Short Thesis

â–  Significant supply is coming in KRC’s markets which will likely lead to rent reductions vs. expectations of rent growth

â–  Economic growth has been particularly strong in KRC’s markets pushing employment well above mid-cycle levels. As such, demand for KRC’s properties cannot get much better, but it can deteriorate significantly

â–  Even if I'm wrong and rents grow faster than in the past, shares are still over-valued

Business Overview  - Kilroy Realty Corporation (KRC) owns, operates, and develops Class A office space on the west coast with a current portfolio of about 14.1mm square feet. The company’s primary markets are San Francisco (45% of profits), Los Angeles (20% of profits), San Diego (17% of profits) and Seattle (16% of profits).

San Francisco Office Market is Red Hot

San Francisco is KRC’s most important market. Currently, San Francisco accounts for about 45% of total NOI. Growth in the tech sector over the past several years has caused San Francisco office market to be incredibly profitable for building owners greatly benefiting KRC. As shown in Figure 1 below, asking rents in San Francisco have grown at 8% annually since 2006 and recently surpassed Manhattan rents for the first time.

 Figure 1: San Francisco Office Market is Red Hot

This massive increase in rents has been driven by a correspondingly large increase in demand with a limited increase in supply. Since 2006 the total number of people working in the San Francisco area has increased 27% while office space has increased less than 5%. Not surprisingly, over the same period office vacancies fell from 11% to a record low of 6.8%.

Large Supply Increases are Coming in KRC’s Markets

Demand spikes such as this almost always provoke a supply response and this is exactly what we are now seeing. As a recent CoStar report states “San Francisco office space under construction is at its highest since 2001”. Since 2005, Class A office space in San Francisco increased ≈1% /yr. However in the next several years this will grow to almost 3%.

Figure 2: Supply of San Francisco Offices to Increase Substantially

Supply Increases Could Cause Rents to Fall

A simple regression of San Francisco office rents to time and vacancy shows a highly statistical relationship with an R-squared 0.99(!).  The resulting equation is:

Rent per sq ft = 59.17 + 1.02 (years) + -2.26 (vacancy rate)

This means that based on historical experience rents should increase about $1 every year and should also increase $2.26 for every 1% the vacancy rate decreases.  Over the past 10 years the number of working people in San Francisco has increased at an average of about 23,000 per year. The number of people working however has increased much faster however as the unemployment rate has fallen to a record low 2.8%.

Figure 3: San Francisco Unemployment Rate and Population Growth

Over the next several years San Francisco is expected to add about 2.35mm sq feet of office space per year. Historically in San Francisco every new worker has absorbed about 95 sq feet of office space (if this seems low it’s because not everyone works in an office) so if population growth continues at the same rate and there is no change in the unemployment rate, this should offset new supply and there would be no change in the vacancy rate. In this scenario, our regression analysis suggests rent would increase about $1/yr or <2% which is below street expectations of mid-single digit increases.

Demand Can Not Get Much Better – But it Can Get Much Worse

Demand for KRC’s properties, particularly San Francisco has been incredibly strong for several years. Obviously this is due in large part to the west coast tech boom. Technology and Media account for 48% of KRC’s revenue which is multiples of the next largest industry. In San Francisco, KRC has 121 tenants occupying 4.9mm sq ft in total. However if you exclude the 7 identified large tenants there are 116 tenants occupying 2.6mm sq feet implying an average size of lease of only 22k sq feet. This would suggest that KRC has meaningful exposure to small tech companies which are dependent on the massive increase in venture capital funding over the last several years. 

At the very least, it does not seem reasonable to expect the San Francisco unemployment rate to stay at 2.8% forever. Unemployment is one of the most cyclical and mean-reverting data series available and it is currently at the lowest level since the late 90’s tech bubble. Of course unemployment can go lower in the short-term but it simply can’t go much lower and it can easily go much higher in the event of an unanticipated macroeconomic shock.  If we merely assume the unemployment rate goes back to historical averages of 5.5% by 2020, the vacancy rate should rise to 10.6% from 7.5% currently (calculations shown in figure 4 below).  Applying a 10.6% vacancy in the regression equation would imply rents actually fall about $2.50 or about 4.5%.  The 10 year average vacancy rate for San Francisco office space is 10.5%.  As such, we are simply assuming vacancy rates return to average over a cycle which gives me greater conviction the assumptions we are using are reasonable.

Figure 4: Vacancy Calculations

Similar Supply Increases are Expected in Seattle

Seattle will also see a large increase in office supply that is unlikely to be absorbed by increased demand. Out of 29 US markets, Seattle ranks second (behind Austin) in terms of new office construction as a percentage of the total market. New construction in Seattle is about 6.9% of current office space vs. a nation-wide average of 2.5%. Similarly unemployment in Seattle is 4.5% which is well below long-term averages. Assuming unemployment goes to 5.5% and using the same methodology we used for San Francisco, the vacancy rate would rise to 12.2% (in line with historic averages) and rents would be flat.

What This Means for the Short

West coast office markets have been historically strong the last several years. This has caused a considerable increase in rents and pushed KRC’s vacancy rates down to levels that are close to the previous peak.  Even so, the market appears to expect rent growth to continue forever (as is generally the case with ‘over-earning’ shorts).  However, if our analysis is correct, rents in the next several years will be much lower than the market is expecting.  Furthermore, it would also be reasonable to assume that an increase in the overall vacancy rate will have a similar impact on KRC’s properties. Any increase in the vacancy rate would further decrease KRC’s revenue relative to market expectations.

Figure 5: Rent Growth Estimates vs. Consensus

Shares are Over-Valued Even If We’re Wrong

Over the last 10 years KRC’s rent per square foot has increased 4.6%. Lets assume that we are wrong about the impact of supply increases / unemployment normalization and over the next 10 years rent increases by 6% annually. Even in this case, shares are still about 20% over-valued (calculations shown in figure 6 below). One could perhaps argue that this valuation is too low because it assumes future acquisitions / development are value neutral whereas in the past development has created value. However, the reason past development was value-creating was because assets were purchased at cheaper point in the real estate cycle. Given today’s full valuation for west coast office assets it does not seem as if new development in these markets will create value.

Figure 6: Shares are Over-Valued Even if we’re Wrong

Key Risks

Inflation – If prices rise for the overall economy the rent KRC is able to charge will increase and thus earnings available for common shares will also increase. Since only 10-15% of leases are renewed in any given year, the impact of inflation would be gradual. This risk appears unlikely as inflation is currently expected to remain low for a considerable period. 

General Economic Strength Offsets Company-Specific Declines Impact– KRC’s business depends on office rental rates and vacancies in west coast markets and office demand is dependent on broad macroeconomic forces. If these macroeconomic forces become materially stronger and remain so, demand for KRC’s offices could increase more than the new supply and rents could increase more than expected. However, given where we are in the current business cycle this seems highly unlikely.