Portfolio Rebalancing

Part 1 of 2: Strategic Asset Allocation

Antti Ilmanen December 2015
Portfolio Solutions Group Decisions relating to portfolio rebalancing can be
Thomas Maloney considered an active investment strategy and
have important implications for portfolio
Portfolio Solutions Group
returns.

In this article, we explore the key considerations

for investors deciding whether and how to
rebalance liquid strategic asset portfolios. Our
results suggest that the ability to capture price
trends has tended to be at least as important as
cost efficiency in driving the relative long-term
performance of different rebalancing processes.

A companion article (Part 2) examines common

misconceptions about the role and implications
of rebalancing, particularly in the context of
actively managed portfolios.

We thank Gregor Andrade, Cliff Asness, Jeff Dunn, John Huss, AQR Capital Management, LLC
Ronen Israel, Lars Nielsen, Christopher Palazzolo, Scott Richardson One Greenwich Plaza
and Rodney Sullivan for helpful comments and suggestions. Greenwich, CT 06830

p: +1.203.742.3600
f: +1.203.742.3100
w: [Link]
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 1

Introduction Risk Implications: A buy-and-hold portfolio (i.e.,

no rebalancing) that is initially well-diversified
In this article, we explore the most important
tends to become less diversified over time, as
considerations for investors deciding whether
winners earn themselves higher weights and
and how to rebalance to strategic weights (say, a
losers shrink to smaller weights. Its risk level is
60/40 stock/bond allocation), combining brief
therefore likely to be less stable, and to rise over
theoretical discussion with an illustrative
time. Exhibit 1 shows ex ante volatility (one
empirical analysis.
measure of riskiness) for 20 simulations of a buy-
Our results suggest that many sensible and-hold portfolio of 3 theoretical assets with
rebalancing strategies may confer similar long- equal starting allocations, equal expected returns
term risk benefits. In order to choose between and constant volatilities. The portfolio risk level
these variants, investors need to take a view not fluctuates as allocations rise and fall.2 We show
only on likely implementation costs, but, more the allocations for a typical outcome, where one
importantly, on future asset price behavior — asset comes to dominate over time. Rebalanced
specifically, the extent to which prices may portfolios are likely to retain their portfolio-level risk
exhibit trends or mean reversion. Trending prices characteristics better than buy-and-hold portfolios.
tend to favor less frequent rebalancing.

Most investors’ strategic portfolios are based on

Exhibit 1: Ex-Ante Volatility and Allocations for
target capital (or dollar) allocations, and we Simulated Buy-and-Hold Outcomes
Ex Ante Portfolio Volatility

follow this long-established approach in the 18%

present article. In our companion article we 16%

discuss the benefits of rebalancing to target risk
14%
allocations and risk levels (i.e., accounting for
12%
differing riskiness across investments and
through time). This is typically a more dynamic 10%
Time
process. 20 simulated buy-and-hold outcomes
Median buy-and-hold outcome shown below
Continuously rebalanced portfolio
Deciding Whether to Rebalance 100%

Few investors even consider the possibility of not 80%

Capital Allocation

rebalancing at all1 — and with good reason, as we 60%

will demonstrate — but examining why we 40%
rebalance helps us to ask the right questions
20%
when choosing how to rebalance. Rebalancing
0%
has risk, return and cost implications, and all Time
Asset A Asset B Asset C
three should be considered. The risk and cost
Source AQR. For illustrative purposes only, not indicative of an actual
implications are foreseeable. The return portfolio or actual investments. Simulated monthly returns over 10 years
implications are harder to predict, but some rules for 3 uncorrelated assets with constant expected volatility of 20%,
expected Sharpe ratio of 0.5 and normally-distributed serially-indepen-
of thumb can be helpful. dent returns. Portfolios start with equal allocations to the 3 assets.

2
For a buy-and-hold portfolio of equities and bonds, the risk level is likely
1
Sharpe (2010) argues that a buy-and-hold (or adaptive) portfolio has the to experience cyclical variations, with equities having a higher weight
advantage of better macro consistency, but it is not clear that this after equity bull markets, and a lower weight after bear markets. Over a
advantage is aligned with investor objectives, given the possibility of longer period, the asset class with the higher expected return (i.e.,
significant drift over time in asset class weights and portfolio risk. equities) will likely come to dominate the portfolio.
2 Portfolio Rebalancing — Part 1: Strategic Asset Allocation

Return Implications: For many investors, the further in Part 2 of this series. Essentially, by
risk implications are sufficient reason to preventing winners from earning higher weights,
rebalance, and only the cost-minimizing and losers from decaying to lower weights,
implementation details (frequency, tolerance rebalancing neutralizes compounding effects
bands, etc.) remain to be decided. But within the portfolio. In other words, a rebalanced
rebalancing to constant weights is also an active portfolio forgoes the very best potential buy-and-
contrarian strategy when compared to a buy-and- hold outcomes, where winning investments keep
3
hold benchmark. Rebalancing to constant on winning and persistent losers fizzle out to
weights involves selling winners and buying losers, small, inconsequential weights. But as shown in
and tends to be profitable during periods when Exhibit 1, it also tends to maintain a lower risk
investments have either similar or mean- level by preventing the concentration of risk
reverting performance, and costly when among winning investments.5
investments have either persistently divergent or
For these reasons, rebalancing tends to reshape
trending outcomes. It is difficult to predict which
the distribution of potential portfolio outcomes,
return pattern will prevail for a given set of
making it narrower and less positively skewed.
investments over a given horizon. Rebalancing is
This increases the median outcome while leaving
“short a regime change” — that is, it can suffer
the mean unchanged, meaning that a rebalanced
when the world changes or when initial
portfolio is more likely to realize positive returns
assumptions turn out to be false.
— and more likely to realize returns exceeding the
A buy-and-hold approach may be preferable for a mean expectation — over the investment horizon.
portfolio of investments with highly uncertain
Cost Implications: Rebalancing incurs costs. Our
expected returns, likely to have widely dispersed
analysis suggests that transaction costs from
outcomes over the investment horizon (or even
rebalancing liquid assets may be modest, and for
fall to near-zero value — examples could be small-
most investors costs shouldn’t dictate whether or
cap equity or venture capital portfolios). A
not to rebalance. They may, however, influence
rebalanced approach may be preferable where the
the decision of how to rebalance.
portfolio components are “tried and true” long-
term investments, such as entire asset classes.4 Deciding How to Rebalance
A related but distinct consideration is whether When designing a rebalancing process there are
the investments exhibit trending or mean- two6 main decisions to be made:
reverting behavior over shorter horizons. This —
 When to rebalance? (How often? Fixed
together with the expected transaction costs —
schedule or trigger-based system?)
may have implications for both the likely return
impact and the preferred frequency of
rebalancing, as we discuss later. 5
Harvey et al (2014) suggest a rebalanced portfolio is susceptible to
larger drawdowns than a buy-and-hold portfolio. This is true for two
The impact of rebalancing on expected returns is portfolios entering a period of sustained investment losses with the same
weights, as the authors illustrate: the rebalanced portfolio keeps buying
rather subtle and much debated — it is discussed the losing investments. However, this analysis misses the tendency of
rebalanced portfolios to be already better diversified at the onset of such
periods. When considered in the context of longer investment horizons,
3
We might equally say that choosing not to rebalance is an active rebalanced portfolios tend to experience smaller drawdowns because of
momentum-biased strategy compared to a rebalanced benchmark. One their lower average risk level (see later analysis). Harvey et al.’s
might argue that a continuously rebalanced portfolio is a more justifiable arguments regarding the complementarity of rebalancing and momentum
strategic benchmark. However, in this article we follow the literature in are, however, persuasive, as we illustrate later.
6
viewing the buy-and-hold portfolio as the passive benchmark. A third decision is whether to rebalance major asset classes only, or also
4
This distinction is discussed in Ang (2014). regional allocations (see NBIM 2012).
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 3

 How much to rebalance? (Fully back to the it may be cost-efficient to use dividend or coupon
benchmark, or only partially?) payments to rebalance where possible (by
Norges Bank Investment Management’s (NBIM) reinvesting in a different asset class), reducing
discussion note on rebalancing (2012) provides a the need to sell underlying securities. Fund
useful combination of theoretical discussion and contributions and distributions may also be
empirical results. Some researchers7 have integrated in the rebalancing process where

designed explicitly optimized rebalancing possible.

processes, but the expected additional benefits of Determinant 3: Expectations of trending or mean-
these may be small for a strategic portfolio with reverting investment performance
static or slow-moving allocation targets (by
If investment returns exhibit trends or mean
contrast, dynamic strategies with much higher
reversion, the rebalancing frequency can affect
turnover may expect significant benefits from
expected portfolio returns. Rebalancing at or
cost-optimized rebalancing).
within the frequency of any mean-reversion
In our view, there are three main determinants of patterns will tend to earn positive returns, while
which method may be preferred for a given rebalancing at or within the frequency of any
investor, relating to risk preferences, costs and momentum effects will be fighting against these
return expectations. and therefore tend to suffer negative returns.
There is evidence that many investments exhibit
Determinant 1: Tolerance of short-term
3- to 12-month momentum,8 which may be
variations in portfolio risk characteristics
exploited by less-frequent rebalancing (annual or
Under plausible assumptions and for reasonable
lower frequency). This can allow winning and
choices of frequency and/or tolerance bands,
losing trends to “play out” and compound
most rebalancing processes will maintain long-
between rebalances.
term risk characteristics almost equally well (this
important observation is supported by the The Pitfalls of Discretionary Rebalancing
empirical analysis in the next section). However, Some investors may consider rebalancing (or not
some will permit a larger short-term tracking rebalancing) an opportunity to express
error away from the strategic benchmark risk discretionary tactical market views — for
characteristics (including allocations, ex ante example, based on the current or expected
volatility and equity beta). macroeconomic environment. We believe tactical
market timing is more difficult than hindsighted
Determinant 2: Expected costs
narratives suggest,9 even without the constraint
These may include:
of expressing views only in the context of a
 Transaction costs (bid-ask and market impact) rebalancing decision. The momentum effects
 Operational costs (internal or external mentioned above may be one of the better market
management expenses or fees) timing indicators, but these, too, are low-
 Tax implications for taxable investors (e.g., if conviction signals when applied to a single

rebalances trigger capital gains tax liability) market. Any tactical views should be applied with
humility and caution, and a predefined, rule-
Higher costs will favor less-frequent or partial
rebalancing, or wider tolerance bands. Note that
8
See for example Hurst, Ooi and Pedersen (2012).
7 9
See for example Sun et al (2006). See AQR Alternative Thinking 4Q2014.
4 Portfolio Rebalancing — Part 1: Strategic Asset Allocation

based rebalancing strategy may be the best way to allocation of any investment under- or overshoots
10
ensure that diversification is maintained. by a given percentage of its target level (see
appendix for details of threshold calculation).
A Simple Empirical Analysis
The first two rows show ex ante measures of
We build a simple strategic portfolio similar to variation. As we would expect, the average
11
the liquid portion of many institutional allocation range depends on the frequency of
portfolios (see Exhibit 2), and examine several rebalancing, with more frequent schedules or
rebalancing approaches using 43 years of tighter trigger thresholds keeping allocations
monthly return data since 1972. To increase the closer to their targets. However, compared to buy-
breadth of events in the sample period with and-hold, all the rebalanced portfolios have more
potential rebalancing implications, we include stable ex ante volatility (second row).
two equity and two bond allocations, as well as a
Similarly, compared to buy-and-hold, all the
small commodity allocation. With an even
rebalanced portfolios have higher returns, lower
simpler two-asset U.S. stock/bond portfolio, the
and more stable realized volatility, and smaller
main conclusions are similar, though results
drawdowns. These performance characteristics
differ in details (see appendix).
improve roughly the same amount regardless of
Exhibit 2: Strategic Portfolio Allocations frequency of rebalancing.
Target
Asset Class Proxy Index A similar pattern is seen in the lower panel,
Alloc.
U.S. Equities 30% MSCI US which shows the impact of the degree of
Non-U.S. Equities 20% MSCI World ex US rebalancing — how far we rebalance toward the
U.S. Govt. Bonds 20% Barclays US Tsy Intermediate*
target allocations. The more complete
Non-U.S. Govt Bonds 20% Barclays Glob Ex US Tsy Hedged*
Commodities 10% GSCI
rebalancing processes achieve tighter allocation
Source: AQR. From January 1972 to December 1972, U.S. Govt. Bond ranges, but all the processes (even rebalancing
proxy is 10-year Treasury. From January 1972 to December 1986, Non- only 25% toward targets) achieve similar
U.S. Govt. Bond proxy is a GDP-weighted portfolio of G6 ex U.S. 10-year
bonds.
improvements in long-term performance
statistics in this analysis.
Exhibit 3 (upper panel) shows the impact of six
The last row in each panel shows that direct
different rebalancing schedules on amount of
transaction costs are relatively modest (a few
variation in allocations, performance and
basis points), even for frequent or complete
turnover, compared to buy and hold. For the
rebalancing (compared to, for example, those
annual and biennial processes we show statistics
faced by active investment strategies). Turnover
averaged across all the possible quarter-end
can be reduced by (1) less frequent rebalancing or
rebalance schedules. The three trigger-based
(2) partial rebalancing, but both also allow more
processes rebalance all investments when the
short-term variation in allocations. Of these two
10
variables, partial rebalancing gives a larger cost
Some authors (e.g., Gort and Burgener, 2014) have argued that
rebalancing using option positions has the advantage of effectively taking saving for a smaller increase in allocation ranges.
the rebalancing mechanism (and any temptation to override it) out of the
investor’s hands before price moves trigger a rebalance. This method has
typically been proposed to rebalance the equity allocation only, by selling The Role of Momentum and Reversal Effects
a straddle with strike prices set according to the chosen tolerance band.
This strategy combines rebalancing and volatility-selling and is beyond
the scope of this short article.
One noticeable result in Exhibit 3A is that during
11
Illiquid investments are difficult and expensive to rebalance. This this period and for these investments and proxies,
difficulty and cost should be taken into account when deciding on the
sizing of any illiquid asset allocations. annual and biennial schedules outperformed
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 5

Exhibit 3: Hypothetical Performance of Different Rebalancing Methods 1972-2014

A. Impact of Schedule
Buy and Calendar-based Rebalance Trigger-based Rebalance
Impact of Schedule
Hold Biennial Annual Monthly +/-30% +/-20% +/-10%
Ante

Avg Allocation Range* 104% 95% 70% 27% 67% 52% 35%
Ex

Vol of Ex Ante Volatility* 2.3% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0%
Net Total Return 8.7% 9.3% 9.2% 9.0% 9.1% 9.1% 9.0%
Performance

Volatility 9.0% 8.0% 8.0% 8.1% 8.3% 8.2% 8.1%

Net Sharpe Ratio 0.32 0.44 0.43 0.39 0.40 0.40 0.40
Vol of Realized Volatility* 2.5% 1.8% 1.8% 2.0% 2.0% 2.0% 2.0%
Max Drawdown -41.2% -29.7% -29.7% -32.8% -32.6% -32.8% -32.8%
Annual Turnover 0.0% 5.1% 9.2% 26.0% 6.3% 9.4% 13.8%
Turnover

Avg # Rebals / Year 0.0 0.5 1.0 12.0 0.4 0.9 2.3
Avg Trade Size NA 10.3% 9.2% 2.2% 16.8% 11.0% 6.0%
Annual Trade Cost* 0.00% 0.03% 0.05% 0.13% 0.03% 0.05% 0.07%

B. Impact of Degree
Buy and Annual Rebalance Monthly Rebalance
Impact of Degree
Hold 25% 50% 100% 25% 50% 100%
Ante

Avg Allocation Range* 104% 101% 86% 70% 42% 34% 27%
Ex

Vol of Ex Ante Volatility* 2.3% 1.8% 1.9% 2.0% 2.0% 2.0% 2.0%
Net Total Return 8.7% 9.2% 9.3% 9.2% 9.0% 9.0% 9.0%
Performance

Volatility 9.0% 8.1% 8.0% 8.0% 8.0% 8.1% 8.1%

Net Sharpe Ratio 0.32 0.41 0.43 0.43 0.40 0.40 0.39
Vol of Realized Volatility* 2.5% 1.9% 1.9% 1.8% 1.9% 2.0% 2.0%
Max Drawdown -41.2% -31.2% -30.1% -29.7% -31.7% -32.5% -32.8%
Annual Turnover 0.0% 3.1% 5.3% 9.2% 10.9% 15.8% 26.0%
Turnover

Avg # Rebals / Year 0.0 1.0 1.0 1.0 12.0 12.0 12.0
Avg Trade Size NA 3.1% 5.3% 9.2% 0.9% 1.3% 2.2%
Annual Trade Cost* 0.00% 0.02% 0.03% 0.05% 0.05% 0.08% 0.13%
Source: AQR. “Avg alloc range” is [(Max-min allocation) / target allocation], averaged across asset classes. ”Vol of ex ante vol” is volatility of ex ante
volatility based on rolling 36-month covariance matrix. Gross total return is annualized arithmetic rate of return. Returns and Sharpe ratios are net of
estimated transaction costs from rebalancing, gross of fees. Risk-free rate is 3-month T-Bill rate. “Vol of realized vol” is volatility of rolling 36-month
volatility. Annual trade cost assumes uniform transaction cost of 0.5% for all asset classes. Allocations are as described in Exhibit 2. Hypothetical data
has inherent limitations, some of which are disclosed herein.

monthly or trigger-based schedules. This is Exhibit 4: Hypothetical Sharpe Ratios for

apparently due to better harnessing of multi- Different Rebalancing Methods 1972-2014
month momentum effects by less frequent, Period
Buy & Annual Rebalance 1M
Hold Mar Jun Sep Dec Rebal
calendar-based processes. This outperformance
1972-2014 0.32 0.43 0.43 0.43 0.43 0.39
by less frequent schedules passes various 1972-1993 0.34 0.36 0.36 0.41 0.38 0.35
robustness checks. As well as the average 1993-2014 0.32 0.50 0.49 0.46 0.49 0.43
outperformance shown in Exhibit 3, each of the Source: AQR. Hypothetical Sharpe ratios net of transaction costs, based

March, June, September and December annual on monthly data. Risk-free rate is 3-month T-Bill rate. Allocations are as
described in Exhibit 2.
schedules outperforms monthly rebalancing, over
the full sample and in both halves of it. These To help understand this outperformance, we
sub-results are shown in Exhibit 4. show in Exhibit 5 autocorrelations observed in
the asset returns used in our analysis. All asset
6 Portfolio Rebalancing — Part 1: Strategic Asset Allocation

classes tended to display positive autocorrelation Investors must decide whether the patterns
(momentum) at frequencies of up to one year, shown in Exhibit 5 are persistent and reliable
and negative autocorrelation (mean reversion) at enough to influence the design of a rebalancing
3- to 5-year frequencies. process. Certainly, evidence of 3- to 12-month
momentum effects has been observed in many
Exhibit 5: Autocorrelations in Asset Excess different asset classes and as far back as data
Returns 1972-2014
permits us to go. Multi-year mean reversion
US INT
Frequency US FI INT FI COM Mean
EQ EQ patterns are also seen in many contexts, but
1 month 0.04 0.11 0.15 0.26 0.17 0.15
perhaps less consistently. Over the course of a
3 months 0.09 0.12 0.00 0.15 0.05 0.08
12 months 0.05 0.11 0.21 0.12 -0.04 0.09
shorter investment horizon of 5 or 10 years, mean
3 years -0.25 -0.37 -0.18 -0.51 -0.33 -0.33 reversion may not occur.
5 years -0.13 -0.45 0.01 -0.65 -0.34 -0.31

Source: AQR. Asset class proxies are as described in Exhibit 2. Allocations Through Time

When we compare the returns earned by buy- Exhibit 6 shows allocations over time for four of
and-hold and frequently rebalanced portfolios, the strategies discussed in this analysis. The
these momentum and reversal effects at different drifting buy-and-hold allocations are clearly
horizons tend to offset each other. The frequently evident. For example, outperformance by U.S.
rebalanced portfolio suffers a drag from short- equities led to above-target allocations in 1999
term momentum, but earns a bonus from longer- and 2014. Also clearly visible is the varying degree
term mean-reversion. Annually or biennially of stability in allocations achieved by the
rebalanced portfolios, however, get the best of different rebalancing processes. It may seem
both worlds in this sample: they behave like buy- surprising that, as discussed above, all of these
and-hold portfolios at shorter horizons rebalancing processes achieve similar reductions
(harnessing momentum), but like rebalanced in volatility and drawdowns.
portfolios at longer horizons (harnessing
reversals). 12 Conclusions

Rebalancing is not a surefire winner: the return

Trigger-based processes appear not to enjoy the
impact will always depend on investment
same advantage. For example, the +/–20%
outcomes. The risk impact may be more
threshold triggers roughly one full rebalance per
predictable: over the longer term, rebalancing
year on average, and achieves tighter allocation
tends to lead to more predictable risk
ranges than a fixed annual process, for similar
characteristics, while seemingly passive buy-and-
turnover.13 However, it underperforms the latter,
hold portfolios are likely to have more variable
probably because it tends to trigger rebalances
risk outcomes.
during trend events (see appendix). A fixed
annual schedule that is indifferent to within-year Our analysis suggests that many reasonable
market moves is more likely to allow trends to rebalancing processes may be expected to achieve
play out. the objective of maintaining portfolio
12
diversification over the long-term (this can also
Rebalancing even less frequently — every three or five years — still
captures this advantage. At 10 years, the benefit begins to fade as full be demonstrated more generally, using simulated
mean-reversion cycles occur between rebalances.
13
It also has the advantage of reducing the element of chance introduced
by an arbitrary calendar-based schedule. The timing of rebalances can
have a significant and discomfiting short-term impact during a major
market event like the Global Financial Crisis of 2008.
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 7

Exhibit 6: Hypothetical Proxy Portfolio Capital Allocations 1972-2014

A. Buy and Hold
40%
35%
Capital Allocation

US Equities
30%
25% Non-US Equities

20% US FI
15% Non-US FI
10% Commodities
5%
0%
1970 1980 1990 2000 2010

B. Annual Full Rebalance (year-end)

40%
35%
US Equities
Capital Allocation

30%
Non-US Equities
25%
20% US FI
15% Non-US FI
10% Commodities
5%
0%
1970 1980 1990 2000 2010

C. Monthly Full Rebalance

40%
35%
US Equities
Capital Allocation

30%
Non-US Equities
25%
20% US FI
15% Non-US FI
10% Commodities
5%
0%
1970 1980 1990 2000 2010

D. +/-20% Trigger-Based Full Rebalance

40%
35%
US Equities
Capital Allocation

30%
Non-US Equities
25%
20% US FI
15% Non-US FI
10% Commodities
5%
0%
1970 1980 1990 2000 2010

Source: AQR. Please see Exhibit 2 for portfolio constituents. Hypothetical data has inherent limitations, some of which are disclosed herein.
8 Portfolio Rebalancing — Part 1: Strategic Asset Allocation

data). When choosing implementation details,

the most important trade-off appears to be
between experiencing short-term variations in
allocations and harvesting historically-observed
price momentum effects. Investors should
therefore balance their tolerance for such
variations against their expectation or belief that
multi-month price momentum effects will
continue in the future.

Cost considerations may be secondary to the

above effects, as our analysis suggests that
expected transaction costs for rebalancing liquid
investments to fixed targets are modest — but
they are also more certain. Cost considerations
tend to favor partial or less-frequent rebalancing.
Partial rebalancing is the more efficient cost-
reduction measure if price momentum effects are
ignored (it gives a bigger turnover reduction for a
smaller increase in allocation drift). In the
presence of momentum, less-frequent
rebalancing may be preferred.

One approach that we do not analyze here is an

explicit momentum strategy, either overlaid on
the rebalancing process — as proposed by Harvey
et al. (2014) — or given its own allocation. This
may be less palatable or less practicable for some
institutions,14 but it may also be a more efficient
and less constrained way to benefit from time
series momentum. Whether investors choose to
integrate or keep separate their views on
rebalancing and time-varying expected returns,
we hope that a better understanding of the
implications of rebalancing choices may help
them to make an informed decision.

14
For a discussion of institutional preferences for contrarian rather than
momentum-based tactical views, see AQR Alternative Thinking, 4Q2014.
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 9

Appendices similar performance and risk improvements

compared to buy and hold, and (2) infrequent
Simple U.S. 60/40 Portfolio calendar-based schedules appear to give the best
Exhibit A1 repeats the empirical analysis for a long-term performance by better capturing
15
simple 60/40 portfolio of U.S. stocks and bonds. momentum effects.
For this portfolio there is less difference between
buy and hold and the various rebalanced Further Analysis on Trigger-Based Rebalancing
approaches. However, the results are broadly Specifying trigger thresholds is as much art as
consistent with those in the main article, namely science. We use a combination of relative,
that (1) all the rebalancing methods produce absolute and volatility-adjusted inputs, in an

Exhibit A1: Hypothetical Performance of Different Rebalancing Methods 1972-2014

A. Impact of Schedule
Buy and Calendar-based Rebalance Trigger-based Rebalance
Impact of Schedule
Hold Biennial Annual Monthly +/-30% +/-20% +/-10%
Ante

Avg Allocation Range* 73% 58% 47% 21% 33% 25% 22%
Ex

Vol of Ex Ante Volatility* 2.4% 2.3% 2.3% 2.3% 2.3% 2.3% 2.3%
Net Total Return 8.7% 8.8% 8.8% 8.7% 8.8% 8.7% 8.7%
Performance

Volatility 10.0% 9.3% 9.4% 9.5% 9.5% 9.5% 9.5%

Net Sharpe Ratio 0.29 0.33 0.32 0.30 0.31 0.31 0.31
Vol of Realized Volatility* 2.5% 2.2% 2.2% 2.3% 2.2% 2.2% 2.3%
Max Drawdown -35.8% -27.8% -28.5% -30.9% -30.7% -30.9% -30.9%
Annual Turnover 0.0% 3.6% 6.7% 19.3% 5.7% 8.2% 11.5%
Turnover

Avg # Rebals / Year 0.0 0.5 1.0 12.0 0.5 1.0 2.4
Avg Trade Size N/A 7.3% 6.7% 1.6% 12.2% 8.4% 4.8%
Annual Trade Cost* 0.00% 0.02% 0.03% 0.10% 0.03% 0.04% 0.06%

B. Impact of Degree
Buy and Annual Rebalance Monthly Rebalance
Impact of Degree
Hold 25% 50% 100% 25% 50% 100%
Ante

Avg Allocation Range* 73% 59% 51% 47% 26% 20% 21%
Ex

Vol of Ex Ante Volatility* 2.4% 2.2% 2.2% 2.3% 2.3% 2.3% 2.3%
Net Total Return 8.7% 8.8% 8.8% 8.8% 8.7% 8.7% 8.7%
Performance

Volatility 10.0% 9.5% 9.4% 9.4% 9.4% 9.4% 9.5%

Net Sharpe Ratio 0.29 0.31 0.32 0.32 0.31 0.31 0.30
Vol of Realized Volatility* 2.5% 2.3% 2.2% 2.2% 2.2% 2.2% 2.3%
Max Drawdown -35.8% -28.2% -28.0% -28.5% -30.0% -30.7% -30.9%
Annual Turnover 0.0% 2.0% 3.6% 6.7% 8.0% 11.6% 19.3%
Turnover

Avg # Rebals / Year 0.0 1.0 1.0 1.0 12.0 12.0 12.0
Avg Trade Size N/A 2.0% 3.6% 6.7% 0.7% 1.0% 1.6%
Annual Trade Cost* 0.00% 0.01% 0.02% 0.03% 0.04% 0.06% 0.10%
Source: AQR. “Avg alloc range” is [(Max-min allocation) / target allocation], averaged across asset classes. ”Vol of ex ante vol” is volatility of ex ante
volatility based on rolling 36-month covariance matrix. Gross total return is annualized arithmetic rate of return. Returns and Sharpe ratios are net of
estimated transaction costs from rebalancing, gross of fees. Risk-free rate is 3-month T-Bill rate. “Vol of realized vol” is volatility of rolling 36-month
volatility. Annual trade cost assumes uniform transaction cost of 0.5% for all asset classes. Allocations are as described in Exhibit 2. Hypothetical data
has inherent limitations, some of which are disclosed herein.

15
Represented by the MSCI U.S. Index and the Barclays Intermediate
Treasury Index respectively.
10 Portfolio Rebalancing — Part 1: Strategic Asset Allocation

attempt to give each asset class a roughly equal Timing of Trigger-Based Rebalances
probability of triggering a rebalance. For Exhibit A3 shows the dates of rebalances for one
example, to calculate tolerance bands for the version of the strategy: +/-30% full rebalancing. It
+/-20% strategy, we multiply by 20% the average also shows past 12-month total returns for each
of the asset’s target allocation (relative asset class on each rebalance date. Highlighted
component) and the average target allocation cells have 12-month excess return exceeding
across all assets (absolute component). Then we average +1 stdev (green) or worse than average -1
average this with a volatility-adjusted equivalent. stdev (red), based on full-period average and
Varying Tolerance Bands volatility. It is clear that rebalances tend to be
triggered after abnormally high or low returns for
Exhibit A2 shows the impact of varying the
one or more asset classes. Trigger-based
trigger threshold on various portfolio
rebalance trades are therefore more likely to
characteristics. Widening thresholds reduces
contradict time series momentum signals than
costs but permits wider ranges of allocations,
price-agnostic calendar-based trades, which may
while over this period returns and Sharpe ratios
explain why the latter process outperforms in our
are unchanged. Partial rebalancing reduces costs
analysis.
with little impact on allocation ranges. Compared
to a fixed annual rebalance, half rebalancing with Exhibit A3: Asset Class Past 12-Month Total
+/-30% thresholds gives both tighter allocations Return at Each Rebalance for +/-30% Full
and lower costs. However, as mentioned Rebalance Strategy, 1972-2014
previously, the price-agnostic calendar-based Trigger US Non-US Non-US Commo
Date Equities Equities US FI FI dities
schedule appears to better capture momentum
1 Jul-73 0.7% 20.4% -1.0% 2.2% 108.9%
effects (higher return and Sharpe ratio). 2 Aug-74 -30.4% -30.5% 3.9% -0.2% 29.6%
3 Oct-77 -7.8% 24.5% 4.9% 22.9% 8.1%
Exhibit A2: Impact of Varying Rebalance 4 Jan-80 14.1% 15.3% 4.3% 2.9% 34.8%
Triggers 1972-2014 5 Feb-82 -10.1% -8.9% 13.8% 17.1% -8.3%
Fixed Trigger Threshold Fixed 6 Aug-86 37.2% 93.7% 18.8% 15.3% 15.5%
Full Rebalance
1M 10% 20% 30% 40% 1Y 7 Sep-89 31.4% 22.5% 9.6% 7.6% 50.8%
Avg Allocation Range* 27% 35% 52% 67% 80% 68% 8 Sep-90 -9.5% -27.0% 8.5% -1.8% 65.9%
Vol of Ex Ante Vol* 2.0% 2.0% 2.0% 2.0% 2.1% 2.0% 9 Sep-95 29.7% 5.8% 10.6% 14.3% 6.1%
Net Total Return 9.0% 9.0% 9.1% 9.1% 9.1% 9.3%
10 Jul-97 52.1% 19.1% 8.5% 13.9% 10.9%
Volatility 8.1% 8.1% 8.2% 8.3% 8.3% 8.0%
11 Sep-01 -27.2% -29.1% 12.4% 9.2% -17.4%
Net Sharpe Ratio 0.39 0.40 0.40 0.40 0.40 0.43
Annual Trade Cost* 0.13% 0.07% 0.05% 0.03% 0.03% 0.05% 12 Aug-05 12.6% 24.8% 1.9% 7.6% 42.6%
13 Oct-08 -36.6% -46.5% 8.4% 4.6% -24.9%
Half Rebalance
Fixed Trigger Threshold Fixed 14 Feb-09 -43.5% -50.3% 5.2% 5.0% -58.1%
1M 10% 20% 30% 40% 1Y
15 Jul-09 -20.7% -22.4% 5.9% 7.8% -53.1%
Avg Allocation Range* 34% 38% 54% 66% 82% 85%
16 May-13 26.3% 29.6% 0.0% 3.3% 3.3%
Vol of Ex Ante Vol* 2.0% 2.0% 2.0% 2.1% 2.1% 1.9%
Net Total Return 9.0% 9.1% 9.1% 9.1% 9.1% 9.3%
Source: AQR. Highlighted cells have 12-month excess return exceeding
Volatility 8.1% 8.1% 8.1% 8.3% 8.4% 8.0% average +1 stdev (green) or worse than average -1 stdev (red), based on
Net Sharpe Ratio 0.40 0.40 0.40 0.40 0.40 0.43 full-period average and volatility. Please see Exhibit 2 for portfolio
Annual Trade Cost* 0.08% 0.05% 0.03% 0.02% 0.02% 0.03% constituents.

Source: AQR. “Avg allocation range” is [(Max-min allocation) / target Finally, Exhibit A4 illustrates the timing of these
allocation], averaged across asset classes. ”Vol of ex ante vol” is volatility
of ex ante volatility based on rolling 36-month covariance matrix. Returns rebalances on a graph of cumulative asset
and Sharpe ratios are net of estimated transaction costs from
rebalancing, gross of fees. Risk-free rate is 3-month T-Bill rate. Annual
returns.
trade cost assumes uniform transaction cost of 0.5% for all asset
classes. Allocations are as described in Exhibit 2.
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 11

Exhibit A4: Cumulative Gross Returns and Hypothetical Rebalance Triggers (+/-30% Full Rebalance)

128.0

64.0
Cumulative Total Return

32.0

16.0

8.0

4.0

2.0

1.0

0.5
1970 1980 1990 2000 2010

Source: AQR. Please see Exhibit 2 for portfolio constituents. Hypothetical data has inherent limitations, some of which are disclosed herein.
12 Portfolio Rebalancing — Part 1: Strategic Asset Allocation

References
Ang, Andrew, 2014, “Asset Management: A Systematic Approach to Factor Investing,” OUP.

AQR Alternative Thinking, 4Q 2014, “Challenges of Incorporating Tactical Views.”

Gort, Christoph, and E. Burgener, 2014, “Rebalancing Using Options,” working paper.

Harvey, Campbell R., N. Granger, D. Greenig, S. Rattray and D. Zou, 2014, “Rebalancing Risk,”
working paper.

Hurst, Brian, Y.H. Ooi, and L. Pedersen, 2012, “A Century of Evidence on Trend-Following Investing,”
AQR White Paper.

Moskowitz, T., Y.H. Ooi, and L. Pedersen, 2012, “Time Series Momentum,” The Journal of Financial
Economics, 104(2), 228–250.

NBIM Discussion Note, 2012, “Empirical analysis of rebalancing strategies.”

Qian, Edward, 2014, “To Rebalance or Not to Rebalance: A Statistical Comparison of Terminal Wealth
of Fixed-Weight and Buy-and-Hold Portfolios,” working paper.

Sharpe, William F., 2010, “Adaptive Asset Allocation Policies,” Financial Analysts Journal, 66(3), 45-59.

Sun, Walter, A. Fan, L. Chen, T. Schouwenaars, and M.A. Albota, 2006, “Optimal Rebalancing for
Institutional Portfolios,” The Journal of Portfolio Management, 32(2), 33-43.
 Portfolio Rebalancing — Part 1: Strategic Asset Allocation 13

This document has been provided to you solely for information purposes and does not constitute an offer or solicitation of an offer or any
advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The factual
information set forth herein has been obtained or derived from sources believed by the author and AQR Capital Management, LLC (“AQR”) to
be reliable but it is not necessarily all-inclusive and is not guaranteed as to its accuracy and is not to be regarded as a representation or
warranty, express or implied, as to the information’s accuracy or completeness, nor should the attached information serve as the basis of any
investment decision. This document is intended exclusively for the use of the person to whom it has been delivered by AQR, and it is not to
be reproduced or redistributed to any other person. The information set forth herein has been provided to you as secondary information and
should not be the primary source for any investment or allocation decision. Past performance is not a guarantee of future performance.
This material is not research and should not be treated as research. This paper does not represent valuation judgments with respect to any
financial instrument, issuer, security or sector that may be described or referenced herein and does not represent a formal or official view of
AQR. The views expressed reflect the current views as of the date hereof and neither the author nor AQR undertakes to advise you of any
changes in the views expressed herein.
The information contained herein is only as current as of the date indicated, and may be superseded by subsequent market events or for
other reasons. Charts and graphs provided herein are for illustrative purposes only. The information in this presentation has been developed
internally and/or obtained from sources believed to be reliable; however, neither AQR nor the author guarantees the accuracy, adequacy or
completeness of such information. Nothing contained herein constitutes investment, legal, tax or other advice nor is it to be relied on in
making an investment or other decision. There can be no assurance that an investment strategy will be successful. Historic market trends are
not reliable indicators of actual future market behavior or future performance of any particular investment which may differ materially, and
should not be relied upon as such.
The information in this paper may contain projections or other forward-looking statements regarding future events, targets, forecasts or
expectations regarding the strategies described herein, and is only current as of the date indicated. There is no assurance that such events
or targets will be achieved, and may be significantly different from that shown here. The information in this document, including statements
concerning financial market trends, is based on current market conditions, which will fluctuate and may be superseded by subsequent market
events or for other reasons.
HYPOTHETICAL PERFORMANCE RESULTS HAVE MANY INHERENT LIMITATIONS, SOME OF WHICH, BUT NOT ALL, ARE DESCRIBED
HEREIN. NO REPRESENTATION IS BEING MADE THAT ANY FUND OR ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFITS OR
LOSSES SIMILAR TO THOSE SHOWN HEREIN. IN FACT, THERE ARE FREQUENTLY SHARP DIFFERENCES BETWEEN HYPOTHETICAL
PERFORMANCE RESULTS AND THE ACTUAL RESULTS SUBSEQUENTLY REALIZED BY ANY PARTICULAR TRADING PROGRAM. ONE
OF THE LIMITATIONS OF HYPOTHETICAL PERFORMANCE RESULTS IS THAT THEY ARE GENERALLY PREPARED WITH THE BENEFIT
OF HINDSIGHT. IN ADDITION, HYPOTHETICAL TRADING DOES NOT INVOLVE FINANCIAL RISK, AND NO HYPOTHETICAL TRADING
RECORD CAN COMPLETELY ACCOUNT FOR THE IMPACT OF FINANCIAL RISK IN ACTUAL TRADING. FOR EXAMPLE, THE ABILITY TO
WITHSTAND LOSSES OR TO ADHERE TO A PARTICULAR TRADING PROGRAM IN SPITE
OF TRADING LOSSES ARE MATERIAL POINTS THAT CAN ADVERSELY AFFECT ACTUAL TRADING RESULTS. THERE ARE NUMEROUS
OTHER FACTORS RELATED TO THE MARKETS IN GENERAL OR TO THE IMPLEMENTATION OF ANY SPECIFIC TRADING PROGRAM
WHICH CANNOT BE FULLY ACCOUNTED FOR IN THE PREPARATION OF HYPOTHETICAL PERFORMANCE RESULTS, ALL OF WHICH
CAN ADVERSELY AFFECT ACTUAL TRADING RESULTS. The hypothetical performance results contained herein represent the application
of the quantitative models as currently in effect on the date first written above and there can be no assurance that the models will remain
the same in the future or that an application of the current models in the future will produce similar results because the relevant market and
economic conditions that prevailed during the hypothetical performance period will not necessarily recur. Discounting factors may be
applied to reduce suspected anomalies. This backtest’s return, for this period, may vary depending on the date it is run. Hypothetical
performance results are presented for illustrative purposes only. In addition, our transaction cost assumptions utilized in backtests, where
noted, are based on AQR Capital Management, LLC’s, (“AQR”)’s historical realized transaction costs and market data. Certain of the
assumptions have been made for modeling purposes and are unlikely to be realized. No representation or warranty is made as to the
reasonableness of the assumptions made or that all assumptions used in achieving the returns have been stated or fully considered.
Changes in the assumptions may have a material impact on the hypothetical returns presented. Actual advisory fees for products offering
this strategy may vary.

Simulated l performance results (e.g., quantitative backtests) have many inherent limitations, some of which, but not all, are described herein.
No representation is being made that any fund or account will or is likely to achieve profits or losses similar to those shown herein. In
fact, there are frequently sharp differences between hypothetical performance results and the actual results subsequently realized by any
particular trading program. One of the limitations of simulated results is that they are generally prepared with the benefit of hindsight.
In addition, simulated trading does not involve financial risk, and no hypothetical trading record can completely account for the impact of
financial risk in actual trading. For example, the ability to withstand losses or adhere to a particular trading program in spite of trading
losses are material points which can adversely affect actual trading results. The simulated results contained herein represent the application
of the quantitative models as currently in effect on the date first written above and there can be no assurance that the models will remain the
same in the future or that an application of the current models in the future will produce similar results because the relevant market and
economic conditions that prevailed during the hypothetical performance period will not necessarily recur. There are numerous other
factors related to the markets in general or to the implementation of any specific trading program which cannot be fully accounted for
in the preparation of hypothetical performance results, all of which can adversely affect actual trading results. Discounting factors
may be applied to reduce suspected anomalies. This backtest’s return, for this period, may vary depending on the date it is run. Simulated
performance results are presented for illustrative purposes only.
There is a risk of substantial loss associated with trading commodities, futures, options, derivatives and other financial instruments.
Before trading, investors should carefully consider their financial position and risk tolerance to determine if the proposed trading style is
appropriate. Investors should realize that when trading futures, commodities, options, derivatives and other financial instruments one
could lose the full balance of their account. It is also possible to lose more than the initial deposit when trading derivatives or using leverage.
All funds committed to such a trading strategy should be purely risk capital.
Information for readers in Australia: AQR Capital Management, LLC, is exempt from the requirement to hold an Australian Financial
Services License under the Corporations Act 2001, pursuant to ASIC Class Order 03/1100 as continued by ASIC Legislative
Instrument 2016/396, (as extended by instrument). AQR is regulated by the Securities and Exchange Commission ("SEC") under United
States of America laws and those laws may differ from Australian laws.
Information for readers in Canada: This material is being provided to you by AQR Capital Management, LLC, which provides
investment advisory and management services in reliance on exemptions from adviser registration requirements to Canadian residents
who qualify as “permitted clients” under applicable Canadian securities laws. No securities commission or similar authority in Canada
has reviewed this presentation or has in any way passed upon the merits of any securities referenced in this presentation and any
representation to the contrary is an offence.
Information for readers in Middle East: AQR Capital Management (Europe) LLP (DIFC Representative Office) is regulated by the Dubai
Financial Services Authority of the Dubai International Financial Centre as a Representative Office (firm reference number: F007651).
Its principal place of business is Gate Village 10, Level 3, Unit 4, DIFC, Dubai, UAE. This marketing communication is distributed on
behalf of AQR Capital Management, LLC.
Information for readers in the EEA: AQR in the European Economic Area is AQR Capital Management (Germany) GmbH, a German
limited liability company (Gesellschaft mit beschränkter Haftung; “GmbH”), with registered offices at Maximilianstrasse 13, 80539
Munich, authorized and regulated by the German Federal Financial Supervisory Authority (Bundesanstalt für
Finanzdienstleistungsaufsicht, „BaFin“), with offices at Marie-Curie-Str. 24-28, 60439, Frankfurt am Main und Graurheindorfer Str.
108, 53117 Bonn, to provide the services of investment advice (Anlageberatung) and investment broking (Anlagevermittlung) pursuant
to the German Securities Institutions Act (Wertpapierinstitutsgesetz; “WpIG”). The Complaint Handling Procedure for clients and
prospective clients of AQR in the European Economic Area can be found here: [Link]
Information for readers in the United Kingdom: The information set forth herein has been prepared and issued by AQR Capital
Management (Europe) LLP, a UK limited liability partnership with its office at Charles House 5-11, Regent St., London, SW1Y 4LR,
which is authorised and regulated by the UK Financial Conduct Authority (“FCA”).
Information for readers in APAC:
This presentation may not be copied, reproduced, republished, posted, transmitted, disclosed, distributed or disseminated, in whole or
in part, in any way without the prior written consent of AQR Capital Management (Asia) Limited (together with its affiliates, “AQR”) or as
required by applicable law. This presentation and the information contained herein are for educational and informational purposes only
and do not constitute and should not be construed as an offering of advisory services or as an invitation, inducement or offer to sell or
solicitation of an offer to buy any securities, related financial instruments or financial products in any jurisdiction. Investments
described herein will involve significant risk factors which will be set out in the offering documents for such investments and are not
described in this presentation. The information in this presentation is general only and you should refer to the final private information
memorandum for complete information. To the extent of any conflict between this presentation and the private information
memorandum, the private information memorandum shall prevail. The contents of this presentation have not been reviewed by any
regulatory authority in Hong Kong. You are advised to exercise caution and if you are in any doubt about any of the contents of this
presentation, you should obtain independent professional advice.
AQR Capital Management (Asia) Limited is licensed by the Securities and Futures Commission ("SFC") in the Hong Kong Special
Administrative Region of the People's Republic of China ("Hong Kong") pursuant to the Securities and Futures Ordinance (Cap 571) (CE
no: BHD676).
AQR Capital Management (Asia) Limited
Unit 2023, 20/F, One IFC, 1 Harbour View Street, Central Hong Kong, Hong Kong
Licensed and regulated by the Securities and Futures Commission of Hong Kong (CE no: BHD676).
China: This document does not constitute a public offer of any fund which AQR Capital Management, LLC (“AQR”) manages, whether by
sale or subscription, in the People's Republic of China (the "PRC"). Any fund that this document may relate to is not being offered or sold
directly or indirectly in the PRC to or for the benefit of, legal or natural persons of the PRC. Further, no legal or natural persons of the
PRC may directly or indirectly purchase any shares/units of any AQR managed fund without obtaining all prior PRC’s governmental
approvals that are required, whether statutorily or otherwise. Persons who come into possession of this document are required by the
issuer and its representatives to observe these restrictions.
Singapore: This document does not constitute an offer of any fund which AQR Capital Management, LLC (“AQR”) manages. Any fund
that this document may relate to and any fund related prospectus that this document may relate to has not been registered as a
prospectus with the Monetary Authority of Singapore. Accordingly, this document and any other document or material in connection
with the offer or sale, or invitation for subscription or purchase, of shares may not be circulated or distributed, nor may the shares be
offered or sold, or be made the subject of an invitation for subscription or purchase, whether directly or indirectly, to persons in
Singapore other than (i) to an institutional investor pursuant to Section 304 of the Securities and Futures Act, Chapter 289 of
Singapore (the “SFA”)) or (ii) otherwise pursuant to, and in accordance with the conditions of, any other applicable provision of the SFA.
Korea: Neither AQR Capital Management (Asia) Limited or AQR Capital Management, LLC (collectively “AQR”) is making any
representation with respect to the eligibility of any recipients of this document to acquire any interest in a related AQR fund under the
laws of Korea, including but without limitation the Foreign Exchange Transaction Act and Regulations thereunder. Any related AQR fund
has not been registered under the Financial Investment Services and Capital Markets Act of Korea, and any related fund may not be
offered, sold or delivered, or offered or sold to any person for re-offering or resale, directly or indirectly, in Korea or to any resident of
Korea except pursuant to applicable laws and regulations of Korea.
Japan: This document does not constitute an offer of any fund which AQR Capital Management, LLC (“AQR”) manages. Any fund that
this document may relate to has not been and will not be registered pursuant to Article 4, Paragraph 1 of the Financial Instruments and
Exchange Law of Japan (Law no. 25 of 1948, as amended) and, accordingly, none of the fund shares nor any interest therein may be
offered or sold, directly or indirectly, in Japan or to, or for the benefit, of any Japanese person or to others for re-offering or resale,
directly or indirectly, in Japan or to any Japanese person except under circumstances which will result in compliance with all applicable
laws, regulations and guidelines promulgated by the relevant Japanese governmental and regulatory authorities and in effect at the
relevant time. For this purpose, a “Japanese person” means any person resident in Japan, including any corporation or other entity
organised under the laws of Japan.
 AQR Capital Management, LLC
One Greenwich Plaza, Greenwich, CT 06830
p: +1.203.742.3600 I f: +1.203.742.3100 I w: [Link]