Wealth Management

BER–5293–0708 www.bernstein.com