Chapter 11: Real Estate Cash Flow Pro Formas & Opportunity Cost of Capital (OCC) © 2014 OnCourse Learning. All Rights Reserved.1

"PROFORMA" = a multi-year cash flow forecast (Typically 10 years.) Show to: Lenders, Investors But the proforma can be more useful than just “window dressing”, if done properly. It is the basic vehicle to implement the DCF valuation and analysis procedure discussed in the previous chapter. The CF proforma presents the numerators in the RHS of the DCF valuation equation. © 2014 OnCourse Learning. All Rights Reserved.2

Uses of multi-year DCF analysis in real estate… Estimate market value of assets CF projections should be unbiased mean mkt expctns Estimate “investment value” of assets CF projections should be unbiased mean subject investor effects (we’ll discuss in Ch.12) Sensitivity analysis & “crash testing” (What if?) Explore upside, downside plausible ranges “Underwriting” for debt finance (sometimes deliberately conservative CF projections) Ex ante performance attribution For investment strategic decision making Ex post analysis for diagnostics Performance attribution for investment management, internal accountability mgt. © 2014 OnCourse Learning. All Rights Reserved.3

2 types of CFs: Operating Reversion (Sale of Property, Sometimes partial sales) © 2014 OnCourse Learning. All Rights Reserved.4

2 ways of defining "bottom line"... 1) Property level (PBTCF, most common in practice):  Net CF produced by property, before subtracting debt svc pmts (DS) and inc. taxes.  CFs to Govt, Debt investors (mortgagees), equity owners.  CFs due purely to underlying productive physical asset, not based on financing or income tax effects.  Relatively easy to observe empirically.  Focus of Chapter 11. 2) Equity ownership after-tax level (EATCF):  Net CF avail. to equity owner after DS & taxes.  Determines value of equity only (not value to lenders).  Sensitive to financing and income tax effects.  Usually difficult to observe empirically (differs across investors).  Will be addressed in Chapter 14. © 2014 OnCourse Learning. All Rights Reserved.5

Typical proforma line items... Exhibit 11-1: At Property, Before-tax Level: Operating (all years): Potential Gross Income = (Rent*SF)= PGI - Vacancy Allowance = -(vac.rate)*(PGI)= - v + Other Income = (eg, parking, laundry) = +OI - Operating Expenses = - OE _____________________ _______ Net Operating Income = NOI - Capital Improvement Expenditures = - CI _____________________ _______ Property Before-tax Cash Flow = PBTCF Reversion (last year & yrs of partial sales only): Property Value at time of sale = V - Selling Expenses = -(eg, broker) = - SE __________________ ______ Property Before-tax Cash Flow = PBTCF © 2014 OnCourse Learning. All Rights Reserved.6

Questions… How forecast vacancy (v)? Vac = (vac months)/(vac months + rented months) in typical cycle. Look at typical vac rate in rental mkt; adjust for non-stabilized bldgs (e.g., gross vacancy in mkt typically > typical stabilized vac). History of vac. in subject bldg. Project for each space/lease: Probability of renewal & Expected vacant period if not renewed. How forecast resale value (“reversion”, V at end)? Divide Yr.11 NOI by “going-out” (terminal) cap rate. What should be the typical relationship between the going-in cap rate and the going-out cap rate?... Usually going-out  going-in (older bldgs have less growth & more risk), esp. if little capital imprvmt expdtrs have been projected. © 2014 OnCourse Learning. All Rights Reserved.7

Exhibit 11-2: NCREIF Same-Property NOI Growth vs Inflation: Gray shade indicates GDP recession © 2014 OnCourse Learning. All Rights Reserved.8

Exhibit 11-3: Average Reported Vacancy Among NCREIF Properties: Gray shade indicates GDP recession Vacancy tends to be cyclical © 2014 OnCourse Learning. All Rights Reserved.9

Exhibit 11-2: As New Competitors Enter the Market, Spread Between Building and Submarket Vacancy Increases for Older Buildings (Source: Torto-Wheaton Research; “TWR Overview &Outlook”, Winter 2004.) Vacancy tends to increase as buildings age © 2014 OnCourse Learning. All Rights Reserved.10

Operating Expenses include: Fixed:  Property Taxes  Property Insurance  Security  Management Variable:  Maintenance & Repairs  Utilities (not paid by tenants) © 2014 OnCourse Learning. All Rights Reserved.11

Operating Expenses NOTE: OE do NOT include: Income taxes, Depreciation expense. Must include mgt expense even if self-managed. Why?... Opportunity cost, “apples-to-apples” comparison with alternative investments that you don’t have to manage yourself. © 2014 OnCourse Learning. All Rights Reserved.12

Capital Expenditures include: Leasing costs:  Tenant build-outs or improvement expenditures (“TIs”)  Leasing commissions to brokers Property Improvements:  Major repairs  Replacement of major equipment  Major remodeling of building, ground & fixtures  Expansion of rentable area © 2014 OnCourse Learning. All Rights Reserved.13

Two truths often not reflected in proformas used in practice in the real world... Realistic long-term rental growth projections in most commercial properties in most areas of the U.S. should average slightly less than realistic expectations about general (CPI) inflation. Realistic long-term capital expenditure projections for most types of commercial property should average at least 10% to 20% of the NOI, or an annual average of about 1% to 2% of the property value. © 2014 OnCourse Learning. All Rights Reserved.14

Simple numerical example (in Appendix 11A: Exh.11A-1) © 2014 OnCourse Learning. All Rights Reserved.15

Section 11.2: “Opportunity Cost of Capital” (OCC) at the Property Level or: WHERE DO DISCOUNT RATES COME FROM?... © 2014 OnCourse Learning. All Rights Reserved.16

Broad Answer: THE CAPITAL MARKETS That is, competing investment opportunities. (This is so, whether we are talking about IV or MV.) © 2014 OnCourse Learning. All Rights Reserved.17

IN DCF APPLICATIONS, KEEP IN MIND WHAT THE DISCOUNT RATE IS... Disc. Rate= Required Return = Oppty. Cost of Capital = Expected total return = r = r f + RP = y + g, among investors in the market today for assets similar in risk to the property in question. © 2014 OnCourse Learning. All Rights Reserved.18

Historical Evidence about R.E. OCC in the U.S. But this particular 41-yr period may be abnormally favorable ex post for bonds & RE, and unfavorable for stocks. Traditionally large-cap stock ex ante RP considered to be ≈ bps; LTGovt Bonds ≈ bps; Institutional RE bps. © 2014 OnCourse Learning. All Rights Reserved.19

Survey avg  bps > Hist.avg Survey Evidence about R.E. OCC in the U.S. What to make of the difference between the blue and the purple lines?... Perhaps a little tinting in the shades?... © 2014 OnCourse Learning. All Rights Reserved.20

IRR – OAR = (y+g) – (y+CI) = (y+Infl-Depr) – (y+CI) = Infl – Depr – CI The brown (IRR-OAR) line should be below the green (infla) line! (probably at least bps below…) “OAR” = “cap rate”, “CI” = capex rate = Avg ann. capital expenditures as fraction of property value © 2014 OnCourse Learning. All Rights Reserved.21

How to "back out" implied discount rates from "cap rates" (OAR) observed from transaction prices in the property market... Cap rate= NOI / V = (CF+CI) / V = y + CI / V = r + CI / V – g Therefore, from market transaction data... 1) Observe prices (V) 2) Observe NOI of sold properties. 3) Therefore, observe "cap rates" = NOI / V. 4) Compute: r = cap rate – CI / V + g e.g.: Data from Real Capital Analytics, CoStar, Reis, etc. realistic ^ © 2014 OnCourse Learning. All Rights Reserved.22

Build up the mkt’s implied OCC (IRR)… IRR = y + g = (caprate – CI) + (Infl – Depr) Typically: IRR = (caprate – bps) + (Infl – bps) IRR = caprate + Infla – bps These days (for “institutional”),   IRR  caprate. (Assuming infl approx 3%) (Shh!... IRR could even be a bit < caprate!... If low inflation. But watch out: It can vary over time (e.g., infl, RE mkt), and across differ types of RE mkts (e.g. “institutional” vs “mom&pop”, Class-A vs Class-B, etc) © 2014 OnCourse Learning. All Rights Reserved.23

r = r f + RP Take the r = r f + RP approach (2006 peak)... For typical 10 yr horizon investment (2006): r f r f = Expected average short-term T-Bill yield over life of R.E. investment, well approximated by 10 yr T-Bond yld – 100 bps (“yield curve effect”). (Bond mkt’s expectation of avg future short-term T-Bill yields over the next 10 years.) r f e.g., if T-Bond yld = 5%, then r f = T-Bond yld – 150 bps = 5% - 1.5% = 3.5%. RP OCC = 3.5% + (2.5%-4%) = 6%-7.5% (or so) RP = 250 to 400 bps for “institutional” investment property (based on NCREIF historical avg, ≈ ½ Stk Mkt RP),  OCC = 3.5% + (2.5%-4%) = 6%-7.5% (or so); RP OCC = 8% - 11% RP = 500 to 700 bps for “non-institutional” investment property (smaller, higher risk, less liquid),  OCC = 8% - 11% Double Checking: Two Perspectives on the OCC Estimate © 2014 OnCourse Learning. All Rights Reserved.24

r = y + g Take the r = y + g approach (2006 peak)... y y = “cap rate” (less CapEx) = e.g., in 2006 in the U.S. this was about 4 to 6% for “institutional” investment property, more like 6% - 9% for “non-institutional” investment property. g Realistic growth rate g = Historical rental mkt growth rate – Historical inflation + Realistic projected future inflation (Bond mkt T-Bond yld – Infla-adjusted T-Bond yld “TIP”) – Property real depreciation rate (≈ 1%- 2%/yr) g Typically g = 0% to 2% in most markets. ry + g  r = y + g = e.g., in 2006 in U.S. ≈ 4% to 7% “institutional”, 6% to 11% “non-institutional”. (Remember: This is meant to be applied to property-level CFs.) Note disconnect with equilibrium r f + RP model If y + g model below r f + RP model  Current pricing is “high.” Double Checking: Two Perspectives on the OCC Estimate © 2014 OnCourse Learning. All Rights Reserved.25

r = r f + RP Take the r = r f + RP approach (today)... For typical 10 yr horizon investment (today): r f r f = Expected average short-term T-Bill yield over life of R.E. investment, well approximated by 10 yr T-Bond yld – 100 bps (“yield curve effect”). (Bond mkt’s expectation of avg future short-term T-Bill yields over the next 10 years.) r f e.g., if T-Bond yld = 2%, then r f = T-Bond yld – 150 bps = 2% - 1.5% = 0.5%. RP OCC = 0.5% + (3.5%-4.5%) = 4%-5% (or so) RP = 350 to 450 bps for “institutional” investment property (based on NCREIF historical avg, ≈ ½ Stk Mkt RP),  OCC = 0.5% + (3.5%-4.5%) = 4%-5% (or so); RP OCC = 5.5% - 7% RP = 500 to 700 bps for “non-institutional” investment property (smaller, higher risk, less liquid),  OCC = 5.5% - 7%. But these seem too low… May reflect Fed intervention (“QE”) Double Checking: Two Perspectives on the OCC Estimate © 2014 OnCourse Learning. All Rights Reserved.26

r = y + g Take the r = y + g approach (today)... y y = “cap rate” (less CapEx) = e.g., today in the U.S. this is about 6% - 8% for “institutional” investment property, more like 8% - 10% for “non-institutional” investment property (non-distressed). g Realistic growth rate g = Historical rental mkt growth rate – Historical inflation + Realistic projected future inflation (Bond mkt T-Bond yld – Infla-adjusted T-Bond yld “TIP”) – Property real depreciation rate (≈ 1%- 2%/yr) g Typically g = 0% to 2% in most markets. ry + g  r = y + g ≈ 6% to 10% “institutional”, 8% to 12% “non- institutional”. (Remember: This is meant to be applied to property-level CFs.) Note disconnect with equilibrium r f + RP model If y + g model above r f + RP model  Current pricing is “low.” Double Checking: Two Perspectives on the OCC Estimate © 2014 OnCourse Learning. All Rights Reserved.27

Watch out for terminology: In Brealey-Myers “capitalization rate” is often used to refer to “r”, the total cost of capital (especially in corporate finance). “r” is also sometimes called the “total yield” (especially in the appraisal profession). © 2014 OnCourse Learning. All Rights Reserved.28

Typical per annum OCC (“going-in IRR”) rates (late 1990s)... For high quality ("class A", "institutional quality") income property:  10% - 12%, stated.  8% - 10%, realistic. Lower quality or more risky income property (e.g., hotels, class B commercial, turnarounds, "mom & pops"):  12% - 15% Raw land (speculation):  15% - 30% © 2014 OnCourse Learning. All Rights Reserved.29

Typical per annum OCC (“going-in IRR”) rates (circa 2005)... For high quality ("class A", "institutional quality") income property:  7% - 9%, stated.  5% - 7%, realistic. Lower quality or more risky income property (e.g., hotels, class B commercial, turnarounds, "mom & pops"):  8% - 10% Raw land (speculation):  12% - 25% © 2014 OnCourse Learning. All Rights Reserved.30

Typical per annum OCC (“going-in IRR”) rates (circa 2011)... For high quality ("class A", "institutional quality") income property:  8% - 10%, stated.  6% - 8%, realistic. Lower quality or more risky income property (e.g., hotels, class B commercial, turnarounds, distressed assets, "mom & pops"):  8% - 12%, realistic Raw land (speculation):  20% - 40% © 2014 OnCourse Learning. All Rights Reserved.31

Variation in Return Expectations Across Property Types © 2014 OnCourse Learning. All Rights Reserved.32

© 2014 OnCourse Learning. All Rights Reserved.33

Note that the difference in OCC tends to be much greater between “instituional” vs “non- institutional” quality real estate ( bps), than between most usage types of property (office, retail, industrial, residential) within either of those two categories. Why do you suppose this is?... © 2014 OnCourse Learning. All Rights Reserved.34

“Institutional” (aka “Investment Grade”) properties (larger, in primary mkts) exhibit different price behavior than smaller (“mom & pop”) properties, as seen in CCRSI… Reflects different sources of financing (non-bank vs bank), different owner/investor clienteles (natl/intl instns vs local/users), different asset mkt segments. © 2014 OnCourse Learning. All Rights Reserved.35

IN DCF APPLICATIONS, KEEP IN MIND WHAT THE DISCOUNT RATE IS... Disc. Rate= Required Return = Oppty. Cost of Capital = Expected total return = r = r f + RP = y + g, among investors in the market today for assets similar in risk to the property in question. © 2014 OnCourse Learning. All Rights Reserved.36