Pagliari: Another take on real estate's role in mixed-asset portfolio allocations Flashcards
the use of annual returns often generates excessive allocations to private real estate:
1) the autocorrelation of price real estate return series understates a) the long-run volatility of such returns and b) its long-run correlation with other asset classes
2) the suggested allocations to real estate are substantially higher than the allocations generally found among large institutional investors
preference for low-risk portfolios:
(unlevered) private real estate
preference for high-risk portfolios:
public real estate (embedded leverage of 40-50%)
potential explanations why investors’ allocation to real estate might be lower than 10-15%:
inclusion of investor’s liabilities, investment management fees and cost, liquidity, information costs, ownership involvement, conflict of interest
the issue of high levels of serial correlation is often found with a variety of privately traded alternative assets
these securities often reflect very different serial correlation characteristics from their publicly traded counterparts
–> their long term portfolio enhancing characteristics may be very different from their short-term characteristics
private real estate returns are proxied by
NCREIF
public real estate returns are proxied by
NAREIT
private real estate represents (largely to its unlevered nature) the
low-risk / low-return variant of public real estate (which is levered)
three components in the standard portfolio optimization technique
- average return
- variance
- correlation
long-horizons would be the preferable prism when investors decide to
private market investments: illiquidity and transaction costs make frequent portfolio rebalancing impractical
de-smoothing models often rely upon unknowable parameters and/or difficult-to-ascertain observations
long horizons allow the data to “speak for themselves”
the longer holding period captures
serial correlation observed in t-bills and other private-market alternative investments
the use of long run investment horizons provides investors with
another perspective by which they can make judgements about their ex ante portfolio allocations
de-smoothing approaches attempt to
recover true market prices from the observed appraisal-based return series while reducing autocorrelation
“constant liquidity value”
to reflect prices that would have been observed had the ease of selling been constant over time
true returns are unobservable and reported returns represent a weighted average of current and previous true returns
reported returns are a smoothed byproduct of true returns
publicly traded REITs are an imperfect substitute for
private real estate
–> these imperfections seem fairly small when viewed from a long run perspective
NAREIT returns are reported ..
with leverage (40-50%)
NCREIF returns are reported ..
on an unleveraged basis
–> increases the return and volatility of return
NCREIF index returns are used as a proxy for the ..
returns available for non-core (value-added and opportunistic) funds
institutional investors can create a risk/return continuum of real estate investment opportunities by
varying their leverage ratio
what dominates the portfolio allocations at the low end of the efficient risk/return spectrum?
t-bills and private-market real estate
what dominates the portfolio allocations at the middle of the efficient risk/return spectrum?
treasury bonds, small stocks, and real estate
what dominates the portfolio allocations at the high end of the efficient risk/return spectrum?
small stocks
real estate occupies an average portfolio allocation approaching 25%
real estate appears as a substantial portfolio weighting in all but the most extreme edges of the high-risk/high-return efficient frontier
when the level of autocorrelation equals zero, the scaled long-term volatility equals the
periodic volatility: their ratio equals one and the volatility of long-run returns is said to be unitless
assets which display significantly different single-period autocorrelation characteristics have a long-term (cross asset) correlation which is ..
more than twice as large as their single period (cross-asset) correlation
–> their beneficial diversification characteristics may be markedly diminished
the Sharpe ratios represent the
interaction of changing means and volatilities as the holding period lengthens
the general effect of using longer investment horizons was to
increase the average Sharpe ratio –> volatilities tend to drop faster than mean returns
dramatic changes in the level of serial correlation in the observed one- and four-year returns
the two real estate return series saw their serial correlation fall by more than 50 percentage points while US equities saw their serial correlation rise by approximately 3ß percentage points
potential downfall of strict application of portfolio optimization is that
small differences in risk/return characteristics may result in unjustifiably large differences in optimized portfolio weights
“floors and ceilings” around the allocation of each asset class:
reduces the allocation to small stocks and increases the allocation to public market real estate
because of the high serial correlation associated with t-bills and private real estate (NCREIF) returns, ..
their infinite-horizon volatilities explode as compared to their one-year volatilities
the return series of private market assets generally display significant serial correlation
–> one-year returns as typically used in mixed-asset portfolio optimizations often fail to fully capture these effects
using infinite time-horizons, the optimal real estate allocations are approx.
10-15% of the mixed asset portfolio (represents upper bound for most investors)
for investors preferring a low-risk portfolio,
(unlevered) private real estate is the vehicle serving this allocation
for investors preferring a high-risk portfolio,
public real estate (embedded leverage of 40-50%) is the vehicle serving this allocation
whether investors prefer private or public real estate vehicles ought to depend on their ..
preference for liquidity, tolerance for short-term volatility, ability to exert control, etc.
