UC Davis Influence Real Estate Investment Decisions Analysis Paper How do forecasting and economic value analysis influence real estate investment decision

UC Davis Influence Real Estate Investment Decisions Analysis Paper How do forecasting and economic value analysis influence real estate investment decisions?Now that you have acquainted yourself with the One Lincoln Street development in the casebook and panel discussion, discuss the following questions:What was it that John Hynes saw in the space market, looking forward from 1999 through 2004, that made him excited about the One Lincoln Street project? What role do you think this played in his willingness to launch this project without preleasing?If you were Frank Mattson at the pension fund (which, as you recall, is supposed to be conservative in its investment policies), how would you have reacted to John Hynes’s enthusiasm about the One Lincoln Street project? What questions would you have asked Hynes? What research would you have wanted to do yourself?How “big” was the ex ante positive NPV in the project? What is the source of that large positive NPV? Is it big enough, and do you have enough confidence in it that you think the pension fund should have done what it did, which was to make an exception to its rule of not investing in development projects without preleasing? Would your answer to the question above change if you knew that the building was sold in 2004 for more than $700 million? MODULE 6 UNIT 4
Casebook: One Lincoln Street project
© 2018 MIT SA+P
All Rights Reserved
sap.mit.edu
Table of contents
1. Introduction
3
2. Statistical overview
4
3. Recent building sales
5
4. Development pipeline
5
5. Projected rental growth rate analysis
6
6. Estimating the stabilized asset OCC
7
7. DCF valuation
10
8. Conclusion and development OCC computation
13
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 2 of 15
sap.mit.edu
Learning outcome:
LO5: Evaluate how forecasting and economic value analysis influence real estate
investment decisions.
1. Introduction
From Module 2, you will recall that One Lincoln Street is a one-million-square-foot, 37-story
office tower in Downtown Boston built between 2000 and 2003. The city of Boston owned
most of the land in the form of an aging parking garage that was far below the highest best
use (HBU) of the site. The rest of the necessary site was privately owned, and the
developer had to purchase those parcels. Under political pressure to stimulate the
development of the site, the city offered its share of the land at a low price.
The project would require a total investment of close to $300 million for land purchase and
construction. The purpose of this case book is to analyze the project from a realistic and
rigorous economic perspective, based on information available to the decision-makers at
the time of the real-world case. The question is whether it made sense to invest in this
development project at that time, and, at a deeper level, how much positive NPV was in
the project. This analysis requires:
•
Estimating the value of the built property upon completion and stabilization
(lease-up).
•
Estimating the present value of the forward purchase of that asset, given that the
project was speculative (no preleasing).
•
Estimating the present value of the construction costs.
•
Computing the net present value (NPV) of the construction project as of the
decision time, as the difference between the above two values.
•
Comparing that construction NPV to the purchase cost of the land to arrive at the
overall NPV of the real estate development project as of the investment decision
time.
You are guided through this process in this casebook, using the principles and tools
introduced in all of the preceding modules of this course. Modules 1 and 3 presented the
basics of investment risk and return and DCF valuation, including the opportunity cost of
capital (OCC) and the NPV decision rule. Module 2 presented the basics of the space
market, relevant for forecasting the rents and occupancy the One Lincoln Street building
could achieve. Module 4 introduced you to leverage, which is relevant in the One Lincoln
Street project not only when thinking about financial leverage implied by the use of a
construction loan in the project, but also to understand the operational leverage inherent
in the development project as elaborated on in this module. The material in Module 5 does
not surface directly in the One Lincoln Street Case, but portfolio theory no doubt underlies
the motivation of Midwest Teachers, the pension fund that served as the equity “money
partner,” to invest in real estate assets such as the One Lincoln Street project. If it weren’t
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 3 of 15
sap.mit.edu
for the portfolio theory discussed in Module 5, the One Lincoln Street developer might not
have been able to find the necessary equity capital to make the project possible.
In the preceding modules of this program, the relevant principles and methodologies were
often presented on a somewhat simplified and abstract level. This was hopefully helpful for
your learning purposes. But, in the real world, the situation is often more complex, and
involves additional considerations not yet explicitly treated in this program. The One
Lincoln Street case study is an opportunity for you to see how the basic principles and tools
presented in this program can be applied in a realistic investment decision and analysis
situation.
As noted, the first and most crucial focus of the analysis is on the basic question of whether
the project presented the external equity investor, Midwest Teachers pension fund, with a
good investment; that is, whether it had a positive NPV. Still, it is worthwhile to try to go
beyond just a yes/no answer to this question, by quantifying the magnitude (the dollar
value) of the NPV.
This level of quantification can be useful in various ways. First (and at a more preliminary
phase), it is the basis for determining the HBU of the site and for guiding the design of the
project. Secondly, it can give an idea of how strongly the equity investor might want to
pursue the project and – related to that – how the money partner might think about its
objectives, terms, and position in negotiating the joint venture (JV) agreement with the
developer. Thirdly, as described previously in Unit 3, quantifying the NPV enables you to
compute the OCC of the project, which in turn enables you to quantify the amount of
investment risk in it, relative to investments in similar stabilized properties. This can help
investors, especially external equity investors, to make a more educated decision about
investing in the project, as risk and return are the fundamental investment considerations.
Turn now to the basic quantification of the One Lincoln Street project analysis, focusing
first on the most fundamental and important issues, the economic rationale for this project
– the value of the asset to be built (also referred to as the “underlying asset”).
2. Statistical overview
Recall from your knowledge of the real estate system (covered in Module 2) that the
fundamental basis of all real estate value is in the space market, or the value of the property
to its users. Therefore, this is where you need to start in the analysis of the One Lincoln
Street project.
Exhibit 1 in the case study contains the history of the space market relevant for this project,
the Boston office market, as of and up to the time of the investment decision for the project.
Exhibit 1 presents the history of construction, net absorption, rents, and occupancy in the
market, i.e., all the key variables discussed in Module 2 of this course. Notice that the rents
and vacancy exhibit a strongly cyclical pattern. This information was available to market
participants at the time of the case.
Studying this history makes it clear that the space market was already long past and far
above its most recent trough. Decision-makers at the time could not have known for sure,
but in fact the space market was nearing a peak.
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 4 of 15
sap.mit.edu
3. Recent building sales
Developers often do not focus much on economic statistics. So, instead of focusing on the
history of the rental market, the developer might have focused more attention on the data
available regarding recent building sales in the asset market, as shown in Case Exhibit 2.
Exhibit 2 shows that recent sales of office buildings similar to the one that would be built
had occurred at going-in cap rates in the mid-7% range, averaging 7.43%. This depicts the
asset market, not the space market, as it focuses on cap rates. At the time, this seemed
like a very low cap rate, i.e., a very high price being commanded by office buildings in
Boston in the asset market. Cap rates had not been this low (property prices had not been
this high) for 10 years, and had never remained at such a level for very long.
In other words, both the space market and the asset market were rather late in their
up-cycles, perhaps nearing a peak. However, the developer, John Hynes, had an insightful
– perhaps slightly optimistic – view of the situation.
4. Development pipeline
Recall from Module 2 how important it is for real estate investors, and particularly for
developers, to forecast the space market, by considering both demand and supply as they
may play out over the near to medium-term future. Case Exhibit 3 shows the development
pipeline of Class A office buildings, which are other new or potential buildings that might
compete directly with the proposed project. Exhibit 3 represents the type of forecast of the
space market noted in Module 2 that is crucial for careful, sure investment decisions. It is
the type of information that was available to John Hynes at the time of the project.
Note that most of the projects that were already underway were mostly already leased.
Hynes may or may not have done a formal “Months Supply” (or “Months Excess Supply”)
quantification as described in Module 2. Nevertheless, Hynes’ familiarity with the Boston
property market and political and regulatory environment led him to have confidence that
other potentially competing projects would have a hard time getting finished before the
proposed project, provided it could start construction right away. This would give the
proposed project a “first mover advantage,” landing in the market at a projected gap period
between demand and supply.
However, the proposed project had no preleasing. If it was to start right away, it would have
to be a purely speculative project. This would make it riskier than if it had some preleasing
agreements in place. In effect, prior to the project’s stabilization upon completion of the
lease-up phase of the project (as described in Unit 1 of this module), the asset to be built
would present an OCC a bit higher than that of a stabilized asset. As suggested in Unit 1,
you might add up to 200 basis-points to the OCC that the asset would have once it became
leased-up and stabilized.
Pension funds are usually somewhat conservative investment institutions and must obey
a prudent investor rule according to government regulations. Therefore, the money partner
in the project (the pension fund supplying most of the equity capital), could not take on too
much risk. However, the pension fund felt that this was an attractive opportunity, and it had
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 5 of 15
sap.mit.edu
a lot of trust in the developer, Hynes, who had a long history in Boston’s commercial real
estate market. Furthermore, the pension fund was bullish on Boston as a place to invest
for the long run.
They decided to go ahead, setting up a joint venture equity partnership, with the pension
fund putting up 90% of the equity capital, and the developer or manager putting up the
other 10%. The equity capital altogether would be about 50% of the total project cost
(including the land purchase), and that equity capital would go in first. The rest of the capital
would come from a commercial bank in the form of a construction loan mortgage. The loan
would have seniority and would be paid back before the equity would be entitled to any
returns.
Thus, in addition to the inherent operational leverage in the project, the equity JV would
include some financial leverage as well. However, because the loan would essentially be
a zero-NPV transaction in itself, the financial leverage would not affect the NPV of the
investment. To simplify the analysis and focus on the fundamental economics, you can
focus on the asset or project level, ignoring the effect of the debt financing at this stage.
5. Projected rental growth rate analysis
Now, think about what you learned about projecting rents back in Modules 2 and 3. Figure 1
shows the relevant historical data that was available to the decision-makers at the time
(taken from the Case Exhibit 1). This data could have been used to compute the trend rate
shown at the bottom of this computation procedure.
Figure 1: Projected rental growth rate analysis.
First, the long-term average real rent growth rate in the space market is computed by
subtracting inflation from the nominal market rent history (over the same period of time,
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 6 of 15
sap.mit.edu
which should be adjusted for the cycle – that is, trough to trough, peak to peak, or midcycle to mid-cycle). For Class A rents, this gives a real annual growth rate of about negative
0.6%.
Second, note that the ratio of Class B to Class A rents tends to be roughly two thirds
(actually 64% in this case). Class A buildings are new, prime properties such as the One
Lincoln Street building will be when it is new. Class B buildings are older, less prized by
tenants and investors, typically what new buildings depreciate to after 50 years or so
(without major renovation). In the absence of real growth in land value, this depreciation
effect would tend to depreciate the rents a new building could charge by a rate of about
0.9% per year (of age), to reduce the rent to 64% of its new Class A level over a period of
50 years, as indicated by the following calculation:
0.641/50 ? 1 = ?0.009 ? (1 ? 0.009)50 = 0.64
This suggests that, in the long run (based on the secular trend), you would project a
building’s rent to grow at a real rate of:
Market trend rate ? Building depreciation rate
= ?0.6% + (?0.9%)
= ?1.5% per year
A reasonable projection for future inflation at the time of the project would have been 2.5%
per year. Adding this to the above real rate, you get a realistic projected long-term secular
trend nominal growth rate for the building’s rents of: Nominal = Real + Inflation = ?1.5%
+ 2.5% = 1.0% per year, ignoring consideration of the market cycle.
This is fundamental information to help you make a realistic projection of the stabilized
property’s future cash flows, without the typical optimistic bias (“rose-tinted glasses”), as
discussed in Unit 1 of Module 3. This information about long-term growth expectations also
helps you to make a rigorous estimate of the OCC of the stabilized property asset, as
discussed in Unit 2 of Module 3.
6. Estimating the stabilized asset OCC
Given a realistic projected growth rate for the building’s cash flows, the next step in
conducting a discounted cash flow (DCF) valuation estimate of the asset to be built is to
estimate the realistic opportunity cost of capital (OCC), to use as the discount rate in the
DCF valuation. Recall from Module 3 Unit 2 that you can do this in two ways: the “y + g”
approach and the “rf + RP” approach.
We see from Exhibit 2 that the current prevailing cap rate for buildings such as One Lincoln
Street is around 7.5%. However, recall that the cap rate is not exactly the “y” cash income
yield rate in the y + g = r formula, as the cap rate is based on the net operating income
(NOI), whereas you need to net out the typical level of capital expenditures (CapEx):
Cap rate = NOI ÷ V
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 7 of 15
sap.mit.edu
y = CF ÷ V
CF = NOI ? CapEx
Typically, CapEx equals 1.5% of property value per year:
y = CF ÷ V
= (NOI ? Capex) ÷ V
= Cap rate ? Capex ÷ V
= 7.5% ? 1.5%
= 6.0%
Here, you complete the y + g estimation by combining the y = 6.0% rate with the g = 1.0%
rate computed previously:
OCC =
r=y+g
= 6.0% + 1.0%
= 7.0%
Is this also consistent with the rf + RP perspective on the OCC?
The risk-free rate in 1999 might have been viewed as equal to the Treasury-bill yield at
that time, which was 4.5%. So, 7.0% will indeed equal rf + RP if an appropriate risk
premium for an investment such as the new stabilized One Lincoln Street office building
would be about 2.5% (250 basis-points).
Figure 2 shows empirical evidence for the actual total investment returns achieved by
properties similar to what the One Lincoln Street project would be once it was stabilized
(leased-up). This is the history of the National Council of Real Estate Investment
Fiduciaries’ Property Index, also known as the NCREIF Property Index or NPI, as of 1999.
This is an index of properties owned by pension funds, i.e., investors similar to the money
partner in the One Lincoln Street project, which typically invested in assets similar to what
One Lincoln Street would be once stabilized. Comparing the NCREIF total returns
(r = rf + RP) to the risk-free Treasury-bill yield over the history preceding the 1999 decision
date, you can see that the average achieved risk premium had indeed been 2.5% since
the inception of the NCREIF Property Index in 1978.
Note:
The inflation is in both the rf rate and the total return “r” rate; hence, it cancels out in the
RP = r – rf derivation. As a result, this is robust to changes in inflation.
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 8 of 15
sap.mit.edu
Figure 2: NCREIF and Treasury bill cumulative total returns.
In fact, 250 basis-points is a bit low for a real estate asset risk premium, judging by the
longer-run historical evidence and traditional assumptions in the asset market. As noted,
400 basis-points might be a more typical real estate risk premium assumption. On the other
hand, in 1999, the yield curve was nearly flat, i.e., long-term government bonds were also
yielding around 4.5%. If you remove the typical 150 basis-points from that long-term bond
yield to adjust for the yield curve (interest rate risk premium), you arrive at a risk-free rate
estimate of rf = 3.0% instead of 4.5%. Combined with the 400 basis-point RP, this again
gives you:
r = rf + RP
= 3.0% + 4.0%
= 7.0%
Thus, whether by the y + g or the rf + RP approach, you get a good 7.0% estimate of the
OCC for the to-be-built One Lincoln Street building (once stabilized, that is, once fully
leased-up).
Interestingly, this realistic OCC, based on actual empirical evidence and rigorous financial
economic reasoning, is 300 basis-points below the discount rate employed by the staff of
the pension fund to analyze the investment at the time, which was 10%.
© 2018 MIT SA+P
All Rights Reserved
Tel: +1 224 249 3522 | Email: info@getsmarter.com | Website: getsmarter.com
Page 9 of 15
sap.mit.edu
7. DCF valuation
Figure 3 shows the DCF valuation of the project’s economic NPV as an investment.
Figure 3: The net present value calculation of the One Lincoln Street project.
First, we started with the pension fund staff’s projection of future cash flows in the stabilized
building. The stabilized phase of the investment is projected in the years beginning with
2004, which was four years in the future at the time of the real estate development
investment decision.
The staff’s original projection for these stabilized cash flows indicated a projected average
annual growth rate of almost 3% per year. But that was a “rose-tinted-glasses” projection.
The analysis of the space market suggests that the realistic long-term trend growth rate
projection should have been closer to 1% per year. Thus, the original projection is adjusted
to close to 1% per year average annual growth, reflecting a “haircut” of the optimistically
biased 3% per year growth assumption. What you see here is a more realistic projection
of cash flows beyond year 2003.
Then, apply the 7% estimate of the realistic stabilized OCC as the discount rate to these
more realistic cash flows, to arrive at a projected present value of the to-be-built asset as
of stabilization at the end of year 2003 equal to $493,643,000. This is the projected
“forward-purchase” value of the asset as of that future time (“VT” in the nomenclature
introduced in this module).
At the analysis time, in late-1999 (“Time 0”), this would have seemed like a rather
aggressive valuation. The pension fund staff wa…
Purchase answer to see full
attachment