COMMERICAL REAL ESTATE

LESSONS LEARNT
THE RETURN OF THE FINANCIAL ENGINEERING DON'T
REPEAT THE SINS OF THE LAST CYCLE
JUNE 2015

The past 6-9 months has seen a significant change in real estate debt markets and, once again, financial
engineering on paper looks a tantalising proposition to enhance returns. The purpose of this paper is to
highlight the impact of various gearing scenarios on our market forecasts, highlight external risks to the
investment environment which could influence the sustainability of gearing, and review recent academic
research on the subject. The aim is to provide guidance to the best risk adjusted strategy for leverage this
cycle and the implications for portfolio construction.

Debt spread highly accretive

The chase for yield and the thawing of risk aversion has seen a noticeable improvement in the availability of commercial real
estate debt from both traditional banks and capital market sources in recent months. The increase in competition has seen
margins fall which when coupled with the general fall in interest rates has contributed to very low borrowing costs for property
investors and developers.

The all in cost of debt for well rated listed and unlisted funds now sits incredibly at around 4%, having fallen almost 100bps in 12
months. Second tier borrowers and developers are also seeing much easier conditions and debt averaging around the 5-6%
mark.

While the interest rate curve has risen in the past month, the dramatic fall over the past year has opened up a very accretive
spread to property income yields, highlighted in the chart below.

Figure 1 All in cost of debt versus property income yield

% Spread
1

-1

-2

-3

-4

2001
2003
2005
2007
2009
2011
2013
2015

Source: IPD/Bloomberg/AMP Capital Debt Advisory

The spread is now over 275bps to the positive. Rental income can well and truly cover interest repayments and there would
appear to be a substantial buffer to counterbalance a rise in interest rates or deterioration in occupancy rates if the economy
stalls. The spread is well greater than the last cycle.

Given this fact, there is growing discussion in the industry about increasing leverage once again, particularly in light of

the fact that interest rates will likely stay lower for longer this cycle, which is also AMP Capitals house view.

The purpose of this paper is to stress test this hypothesis using our house view economic and property market forecasts and
provide guidance to the best risk adjusted strategy for leverage for the operating environment ahead.

Enhanced returns in the short term.

Gearing typically amplifies property returns at both the asset and fund level, highlighted in the chart below. The chart tracks the
level of gearing of the IPD unlisted property fund index (the series inverted for illustrative reasons) against returns of the fund
index and overall direct property market.

Figure 2 Direct Property, Fund Returns and Gearing

25

-10

20

15

%Gearing

Return
10

-15

Total
0

-20
GAV

Property Fund Returns (Pre Fee GAV %)

(inverted)

-5

IPD Direct Property Returns

-10

Gearing % GAV

-15

-25

1996
1999
2002
2005
2008
2011
2014

Source: IPD/AMP Capital

With such a wide spread, conditions are conducive for positive enhancement to returns in the short term at the fund level, but also it is
going to allow investors to bid higher for assets in an ultra-competitive, undersupplied investment market.

Historically buyers have used leverage to meet client benchmarks when it has positively contributed to that goal but over time
this starts to create market imbalances as the marginal buyer effectively is pricing assets on a levered IRR basis. The levered
buyer is more exposed to the interest rate curve and refinancing risk over time. This then introduces

LESSONS LEARNT COMMERICAL REAL ESTATE

01

additional interest rate volatility into asset pricing over and

above property market, capital market and asset specific
risks. Gearing at a fund level then amplifies the trend.
Overtime this imbalance becomes obvious when equity
IRRs start moving 150-200bps apart from equity discount
rates like we saw in 2007.

Thus in this paper we have focused more on the impact on

direct market performance to illustrate the risk/return tradeoff of various gearing levels as a proxy.

individual market, the outlook for property fundamentals in

each of those markets, and the outlook for interest rates.

Our pricing models suggest market returns could be

enhanced on average by c50bps p.a over the next three
years (with the accretion reducing each year) with low to
moderate gearing (up to 20% LTV) being applied. This is
highlighted in the chart below.

Figure 3 Enhanced returns from conservative gearing

We have completed sensitivities on our property house
view forecasts to illustrate the impact of various levels of
debt on market pricing and returns. These are based on
the spread between passing yields and cost of debt by

0.6
0.9

@
LVR
0.8

0.5
0.7

%
20%

Office

Industrial

Return

0.1

0.4

0.3
Composite
Retail

Additional
0.2

Higher gearing amplifies the cycle but starts to detract

from returns in the medium term
However this benefit starts to wane as gearing rates rise
and the market moves through the cycle.

The data below (and findings of similar international

academic research detailed later in the report) clearly
conclude that high gearing delivers poor risk adjusted
returns for long term investors as volatility amplifies
substantially the higher the gearing rates used. This is
clearly borne out by the much

2014
2015
2016
2017
2018
2019

Source: AMP Capital Investment Strategy/AMP Capital Debt Advisory

The level of accretion varies by submarket, ranging from

over 70bps p.a for some of the smaller higher yielding
office and industrial markets to less than 30bps p.a in
markets such as Perth office where fundamentals are
deteriorating. The pricing models suggest slightly stronger
cap rate compression is likely moving forward (c25-50bps)
at these levels of gearing across all property sectors.

The spread slowly narrows over time as cap rates fall and
interest rates rise but there appears to be meaningful
excess returns in the next 12-18 months from using low to
moderate levels of debt. This is also being reflected in the
table following where the low to moderate gearing is
producing a slightly higher risk adjusted return against
ungeared market returns over the next 5 years.

weaker risk return ratio in the table opposite for higher

geared portfolios. The chart below highlights the excess
volatility in market cap rates under a high leverage
scenario (40-50% LTV), compared to a lower geared
scenario.

Figure 4 Impact on market pricing as gearing rises

7.5

Cap

Rate %
6.5

5.5

20% Gearing

rates start falling (the 5 year average return from this

strategy falls quite alarmingly in the year ahead).

Under this scenario market cap rates could compress up to

100bps in the next year or so compared to 25-50bps on the
more conservative scenario as leveraged IRRs are
increasingly adopted when acquiring assets.

40% Gearing

But its on the downside where this model fails. This

scenario has higher sensitivity to interest rate movements
and pricing risk because the gap between the leveraged
and equity value widens the further through the property
cycle you go. This pretty much ensures that cap rates and
pricing have to adjust back once the spread starts to
narrow, again in a more amplified way that you would
normally expect.

Thus higher leverage is producing poorer market risk return

ratios than lower gearing.

Table 1 Impact on market returns of various gearing ratios

Total Return
Total Return

% p.a
2014

% p.a

2015

Risk

2016
2017
2018
2019

(2015-2019)
(2016-2020)
Return Ratio

2020

Source: AMP Capital Property Investment Strategy

While the high leverage is adding some attractive

additional returns initially, this reverses quite quickly as cap

Minimal Gearing
7.7%

7.2%
4.3x

15-20% Gearing
8.4%
7.9%

Source: AMP Capital

The key conclusion from this analysis is that in moderation,

debt can be a positive for returns as long as it is not too
excessive to the point that asset pricing becomes clearly
unstable from an equity perspective.

4.7x

Other factors supporting conservative gearing

Economic momentum is fragile in Australia.

40-50% Gearing
8.6%

The risk here is that the sluggishness in the economy

continues and the spread that looks healthy on paper is
much narrower in reality (or eroded) because of vacancies.

6.7%
1.2x

Sub markets/assets to watch for negative impact on occupancy

rates in a protracted slowdown are the resource states,
secondary assets, suburban office assets, sub regional

LESSONS LEARNT COMMERICAL REAL ESTATE

malls, secondary industrial, and any markets which are

seeing higher levels of construction activity. These are most
at risk from a leasing/tenant retention perspective in a world
of ongoing sluggish economic momentum.

What might work well on paper today could become

unsustainable in three years time when income security
diminishes.

Interest rates could rise quicker than the Central bank

cash rate movements in the short term.

At the moment the outlook for growth and inflation is tepid

and AMP Capitals fair value indicator, as shown below,
suggests interest rates will rise slowly, only 75-100bps over
the long cycle.

02

AMP Capitals investment strategy team has developed a

fair value indicator for Australian 10 year bonds.

As highlighted in the chart below, the fair value estimate has

a strong directional relationship with actual pricing giving
confidence of accuracy of trend. It is an amalgam of real
policy interest rates (ie the Cash rate less the RBA CPI
target), real GDP growth, and implied inflation expectations.
The underlying assumptions are highlighted below.

Table 2 Australian 10 Year Bond Fair Value Index

Assumptions

Real Policy

-0.5
2.9

Average RBA

2.0

Interest Rate
Australia
statistical
2016
Fair Value 10
(Cash rate
Potential
CPI measure

4.0
0
2.8
2.2

Year Bond
less CPI
Real GDP
plus inflation

Yield

2017

target)

4.1

Growth

0.5

expectations

2.7
2.4

2018
4.3
1.0
2.7
2.5
2015
3.8

Figure 1 Actual Australian 10 Year Bond Yield vs Fair

Value

2019

8
4.5
1.5
2.7
2.5

2020
4.5
1.5
2.7
2.5

Source: AMP Capital

While the model is clearly suggesting rates will stay lower for
longer, there are two key risks on the upside to the fixed
interest/interest rate curve which could narrow the spread
(and reduce the accretion) more quickly than expected.

Firstly, current pricing is not at the fair value estimate, it is in

fact 80bps below! Thus as we saw in 2013/14, signs of
inflation and growth returning in international markets could
see the yield curve rise over the course of the next 12-18
months. In fact over the last few weeks there are early signs
of this starting to occur.

Secondly, the lower for longer outlook for rates is based on

economic growth trudging along in the 2.7-2.8% p.a range
over the next 4-5 years. If say the global economy picks up
more quickly in 2017/18, the interest rate curve will also lift,

Yield

1
Fair Value

3
0

1998
2002
2006
2011
2015

Source: Bloomberg/ABS/RBA/AMP Capital

Australia 10 Year Bond Yield

Likely to be bouts of excess volatility in investment

markets potentially causing capital surges or crunches
for real estate.

At the moment real estate is benefiting from the good side of

this volatility in the chase for yield. But this could easily
reverse if something happens in equity or bond markets.
This could create an illiquidity event for real estate, on either
the equity or debt side (or both), again exposing a high debt
strategy.

There has been academic research completed since the

GFC regarding the impact of debt on fund returns. Two key
studies were completed in 2012 by;

Baum & Kennedy on Global Real Estate Returns, and

Firstly, no-one really knows what will happen when
quantitative easing stops in Europe and Japan and interest
rates and currencies start rising. Volatility in share markets
historically causes excess volatility in real estate liquidity off
the back of the allocation effect.

Another risk around this thematic is a bond market rout, as

fixed interest investors try and liquidate deflating investments
when inflation and the yield curve starts to rise (and we are
seeing early signs of this shift). This will blow back into
higher costs and possibly reduced availability of debt for
banks and investors. Clearly a higher leveraged real estate
investor is going to be at risk of refinancing and price
adjustment if this occurs.

But there are so many other risks that could create volatility
for investment markets that again add risk to a high debt
strategy in the new normal post GFC world.

Academic Research

Farrelly & Matysiak on UK Institutional Funds

Both of these highlighted that high leverage is detracting

from risk adjusted returns over the longer term.

The Farrelly and Matysiak research clearly highlighted there

was an asymmetric impact based on an analysis of UK
institutional funds the downside risks are greater because
the debt accentuates tail risk events. This research
highlights that return is not being compensated enough in
relative terms on the positive side of the cycle to
compensate for the downside risk when things turn
negative.

LESSONS LEARNT COMMERICAL REAL ESTATE

03

Further bank and non bank regulation

Lastly, like we are seeing in the residential sector, regulators

are closely watching financial stability this cycle which could
cause some change in debt liquidity if things start to get out
of control. Again this is a risk for refinancing that could be
another potential problem for a highly leveraged investor.

Conclusion

Clearly the case for elevated gearing levels is weak but the
research clearly indicates that returns can be enhanced with
the use of sensible levels of debt at this point in the cycle.

There is also a higher probability of stronger cap rate

compression than we were expecting 6-9 months ago
because of the significant shift in debt markets in recent
months, particularly in cities such as Sydney and Melbourne
where property fundamentals are improving.

While there are still lots of wounds healing from the GFC
and a raft of externalities which should see conservatism
continue, we remain less convinced that pricing discipline
will continue the longer the chase for yield and excess
capital environment remains given the behaviour of
markets in the past in real estate and other asset classes.

With global real estate advisory firms suggesting there is a

4:1 ratio of capital to property for sale, unless there is a
slowdown in the chase for yield, the forced buyer will
inevitably start using more debt to win the deal based on
past experience. This then starts to transfer leverage risk
into market values as IRRs start shifting to a leveraged
basis.

Our view is that the market is likely to move slightly beyond

the moderately conservative gearing scenario as the highly
leveraged buyer starts to set new pricing benchmarks.
This then amplifies the cycle somewhat, highlighted in the
chart below which compares our current outlook with the
previous profile.

Figure 2 Total Return Outlook 2015-2020

Total
7
11

10

Return
9

6
8

Latest House View

2015
Previous House View

2016
2017
2018
2019
2020

Source: AMP Capital

In the next couple of years moderately conservative gearing

can enhance returns but watch closely for asset values
rising to or above historical peaks (in the absence of a
similar shift in rents), the narrowing of the spread between
debt costs and property yield, and the decline in lending
margins to very low levels. These are all leading indicators
of instability. Maintaining a bias to high quality dominant
retail assets or scarce assets will protect portfolio returns
from this volatility.

Remember while gearing does add to performance in a

rising market, the negative impacts are amplified on the
downside.

CONTACT DETAILS

For further details on Australian, New Zealand and international real estate markets www.ampcapital.com

Michael Kingcott

Tim Nation

Head of Real Estate Strategy & Research

T: +61 2 9257 1639

Head of Real Estate Capital

T: +61 2 9257 6129

E: [email protected]

E: [email protected]

Important note: While every care has been taken in the preparation of this document, AMP Capital Investors Limited (ABN 59 001 777 591, AFSL 232497) and AMP
Capital Funds Management Limited (ABN 15 159 557 721, AFSL 426455) make no representations or warranties as to the accuracy or completeness of any
statement in it including, without limitation, any forecasts. Past performance is not a reliable indicator of future performance. This document has been prepared for the
purpose of providing general information, without taking account of any particular investors objectives, financial situation or needs. An investor should, before making
any investment decisions, consider the appropriateness of the information in this document, and seek professional advice, having regard to the investors objectives,
financial situation and needs. This document is solely for the use of the party to whom it is provide

LESSONS LEARNT COMMERICAL REAL ESTATE

04