May 23, 2018 - 7:01pm EST by
2018 2019
Price: 41.40 EPS 0 0
Shares Out. (in M): 55 P/E 0 0
Market Cap (in $M): 2,310 P/FCF 0 0
Net Debt (in $M): 961 EBIT 0 0
TEV (in $M): 3,519 TEV/EBIT 0 0

Sign up for free guest access to view investment idea with a 45 days delay.

  • Warren Buffett Personal Account




We have been meaning to write this for a while and with SRG up some 10% in the last few days we may have dragged our feet for too long but we’re here to tell you that you haven’t completely missed the boat.  In our estimation, SRG is conservatively worth ~$50, with longer term upside potential of $60-70 in a few years. There are a couple ways to value SRG but our favorite is a SOTP (sum of the properties) that requires relatively few assumptions and can be assembled from public information.  Skip to the bottom to see this SOTP valuation. We hope that others will find this method useful and more reliable than a traditional DCF which requires a great deal of assumptions on the pace of Sears lease terminations, the pace of redevelopment, and redevelopment economics.


Perhaps more importantly there are a couple good catalysts coming up in the next quarter or so that we believe will take the stock price there and beyond.  So far, Seritage has been doing almost all retail-to-retail conversions. In the next 6 months we believe SRG will announce 1-3 major retail-to-residential mixed-use projects, which should be much more profitable than the retail-to-retail projects and cause a step change in perception of the company as the “story” changes.  


The Basics


There are a number of good write-ups on Seritage so I’ll keep this brief.  Basically they own a lot of the land that many Sears and Kmarts sit on. Sears and Kmart aren’t really viable retail concepts and they’re paying low rents so the idea is to redevelop all the properties to be occupied by other tenants paying higher rents such that at the end state they’re a diversified REIT with few if any Sears/Kmarts as tenants.  The questions are all around how fast they can do this and how much does it take to do it and how much rent will they be getting when they’re done. I believe the process is best illustrated by a diagram showing the pathways by which this process can go forward. (Some have also questioned how SRG will fund all this redevelopment but the recent JVs have probably done a convincing job of showing how this will be done and are probably why the stock has rallied in recent days.)



The truth can be a little bit more complicated because often SRG just recaptures ½ of the Sears box, and then the other half can later also be recaptured or maybe gets terminated and Sears vacates… also a property can be outright sold or JV’d at any time… but you get the basics.  


Note that both because SHLD is likely to go bankrupt one day, and because once a property is done being re-developed the rent is much higher, we want this process to go as fast as possible.  Particularly we want to see properties being “recaptured” because this means that a redevelopment project is imminent… why else would SRG kick out their rent-paying tenant unless they were about to do something with the space?  Indeed, on the “recaptured” side, properties don’t lie fallow for long. We don’t want properties to be “terminated” because then the SHLD rent-paying tenant leaves and another one isn’t necessarily waiting to come in. Indeed, on the “terminated” path above, some of the properties have been “lying fallow” (i.e. vacant) for numerous quarters.


What’s been going on lately?

SRG actually couldn’t start terminating their leases until 2017, but since then they’ve been terminating them.  (They are limited to only terminating 20% of rent each year)


Terminated Gross Leasable Area (GLA) by quarter.  Shows future quarters because SHLD has to give notice so we know ahead of time:

It’s not part of the core thesis, but from a timing perspective it is nice that no leases will be terminated in the upcoming Q2.


Recaptured GLA by quarter:

This pace of recaptures has picked up lately, which is great.  Also note that in the latest quarter they did all 100% store recaptures (as opposed to just recapturing half the store), which is a good sign of increased pace and lower ultimate reliance on SHLD as a tenant.


So we have a lot of properties moving through the flow diagram above, which is a good thing (at least on the recapture side) but when properties are “lying fallow” or “in redevelopment” they don’t product any rent.  Hence, all the current metrics are deteriorating, which I believe is a large part of why, until recently, the stock had been declining.



Funds from operations has been falling in 2017, before this it had been erratic but without a clear downward trend.


Company FFO (a sort of Adjusted FFO):

Again, falling in 2017 and relatively stable before.


I think most would agree that these aren’t the right measures to focus on, because the more properties SRG redevelops, the worse things look but the better they really are… but I would venture to guess that many investors indeed focus on these numbers because they’re what you normally focus on for a REIT.


The most forward-looking number the company publishes is “Annualized Total NOI” which counts rent on Signed-Not-Opened leases which aren’t paying yet.  The stock really sold off when this started trending down a few quarters ago.


Annualized Total NOI:

In a way this indicates that SRG isn’t signing enough new leases to offset the rent from SHLD tenants leaving.  But it’s also not really true because 1) properties “in redevelopment” are not fully leased up before they are completed, meaning that more 3rd party leases are coming and more interestingly, 2) there is a significant amount of property that is “lying fallow” on the recapture side, meaning that there is a growing amount of property which does not have anyone paying rent right now, but will soon be the subject of a profitable redevelopment project.


“Recaptured GLA that is currently ‘Lying Fallow’”.  You can create this figure by comparing the list of recaptured properties with the list of projects and then looking up the GLA for each recaptured store that doesn’t have a project in the 10K.

As you can see this is a relatively new phenomenon.  Previously all, and still most, recaptured properties went/go straight from occupied into the official project list, skipping the fallow state all together.  What do these properties that languish for a quarter or two “lying fallow” have in common? They’re the biggest most ambitious projects on the most valuable land.  We also believe they’re on the verge of redevelopment.


Properties currently lying fallow as a result of being recaptured but not having an announced project yet:

  • Valley View (i.e. Dallas Midtown), TX.  Fallow since Q2 2017 (by far the longest stay in fallow state for a recaptured property).  They are planning an absolutely massive mixed use project here. Fun fact: this property is part of Dallas’ bid for AMZN HQ2.
  • Redmond (i.e. Overlake i.e. Seattle).  Partially fallow since Q3 2017, fully fallow Q1 2018.  They are planning a big fancy mixed use development here too.
  • Orland Park, IL. Fallow since Q4 2017.  Not sure what they’re planning.
  • Hicksville, NY.   Fallow since Q4 2017.  They’re planning a bunch of multifamily residential over retail, right on the LIRR less than an hour from Manhattan.
  • Westminster, CA.  Fallow since Q4 2017.  Have reason to believe they’re planning multi-family residential here too but no hard evidence
  • Boca Raton, FL.  Fallow since Q4 2017.  They’re planning upscale small shop and retail.
  • Clearwater, FL.  Fallow since Q1 2018.  Not sure what they’re planning.
  • Chicago, IL (4730 West Irving Park Road).  Fallow since Q1 2018. Not sure what they’re planning.
  • Asheville, NC.  Fallow since Q1 2018.  Not sure what they’re planning.


Key Insights

Key Insight #1: the most value will be created when Seritage is able to repurpose property from retail to residential.  We have been told that this will be true on a project by project basis by speaking with many real estate professionals.  We furthermore believe that this will be true for SRG as a stock as well. So far, almost 100% of the projects have been retail to retail and none have been retail to residential (there have been two projects that incorporate office space).  We believe the stock price will be at an inflection point when SRG demonstrates that it can perform some retail to residential. We believe this is imminent.


Key Insight #2: SRG hasn’t pursued the best projects first, they’ve pursued the easiest projects first.  Bears often say you can’t extrapolate project results to the entire SRG portfolio because of differing quality, and that’s true, but it doesn’t mean that the best projects have already been done.  As stated above, we believe the best projects require substantial rezoning and approval and are yet to be started. We believe their inception is imminent.



The catalyst will come when SRG fully announces a major mixed use project (Hicksville, Redmond, and especially Valley View).  This will be good both because we expect the return numbers to be excellent (will show up in their “Total Projected Project Income” figure), and because it will change the “story” around SRG such that it won’t be perceived as just another mall REIT anymore.  These projects are likely to be paired with a JV announcement as well since they require so much capital to complete. As we have seen in recent days, these JV announcements serve as catalysts on their own because they both value the individual projects as well as demonstrate how SRG intends to finance the substantial project expenditures.


Why do we believe these projects will be announced soon?  Again, this is because SRG has already voluntarily decided to “recapture” the space, kicking out the rent-paying SHLD tenant, and now the property is “lying fallow”.  Why would they do this unless they were ready to get going? Most recaptured properties don’t lie fallow at all. Mega retail to retail projects La Jolla UTC and Aventura both laid fallow for two quarters before being formally announced.


I believe these 3 projects are enough to move the needle:

  • Valley View (Dallas).  Development planned since 2012 (remember what we said about best vs easiest?).  Full plan calls for 800 residential units, 200 hotel keys, 200k sqft retail, 1MM sqft office in two towers.  Previously a 229k sqft detached Sears so we are talking major densification. This is the whale that’s sitting out there.  We believe it would already have been underway by now were it not for some permitting delays. If they announce a JV on this development I expect the stock to go to $50 in a flash.  Be cautioned though, when they announce it they will probably just announce “phase 1” i.e. and so the project won’t look as big as above on first announcement.
  • Redmond (Seattle).  Full plan calls for 500 residential units, 210 hotel keys, 186k sqft retail, 267k sqft office.  Formerly a 255k sqft detached Sears. They just kicked out their Sears tenant who will take 90 days to vacate.
  • Hicksville.  Full plan calls for 596 residential units and 240k retail space. Formerly a 332k sqft detached Sears.  Sears vacated in April.


Sum Of The Properties Valuation Methodology

Ok that’s great we probably have some catalysts to get us where we’re going but where is that?  What’s it worth? We find doing a DCF too difficult as it relies too much on all sorts of assumptions on project economics and redevelopment pace which we don’t feel comfortable making accurately.  We can do an SOTP with relatively few controversial assumptions.


We value the properties by dividing them into two main buckets: those that have a redevelopment project announced and those that don’t.  To this we will add individual estimates for the retail to residential mixed use “fallow” projects discussed above and a value for the JV properties.  The updated result is a value of $47.28 per SRG share, which we believe is actually conservative for a few reasons we’ll discuss. In the absence of new information we wouldn’t sell before $50.  Review of methodology below. Skip to the bottom to see the figures.


Project Properties

SRG actually releases the total “Projected Annual Income” for all of their wholly owned development projects since inception.  This number is only available in aggregate. This of course requires one to take SRG at their word for what kind of income their projects will generate, but we figure their assumptions are better than ours.  Think of this as what the projects will be earning in a few years. We multiply this number by the average EV/Rev for GGP, SPG, and MAC, assuming that the completed redevelopment projects are of similar quality.  This number is currently 14.2, which is more conservative than the 17 multiplier SRG uses in their presentations.


Of course, it will take a lot of money to get there, so we subtract out the projected project costs, and then add back the project costs already spent, which is helpfully available in the 10K.


When projects properties are sold, they are not removed from this project list, so we have to subtract them out to avoid counting something that is not there.  So far this only applies to Santa Monica (with its recent JV) and Wayne, NJ (contributed to the GGP JV in 2017).


Non Project Properties

This includes both “Occupied by SHLD” properties and “lying fallow” properties.  Here we will use the individual property by property appraisal data in the July 2015 CMBS document, which gives no credit for redevelopment potential.  Every wholly owned property was assigned two appraisal values in July 2015: an “as-is” appraisal by Cushman & Wakefield which assumes SHLD stays as tenant, and a Hilco “Dark” estimate which assumes SHLD tenants vacate immediately.  For conservatism, we chose the lowest of each of these two values for each individual property. We then add up all the values for these properties, not including any that have been sold, have an announced project associated with them, or are dealt with via our individual estimates.  However, we assume that these are lower quality properties and their values might have fallen since July 2015. Therefore we adjust the final number down by the percentage that the EVs of lower quality mall REITs (KIM, DDR, KRG, BRX) have declined since then.


It’s worth emphasizing that for each of these properties we are valuing them at the lower of the appraisals with both SHLD as a tenant and not as a tenant.  Thus, this valuation does not depend on SHLD remaining a tenant.  In fact, it sort of represents a worst case scenario where SHLD declares bankruptcy and then can “shatter” the master lease into a bunch of individual leases, the worst of which they walk away from and the best of which they keep (of course what is “best” for SHLD is probably worst for SRG).


Value in JV Properties

SRG reports the NOI of unconsolidated joint ventures so we annualize the last quarter (because some were added midway through Q4 2017) and multiply this number by weighted average of the EV/NOI multiple for GGP, SGP, and MAC.


Note that since we are using NOI to value these properties, the value for any property not currently producing income will be missed.  That means we are conservatively ignoring the GLA associated with redevelopment on 3 of the 23 JV properties. It also means that we are ignoring the Santa Monica JV (though we are accounting for the remaining 50% interest in this property under the “Project Property” section).


Individual Project Valuations

So far we’ve been limiting our assumptions, but here we’ll be making them as we try to value the 3 pending mega mixed use projects: Valley View, Redmond, and Hicksville.  We spoke with numerous real estate professionals and online resources such as CBRE reports for regional construction and rent costs. Most useful was a long discussion with an associate who works with a large and well-recognized developer, who has a lot of experience in projects of similar size and scope.  The only constant we received in all these consultations was that “it depends” on the specific property, but we have attempted to synthesize the different estimates into a cohesive estimate of value. You can argue with the inputs, and we invite you to play with your own assumptions.


Our method was to add up the annual rent for each type of space that is planned to be created for each property, and then divide by what we were informed were likely cap rates based on current conditions to get a final value, which we then discounted by 30% to account for the reality that these won’t be completed for some time (4-5 years most likely).  From that we subtracted our cost estimates.. Here are the input assumptions for Redmond, which ended up not being that different than those for Valley View or Hicksville. The result is a big number. Valley View is bigger. Most of the value is in these mixed use densification projects. Little reason to own the stock if you don’t believe in these projects.


That may be hard to read so here are the key assumptions:

Residential: 500 units, 920 sqft per unit, $350 cost PSF, $41.19 rent PSF, 4% cap rate

Hotel: 210 units, assuming upper upscale, $325k per key, value $450 per key

Retail: 185,800 sqft, $367.50 cost PSF, $54 rent PSF, 6.5% cap rate

Office: 266,800 sqft, $300 cost PSF, $42 rent PSF, 6% cap rate


Here is the city of Redmond’s website for the development:


In addition to the fact that SRG has recently served a recapture notice to their Sears tenant (in Q1 2018), there was recently another sign that this project is going forward:  Seritage’s Q1 2018 earning release noted that they had formed “an agreement with a residential developer to pursue a multi-family development [on the property]... The agreement values the 2.5 [of 15] acre property at approximately $16 million.”


Note that while our individual cost and value estimates may turn out not to be conservative, we are being somewhat conservative in that we are only estimating 3 projects in this fashion, when in reality there are likely to be more residential mixed use projects in the future.


What is conservative about this method?

This valuation methodology gives credit for all projects currently announced and those projected for Hicksville, Valley View, and Redmond, but ignores all other further development potential.  


This method ignores value-creating redevelopment potential at 158 of 225 wholly owned properties.  If, on average, these properties were redeveloped and with economics along the lines of the other current projects (final valuation of which we calculate at 127% of appraised value) it would represent an additional $10 per share.  Some properties may fail to achieve 127% of their original appraised value, but as stated earlier we believe some of the best projects are yet to come. While the projects resulting in 127% of appraised value (net of costs) may not be representative of the whole portfolio, to date Seritage has not done a single real estate transaction that values any property below its appraisal value.  See appendix for a list of these transactions. Additionally, Seritage has yet to complete (or even formally announce as a project) a single retail-to-residential project. However, we know some are in the pipeline, including a multifamily development in Newark, CA for which they just announced a partnership in the Q1 2018 press release.


Dozens of the 68 projects were only partial recapture projects, meaning that they have further redevelopment potential because SHLD is still yet to be removed and replaced with a higher paying tenant.


Further redevelopment potential remains even on full recapture projects like Aventura and Santa Monica. Both of these projects are just “phase 1”, and further development is planned.


As mentioned earlier, this method should be robust to a Sears bankruptcy as the estimate for non-project properties explicitly assumes a scenario where SHLD vacates as a tenant.


Final Sum Of The Properties Valuation


Risk of Sears Bankruptcy

Basically it has been our opinion for some time that Sears CEO Eddie Lampert can keep pulling rabbits out of the Sears hat (by selling assets) to keep SHLD alive for the foreseeable future, but that it eventually will go under.  People have been predicting imminent bankruptcy for years but to date the asset sales have clearly sufficed. As an aside we think what he is doing with Sears only makes sense in the context of Seritage: even though SHLD is ultimately doomed he wants to keep it on life support and paying rent to SRG long enough to allow SRG to thrive.  In any case, the recent news where Eddie basically offered to buy all of SHLD’s remaining good assets for a few billion shows that Eddie is committed to funding SHLD himself for a while if need be. The bonds have responded by rallying. We think they go BK at the end of 2019 at the earliest, by when SRG will have completed many of its current projects and will be well on its way to self sufficiency.  


Some have suggested that this will be a boon to SRG; we disagree. SRG is already re-purposing the portfolio as fast as they can and they have their hands full.  We do expect them to weather a storm and at least the large overhang will be cleared up from an investor perspective. The stock might even pick up some sell side coverage after such a risk has come and passed.




Appendix - Property Transactions and Valuation Versus Hilco Dark Appraisal


  • In Q3 2017 SRG sold 5 properties into the GGP JV for $57.45, valuing the entire portfolio at $115MM.  The Hilco Dark appraisal for these properties was $101MM, so they were sold at a 14% premium to this estimate.
  • In Q1 2018 SRG sold 4 vacant former Kmarts for $13.5MM.  The Hilco Dark appraisal for these properties was $9.3MM so they were sold at a 45% premium to this estimate.  This transaction is notable because it shows there is potentially still value in vacant properties where SHLD has terminated their lease.
  • In Q1 2018 SRG sold a 50% interest in their Santa Monica property for $50MM, valuing the total at $100MM.  Though SRG had likely put some money into the property, this represents a 33% premium to the Hilco Dark appraisal.
  • The Q1 2018 earnings release had information on 3 transactions occurring in Q2:
    • Redmond had a transaction valuing 2.5 acres of a 15 acre property at $16MM, which is a valuation of $96MM applied to the whole area.  The Hilco Dark appraisal was $11.3MM so this portion appears to be sold at a 734% premium to this estimate. This property had just been served a recapture notice.
    • Newark, CA had a transaction valuing 4.5 acres of a 10 acres property at $20MM, which translates to $44.4 for the whole property.  The Hilco Dark appraisal was $10MM so this portion appears to be sold at a 344% premium to this estimate. This property had not yet been recaptured or terminated.
    • San Bruno CA was sold to an adjacent mall owner for $42MM.  The Hilco Dark appraisal was $20MM so the property has been sold for 400% of appraised value.  This property had not yet been recaptured or terminated.
  • In Q2 2018, per a recent press release, SRG sold 50% of West Hartford (Corbin’s Corner) for $23MM.  This property had been under redevelopment for some time, so the comparison isn’t apples to apples, but the implied $46MM for the property as a whole represents a 100% premium to the Hilco Dark appraisal.  This property had been recaptured and was almost finished with redevelopment.
  • In Q2 2018, per a recent press release, SRG sold 50% of San Diego/La Jolla for $44MM.  The $88MM valuation for the whole property represents a 60% premium to the Hilco Dark appraisal.  This property had been recaptured and had recently begun redevelopment.
  • For context: the entire company is now trading at a EV of about a 45% premium to the Hilco Dark Appraisal
I do not hold a position with the issuer such as employment, directorship, or consultancy.
I and/or others I advise hold a material investment in the issuer's securities.


Discussed above.  Announcement of residential mixed use projects and the increase in total projected annual income for redevelopment projects.

    show   sort by    
      Back to top