Looking for Safety in REITs
Board: Real Estate Investment Trusts
I really enjoyed reading XOT's post of Nov3. He is certainly right that REIT pfd investors must weight the importance of dividend yields between issues in the context of creditworthiness, balance sheet strength, upside and downside potential, call protection, etc. An 8.5% yield may be totally uninteresting vs. a peer with a 7.5% yield if the former is a major credit risk.
I think this exercise is most valuable when comparing REITs that do business within the same property sector, i.e., comparing REGpG with GRTpI makes sense, but comparing TCOpK with GRTpI would be an ever better exercise. Similarly, we might compare KIM's pfds to REG's.
On the other hand, comparing AHTpD to GRTpI doesn't work as well, given that they are in totally different property sectors.
So would it be useful to compare risk among the various property sectors? Perhaps - but it's not likely that any two CRE experts would agree on this. It's pretty clear to me that lodging is the riskiest of sectors, due to the lack of any long-term leases and the operating leverage involved in owning hotels. All other sectors are inherently less risky. Office might be the next riskiest (despite long-term leases), due to the elongated cycles here and the long-lead times of new development. New developments are very difficult to shut down once they are started. However, CBD properties are probably at less risk than those of suburban markets where developments can proliferate and rental rates are more volatile.
Industrial is less risky than office; leases are shorter, but excess supply from new development has been less of a problem historically than in the office sector, and warehouses and distribution facilities are usually not something that users can downsize like they can office space. And there are less cap ex requirements here, which can really hurt the economics of office leasing in weak markets.
The other major sectors, i.e., retail and residential, have been quite stable historically. Cycles in the apartment sector have been mild, and hard times often bring in new renters as we have seen in the last few years. Big risks here are overdevelopment, and really weak economic conditions when jobs are being lost - sure people have to live somewhere, but in bad times they may double up or live with Mom. Unlike in other sectors, vacancies can always be filled if apartment owners are willing to cut rent and provide concessions.
Neighborhood shopping centers sell necessities, not luxuries that can be dispensed with in tough times, and are often supermarket-anchored. So there is some inherent stability in this sector, but risk also. Strip center REITs encountered difficulties during the last recession partially because of struggling mom 'n' pop tenants, but also due to development - developers were chasing rooftops during the housing boom. Supermarkets are also facing new threats from discounters and boutique grocers. Right now, however, we are in the early stages of the upward section of the cycle, with rising occupancies and firming lease rates.
Mall properties can be even more stable, due to the lack of new development, the nature of the tenants (national and regional, not mom 'n' pop); however, malls are tricky, as those that are not productive and/or those located in tertiary markets have not done well recently and some are at risk of closure, especially if a large and weak tenant, such as JC Penny, leaves. For this reason, a pfd issued by PEI is more risky than one issued by TCO. I think the 7.6% provided by PEIpB, for example, is much too skinny a premium vs. the TCOpJ, at 7.3%. There ought to be at least a 50-60 bps gap.
We should also overlay a REIT's business strategy on top of all the other factors we use, i.e., a heavy commitment to development projects warrants a higher pfd yield when all other factors are equal. A REIT's track record of development is a related consideration. There have been few really successful developers in REIT world, so we should watch those development pipelines carefully.
I am over-generalizing in the foregoing thoughts, but perhaps they might be slightly useful.