The risk curve

June 25, 2007 | Finance, Primer posts

Just as we have a yield curve, finance recognizes a risk curve.

The yield curve is a two-dimensional representation of the market’s tolerance of long-term debt investments, with maturity as the horizontal axis, and interest rate as the vertical.



As maturity lengthens, interest rates should be higher


In a ‘normal’ world, the longer a loan’s maturity, the higher its interest rate.  That’s common sense, since at longer intervals, macroeconomic policy — inflation and central bank interest rates — are less and less predictable. 


The yield curve posits that all instruments have the same risk profile, so it uses Treasury notes of varying maturities.  Its core principle — that yield correlates to a debt’s features — should allow us to plot a risk curve, similar to the yield curve.  While we use the same vertical axis — yield relative to a base — we plot spreads on a different horizontal axis, the instrument’s risk. 



I need … we need to plot on a different axis, Guido


To get there, we have use instruments all of which have the same maturity.



Make measures of median mature mangoes


The abstract debt risk curve thus looks like this:




The origin — the theoretically risk-less investment — is sovereign debt.  (It’s not really risk-less, but every other financial instrument is denominated in the currency, and if the currency collapses, everybody else does too.)  In the US, that’s the Treasury bill.  Then, moving left to right, we introduce default risk bit by bit, and as we do, spread relative to the safe investment rises.


We can even populate this a bit.  If we believe that the rating agencies know what they are doing, then the distinctions among ratings ought to lie neatly along the risk curve line, with all AAA-rated instruments of similar maturity clustering tightly around a particular spread.  In reality, spreads do cluster, but this may be reverse-engineering — spreads concentrate toward ratings, rather than ratings to spreads.




Which raises the question — where do we put the GSEs, Fannie Mae and Freddie Mac, on this chart?


If they were evaluated purely as private companies, their spreads ought to be in the AAA range.  (Some would argue they should have their securities downgraded because the GSEs maintain very low capital ratios, but let’s give them AAA.)  Yet in reality, GSE spreads are barely above Treasuries.  They are, in other words, under the risk curve … except for the widespread perception that they are, in fact, safer than a AAA, because Uncle Sugar will bail them out.


Sheltering under the risk curve is why the GSEs make money, and in particular why they are so eager to expand their lending into as many arenas as possible.  Lever that spread advantage over more and more dollars, and your profits go ever higher.


In defining a risk curve, we need not limit ourselves to debt; we can also apply it to equity as well.  Once again, the vertical axis is spread and the horizontal axis is increased risk, but the origin — the zero point — is different.  We have to find the lowest risk form of real estate equity, which I will suggest is the long-term, fully collateralized, triple-net lease with a major credit tenant.


(By the way, low-risk equity looks suspiciously like … debt.  Indeed, debt and equity are abutters: low-quality debt looks like good equity.  That’s also why equity should always be more expensive than debt.)


In income-producing real estate, one might populate a risk curve something along the following lines:




Working upward from safest to riskiest, we have:



With that many signs, it must be a safe investment


·         Credit lease.  Long-term lease, matching the projected financing, single tenant, creditworthy.  Like renting buildings to the US government.  In a word, safe.

·         Land lease.  Land won’t run away, and if there are valuable improvements atop the land, it takes a Biblical disaster — earthquake, flood, famine — to reduce its value.

·         Office.  A few tenants, large corporations with balance sheets and market positions, low turnover, multi-year leases, and the ability to pass through operating costs.  Obviously, office vary in safety based on market strength, large tenant strength, building appeal, and other factors, so this is less a Bohr electron point and more a cloud of possibilities that vary with each property.

·         Retail.  Like offices, they have only a few tenants, usually large ones, except that mid-size and smaller retail tenants have the annoying habit of becoming insolvent.  Like office properties, retail centers vary widely in appeal, so their yields likewise slide up and down the risk curve.

·         Apartments.  More tenants per property, more turnover per tenant, more management activity to operate the property, all make apartments generally a bit riskier (or management intensive, which often translates into the same thing) than offices.

·         Serviced apartments.  When the tenant population becomes specialized — students at one end, congregate-living at the other — the operating budget per apartment rises.  That more management attention, a more complex business, and higher risk.

·         Hotels.  You rent thirty times a month instead once, turnover is over 85% a night, there’s always a lot of competition, and customers are fickle.  No wonder the industry has learned to separate operators from property owners, and share the risk between them via operating agreements. 

·         Operating business (e.g. restaurants).  Any use for rental property that depends on the health of the underlying business is risky; further, when the property is designed for a limited group of users (e.g. restaurants), the replacement market is less deep.  Add to that the need for special configurations (restaurants need kitchens, for instance, which means more plumbing and electricity) and it’s little wonder that properties used for operating businesses are even riskier than hotels.

·         Single-use facilities (e.g. sports clubs).  If a property can be used for one thing only, and that thing is not a necessity, then when the fad wanes, what can you do with the property?  Ask anyone who invested in a bowling alley during the 1950’s, a racquetball club in the 1970’s, or a golf course in the 1990’s, how much fun it is to figure out an alternate use.



Pretty risky investment: there aren’t many other uses for a crocodile-fighting pit


Remarkably, given its evident utility, as far as I know nobody actually publishes a risk curve.  I speculate this is because it’s so hard to get reliable data, whether it’s on equivalent-maturity debt instruments (there just aren’t enough of them, and there’s no standard like the Treasury).


[Readers who have seen a graphic depiction of a risk curve should email me at davidalexandersmith {at} yahoo [dot] com, to be feted with fame, glory, and conspicuous mention in a future AHI blog post.]



Yes, I’ve seen a risk curve, it’s back in the forest here