Exit Cap Assumptions
I was wondering what you guys do for exit cap assumptions and also where you might find some of the data/projections that fuel those assumptions. I am largely concerned with Multifamily underwriting here but of course would be happy to hear about all asset classes.
Curious to hear peoples thoughts on the pros and cons of the following:
1.) Flat exit cap - same as yr 1 stabilization for value add, or for core maybe forward yr 1 cap rate, with nothing else. Just an IRR matrix showing deviations in either direction, but with the mean cap selected for target exit.
2.) Same assumptions as above but with some upward slope of the cap rate to account for actual asset level decline. Buildings after all only go down in value. Maybe something like 10 bps a year.
3.) Exit caps relative to 10yr T-Bonds with the historic margin between caps for that asset class vs bonds (160~200bps?)
4.) Simple reversion to the mean for the submarket and asset class cap rates from historical data
5.) Relative to interest rates. Interesting to note that for CRE this is not the driver but if you buy SFR on the side or are underwriting small SFR portfolios consumer debt drives that market more than commercial debt since valuations are based on comps, so, that's at least one asset class I can think of that should rely on rates more.
6.) Some cool stochastic tool.
7.) Some other way I am not thinking of, ha.
For those using T-Bonds I would love to hear what source you use to pull the projections or if you alter it in any way.
I'll start.
For #1 I'll say that a flat cap seems for me anyway, to ebb and flow with the market. In a post downturn market it may be conservative as you would expect caps to maybe compress, but as the bull run lengthens it seems optimistic. You could say that for something like a 10 yr hold you would expect to almost bypass the next cap rate expansion but then why would caps normalize at where they are today? For this reason I think some other methods listed here are better. I am however sympathetic to the argument that all the other assumptions are so artificial that this is almost the more justifiable way to go, as it and least explicitly assumes less. However doesn't it implicitly assume just as much?
For #2 I think it feels somewhat arbitrary. While it's almost certainly true that your asset will decline in condition, forcing you to provide more value by reducing the rents for the same square footage in order to remain competitive, I think this can easily get washed out by other economic forces. I think more high level economic views are more valuable if the sponsor/operator has the know how to do so. As far as a combination of this and other methods, I suppose I am agnostic. On the other hand, this is the only option here relying basically on physics, so...
For #3 This makes the most sense to me. Interested to hear where people are getting their projections.
For #4 This makes the second most sense to me, but why not cut to the chase and use bonds?
For #5 I am inclined to underwrite small scale SFR with WSJ Prime Rate plus something like 200 bps, as this is what many smaller lenders or credit unions are underwriting for their resets on ARM's. For CRE why not just use bonds?
For #6 Frankly don't know how to use nor do I use any tools like this but it's an interesting idea. Would love to hear from anyone doing this.
For #7 Ditto ^
I think it is more subjective than a formula. For example, trophy core assets should experience less cap-rate movement than run of the mill deals due to scarcity. Also, long-term bond yields in the USA are substantially higher than EU/Japan. Therefore, I am not sure how much merit rate rise forecasts have based on the current dynamic.
Generally, on our typical 3-5 year hold we are underwriting 50 bps above today's market cap-rate but stress testing up to 100 bps. If cap-rates rise 200 bps in 3-5 years the whole development market is underwater.
You are over thinking exit cap rates. Remember, folks buy for different reasons so this can be subjective. A high net worth family might buy a deal that should trade at a 4.5% cap at 4.0% cap and put 20 year LIFECO money on the deal. They are less worried about the exit cap becuase it's a long term hold and they want to shelter capital than appreciation aka IRR driven. They are less worried about a 15%IRR for value add and might be ok with a 7 or 8% IRR (more of an EGO play in the 50 yard line of town). I am seeing that happen right now in a core market I am in. Also, it might be 1031 exchange funds so they can pay more.
My general rule of thump to IRR is 10 bps increase by year to market cap rate. For example, if market cap rate for comparable properties is 5% today, I will use 5.3% in year 3, 5.5% in year 5, and 5.9% in year 9. Depending on the location, I might be more aggressive or conservative based on local market dynamics, etc. No one knows where the hell cap rates are going!! It's a mix of buyer appetite (aka cost of capital), treasury rates, best use analysis and so much shit that's out of your control to "guess".
I have even seen groups use lower exit cap rates at exit in this market but it's made up numbers to make unsophisticated investors bite for syndicated deals and show CRAZY IRR. Look at some of the deals on crowdstreet (it's voodoo underwriting).
Middle of the road value add deals today are generating a 12-13% IRR in core markets like DC, chicago, arizona, seattle, etc if you use standard underwriting of 3% rent growth and expanding cap rates. The dick heads showing 20% IRR over 5 year for value add are assuming 4-6% rent growth or they haven't budgeting for renovations or cap ex items that will get them fucked on the back end.
End case, it all depends on the strategy or FAT FINGERING of the exit cap rate along with LEVERAGE. The last 5-8 years have been strange due to cap rate compression after a major recession along with insane rent growth that has made every broker that used INSANE numbers in OM look like geniuses. The market is at an inflection point, so be smart about your assumptions.
Depends a lot on market/product type. For industrial, for example, if you underwrite 100 bps spread on exit (10 year hold) in any market on the west coast, you'll be hard pressed to win any deal assuming you're an IRR driven buyer.
In general, I never compress exit caps. This should be 100% upside and stuff that isn't baked into your pro-forma. Nor should it be, since it's all based on market timing (fool's errand to project).
You're equating building value to depreciation. Would advise you not to do this. Functional obsolescence should be reflected in capital expenditures, not in your yield/exit. I.e., the value of the site/building/land itself will continue to go up. The capital systems/capital required is what will take a hit. The way to mitigate that is with a maintenance/capital plan. This will directly manifest in the rental rates.
Posted this years ago and forgot about it, but this is a great answer. I mean it’s possible that over a very long hold you might include an escalation in caps because regardless of good asset management the building may simply be less competitive because it’s just fundamentally outdated, but for a standard deal you’d be better with this protocol instead and in the prior case I’d just as soon not bother modeling it in. Kudos.
More art than science. But if the market cap rate today is 5%, year 1 I would assume 5% exit cap. In year 3 I would do 5.25. In year 5 I would do 5.5% in year 7 I would do 5.75% and year 10 I would do 6%. If rents are always under market, there is an argument someone would purchase at a below market cap rate (less than listed above).
Aut nesciunt quia vel. Aut sit et laborum nostrum. Aut vero ullam sed ut. Ipsum culpa porro ea deserunt aut praesentium magnam.
Omnis velit earum est est ipsa minus sapiente. Beatae deleniti et et dolorem voluptates consequatur.
Velit rerum laboriosam et nemo qui. Dolores maxime magni adipisci ratione. Omnis ut maxime quasi repellendus ea. Aut perferendis eos et nesciunt est.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...