jump to navigation

Amazon earnings off another 1% due to real estate …. or so they would be if the new lease accounting was already in effect July 28, 2010

Posted by Bob Cook in Company Case Studies, Financial Planning & Analysis, Lease Accounting.
Tags: , , ,

Earlier this week, I posted about Amazon’s disappointing results… about how the market reacted negatively to its increasing operating expenses and how real estate expenses likely are contributing to its higher-than-desired cost structure.  Let’s take a look, though, at how expenses would have been even worse if the new lease accounting standard, being created by FASB and IASB, was in effect now.

(For a good review of the new accounting standard and its implications, see my webinar series which is available on-demand on TRIRIGA’s website.)

As I’ve posted previously, in the year that the new lease accounting is adopted, a company is going to see its lease-related expenses increase by 5% to 10% or even more where the remaining terms of leases are exceptionally long.  We can use the operating lease obligation amounts shown in Amazon’s latest 10-K to estimate what the effect on Amazon will be. 

The 10-K shows minimum operating lease obligation to be $162 M (2010), $146M (2011), $130 M (2012), $122 M (2013), $115 M (2014), and $317 M (2015+).  Assume the last amount is actually distributed evenly over the years 2015 through 2019.  For simplicity, we will ignore the fact that there may have been free rent periods or fixed rent increases that may make acrued expenses differ from cash flow.  Let’s also assume that Amazon’s applicable incremental borrowing expense, which would be used to discount the obligations to determine how much lease obligation goes on the balance sheet, is 3.5%.

Under current accounting, none of the operating leases would be on the P&L, and the P&L expense for these operating leases would be $162 M for 2010. 

Under the new accounting, though, Amazon would record an $850 M right-to-use asset on the asset-side of its balance sheet and an equivalent amount as a lease liability on the liablity-side.  This will set in motion P&L accounting for the leases.  There, if the new standard were in effect, Amazon would record expense of $174 M … an increase of $12 M over the expense under current accounting … or an increase of 7.5%.  See chart, below.

The impact will be even more if Amazon’s leases have options to renew that are likely to be exercised.    The obligations related to the renewal periods would also go onto Amazon’s balance sheet and, due to the way the accounting arithmetic works, result in an even higher expense in the first year of adoption of the accounting standard.

How would this $12 M affect Amazon’s bottom line?   Let’s use, for illustration, the company’s Net Income for FY2009 of $902 M.  This $12 M of additional expense, adjusting for taxes, would decrease net income by about 1% … not an insignificant number.  To put the number in perspective, if the market were, therefore, to cut Amazon’s valuation by 1%, approximately $500 M of share value would be lost.

Whether one agrees with the logic of the new accounting or not, the financial effect is going to be something that cannot be ignored.


Current vs New Accounting for Amazon’s Existing Operating Lease Portfolio  Clarification:  The  information in this chart shows the P&L impact of operating leases that existed as of 12.31.2009.  It does not assume any renewal of those leases, nor does it make any assumptions regarding the signing of new leases.

  2010 2011 2012 2013 2014 2015
Current Accounting            
Rent 162 146 130 122 115 63
  Total 162 146 130 122 115 63
New Accounting            
Depreciation 144 128 113 106 99 53
Interest 30 25 21 17 14 10
  Total 174 153 134 123 113 63
   % Increase 7.5%          


1. Richard L. Podos - July 28, 2010

Great post Bob, but I am not following the math.

I get an $854M asset and liability number, but what is your methodology to derive the depreciation and interest as above?

Also, where do you get the 3.5% incremental borrowing rate? Is that a blend adjusted for each year of obligation (i.e., very very low for earlier expirations, and successively higher further out?


2. Bob Cook - July 28, 2010


I have a proprietary model that I’ve built that takes information provided in company 10-k’s, teases out of that data the rental streams likely attributed to tranches of leases expiring in each of the upcoming years, and then calculates the PV for each and prepares depreciation schedules and liability amortization schedules.


3. Richard L. Podos - July 30, 2010

So then, I was correct that you are effectively ending up with a very low incremental borrowing rate on average, as at the day (i.e., fiscal quarter) of adoption, the remaining lease terms are of a shorter average life than in ‘reality’ / inception?

I had not thought of that as yet another problem with the proposed implementation!

Let me say that a different way for everyone:

In the quarter that a given corporation adopts the new lease accounting, they will have to assign a discount rate for each lease obligation (along with all the other projections such as contingent rent, renewals, and currency exchange… I’ll come back to the in a bit).

The discount rate per lease will be as per remaining lease term, and not original term. By nature, a given portfolio will have many leases with remaining term less than average original term. Thus, at adoption, the blended average incremental borrowing rate will be lower than the would have been the case if all the leases had been treated this way from inception. The consequence is that where your company’s average 10 year borrowing rate might be 5.0%, the application (as Bob did for Amazon) will actually be at something like 3.5%, which thus makes the NPV of the lease obligations (both asset and liability) *much* larger.

That’s effed up! And also, not consistent with the Boards’ goals of transparency and accuracy for the end-users of financial reporting (i.e., investors). Meaning, for instance, Bob’s example of a 1.0% decline in Amazon’s EPS is an incorrect fiction (meaning Bob did it correctly, but the result is nonsense). Going back to contingent rents, renewal options, and currency exchange, I could make the argument that Bob probably *understates* the impact!!

There’s also a brain-pretzel in here that I haven’t fully absorbed, where companies with longer average lease life get the benefit of a higher discount rate…

1) I would hope the ratings agencies and analysts understand this fallacy over the next couple of years.

2) Yet another reason for the corporate real estate community to rally, in the form of comment letters to the Boards, through the end of 2010.

Good stuff, Bob,

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s