A fellow S.A. contributor recently wrote an article arguing that Amazon (AMZN) overstates their free-cash-flow by leasing, rather than purchasing, certain assets -- thereby, eliminating the capital expenditures required to purchase the assets. While a very well-written and interesting analysis, I disagree with his conclusions for several reasons.
First of all, and most simply stated, the analysis is detached from reality. Amazon does, in fact, lease these assets. As such, the company's cash flow statement (rightfully) reflects this reality and there's no compelling reason to shift the valuation focus from actual free-cash-flow to an adjusted number that pretends that the company is purchasing assets that they are not.
To be fair, I do believe that lease-adjusted credit ratios are a useful tool for credit analysis (i.e. using Adjusted Debt/EBITDAR) -- as operating leases do have some debt-like characteristics, from a credit perspective (to the extent of their committal nature and term). That said, I believe that these adjusted credit ratios should be just be one minor tool for credit analysis and actual debt/cash-flow ratios are more relevant and important.
Now, while I do see value in lease-adjusted-ratios for credit analysis, I believe that they are of little use for valuation purposes. The author implies that leasing overstates free-cash-flow -- allowing analysts to produce favorable discounted-cash-flow (DCF) models that overstate Amazon's value. I believe that this is simply not true. In fact, over the long-term, I would argue that purchasing the assets (rather than leasing) would actually benefit the company's free-cash-flow and DCF valuation.
Operating leases can be broken down into two cost components (principal/depreciation and interest). Simplistically, over-time, these leasing costs effectively replicate the cost of purchasing the asset. As such, free-cash-flow (whether you buy or lease) is conceptually/simplistically a wash. There is clearly a timing difference (higher up-front outflows when buying), but this is factored into DCF models which, by their nature, adjust/discount for the timing of cash flows.
But, there's more. Above the principal/depreciation and interest cost components, lessors also (typically) incorporate a profit margin into the leasing cost. As a result, over time, Amazon will arguably spend more leasing these assets than they would be if they purchased them outright. While leases give a company more flexibility and involve less risk, they can actually be more costly, over time (as you are paying a profit margin to the lessor/middle-man). Relative to buying, they can actually have a negative long-term impact on free-cash-flow and, resultant, DCF valuations. The opposite of what the author argues in his article.
I also believe that making these kind of "what if" adjustments can be a slippery slope. If adjusting for operating leases, why not also adjust for other assets that the company utilizes but doesn't own? Should we capitalize the cost of all of the airplanes that ship Amazon's products? By outsourcing these costs to transportation companies, are they avoiding the purchase of all of the planes, etc. and effectively financing their transportation infrastructure off-balance sheet? I greatly exaggerate, but you get my point.
In summary, I believe that focusing on lease adjusted free-cash-flow makes little sense from a valuation perspective. It ignores reality -- both the near-term benefits of leasing (less capital required, more flexibility/less risk) and the longer-term negatives (higher costs, as you pay a middle-man). I don't have a strong view on Amazon's valuation and the shares may, indeed, be overvalued, but not for the reasons stated by the author in his article.