Letter number 77 of October 2013


News : Accounting for operating and capital leases – Errare humanum est sed perseverare diabolicum!

The IASB and the FASB have a joint project that is already well underway. They are seeking to carry out a root-and-branch overhaul of the method for accounting for operating and capital leases, on the pretext of improving the financial information available to investors.  For our part, we believe that this is a giant leap… backwards, by about 30 years! 

Briefly, it will involve accounting for operating leases in the same way as capital leases, by creating on the asset side of the balance sheet, a rental right by discounting future cash flows, the counterpart of which, on the liabilities side of the balance sheet, will be a financial debt.

After public consultation, which did not reveal great enthusiasm for this project (and that’s a euphemism), it was sent back to the drawing board in 2011.  It re-emerged in May this year, in different version, for public consultation which has been completed in mid-September.  We may as well say straight away that the differences between this new version and the initial draft, as far as firms are concerned, involve small details and not, unfortunately, the substance of the project, which, in our view, is just as bad as it was.

In business life, there are two main types of leases:

- operating leases, under which the owner of an asset makes it available to a lessee for a defined period in exchange for the payment of a rent and the return of the asset on expiry of the rental agreement, unless this lease is extended; and

- capital leases, under which a company makes an asset available to a lessee for a period that is close to the asset’s life span, usually with an option to purchase the asset at a price that is lower than the residual value of the asset at the end of the leasing period, in exchange for the payment of a rent on a regular basis, part of which corresponds conceptually to remuneration for making the asset available and another part to the anticipated payment of a part of its price.

These two different types of leases are translated in the accounts by two different accounting treatments that correspond to the different economic natures of these operations.

Under an operating lease, rent paid is booked on the income statement under operating expenses.  Nothing is booked on the balance sheet.

Under a capital lease, the asset is recorded on the asset side of the lessee’s balance sheet, with a corresponding entry on the liabilities side of a financial debt that is equal to the current value of outstanding rent.  On the income statement, the rent paid disappears and gives way to financial expenses and the depreciation and amortisation of the asset.  It’s as if the lessee had acquired the asset and had taken out debt to do so.  There is nothing shocking about this treatment which, on the contrary, reflects the intention of the firm which is to enter into a financial lease agreement.  It will use the asset as if it were the effective owner thereof as it knows that in the end, it will become the owner, while financing the asset using debt during the interim period.  For the firm, a capital lease is, above all else, a loan.  

A great leap forward was made around 20 years ago when the accounting authorities decided to interpret capital leases economically, rather than in their strict legal sense (legally speaking, the lessee is not the owner of the asset).  Accordingly, it was no longer necessary to make adjustments to the accounts by recording the asset on the asset side of the balance sheet and a debt for the same amount on the liabilities side, and to restate the leasing repayments as financial expense and depreciation and amortisation.  This was definitely a step in the right direction.

Once again, the IASB and the FASB are seeking to have operating leases treated in the same way as capital leases.  The value of the right to use the asset will be entered on the asset side of the balance sheet.  Parallel to that, a debt of a financial nature will appear on the liabilities side which will gradually be reduced.  Rent will be replaced by financial expense and depreciation and amortisation, which will reduce the value of the utilisation right each year.

The value of this utilisation right will have to be revised if a change in the economic and financial terms impacts negatively on its value, leading to a depreciation on the income statement.

When compared with the initial project, this one introduces a new provision by classifying leases into two types – A and B – the main accounting consequence of which is a different amount and accounting method for the rent on the income statement.

Type A groups together capital or operating leases for which the rental period corresponds to the major part of the economic life of the rented asset or for which the discounted value of rent is equal to practically the whole of the fair value of the asset. In other words, “long” leases compared with the residual life of the leased asset.

Type B groups together capital or operating leases for which the rental period corresponds to an insignificant part of the economic life of the rented asset or for which the discounted value of rent is equal to practically the whole of the fair value of the asset. In other words, “short” leases compared with the residual life of the leased asset.  Real estate leases, unless an exception is made, will be classified as type B.

For A-type leases (mainly long leases whether on real estate assets or not), the rent expense recognised on the income statement will be different from that recognised as cash and effectively paid.  But for the whole of the duration of the contract, all of the expenses recognised would correspond to the amount of rent effectively paid.  In fact, the new project requires that the present value of the rent paid over the duration of the contract, discounted at the contract rate (for a capital lease contract) or at the cost of the company’s marginal debt (for an operating lease), be depreciated on a straight-line basis over the life of the contract.  To this depreciation expense, which is constant every year, would be added interest expenses, calculated as the product of the interest rate seen above, multiplied by the present value of the residual rent. The total expense, the sum of a constant depreciation and decreasing financial expense, would thus decrease.  The two expenses would appear separately, one under operating expenses and the other under financial expenses, on the income statement.

For B-type contracts (which would mainly concern short leases, whether for real estate assets or not), the rent expense on the income statement would be constant every year (and would correspond to rent effectively paid).  It would result implicitly from the sum of a decreasing financial expense corresponding, like above, to the product of the interest rate multiplied by the present value of the residual rent, and complementary increasing depreciation and amortisation. But this would not appear on the income statement because there would be no breakdown between the two components and only a single operating charge would appear.

On the cash flow statements, B-type contracts would reflect, like today, a rent under operating flows.  A-type contracts would reflect the reimbursement of the principle under financing flows and the interest under financing flows or operating flows.

In other words, for B-type contracts, there’s nothing new on the income and the cash-flow statements.  It’s only on the balance sheet that the revolution takes place, with the booking of a utilisation right and a “rent liability” (i.e. its debt) on the other side.

A-type contracts are in fact leases, like capital leases are today.  On this point, the difference between A-type and B-type leases replaces the difference between operating and capital leases.

The accounting standards authorities base their point of view on the following assumptions:

- a capital lease and an operating lease basically cover the same economic act and it is not normal for them to be treated differently in the accounts;

- it is difficult to draw a clear distinction based on principles, and not on rules, between an operating lease and a capital lease and it leaves too much room for manoeuvre to managers if they are left to make this choice, which affects the comparability of accounts from one company to the next;

- since investors make adjustments to operating leases in order to analyse them as capital leases, why not do it for them.

We’d like to voice our firm disagreement with these positions.

The spirit of an operating lease has nothing in common with the spirit of a capital lease.

A firm uses an operating lease because:

- it does not have the financial resources, either now or in the future, to buy the asset;

- it wishes to retain the flexibility of being able to return the asset when the lease expires and to rent another one which is better suited to its future needs, or even to purchase one;

- it prefers to allocate its financial resources, which more often than not are limited, to other areas that it deems more efficient for the firm –  R&D, external growth, advertising spending, etc.

A firm makes use of a capital lease because it wants to use the asset as if it were already the owner thereof, and in the long term, to acquire it.  But today, it cannot or does not wish to acquire it without taking out a loan.  Because the lender remains the legal owner of the leased asset until expiry of the contract, it is a form of credit under which the lender enjoys a very solid guarantee. This means that it is able to offer attractive interest rates to the borrower.

Accordingly, the economic and financial logic behind operating leases is totally different from the logic behind capital leases.  It’s a pity that the IASB doesn’t see things this way and that it is set on a course that is going to make the reading of accounts more complicated. It will unnecessarily increase the work load of those whose job it is to prepare them, and contrary to what it claims without advancing the slightest bit of evidence, it will complicate the lives of those who use accounts on a daily basis.

More seriously, by complicating the accounts, the IASB and the FASB are making them more opaque and thus less credible for mere mortals, whereas the simplicity of accounts is the keystone of their acceptability.

It is, however, true that in certain sectors, such as mass market retailing, the hotel sector and the aviation sector, companies use operating leases on a massive scale, often selling assets they own outright, only to rent them back again from their new owners under operating leases.  It is also true that rating agencies and banks adjust the accounts of airline companies and hotel groups to factor in their undertaking to pay rent, which could take a heavy toll on free cash flows.  Accordingly, Standard & Poors capitalises Accor’s fixed rents, but does not adjust rents that are indexed to sales.

What is the share of these sectors in the whole of the economy? Around 10% at the most.  Wouldn’t it be sufficient to request that this information be included in the notes to the accounts to enable those who so wish, a tiny minority of investors - and not the majority as the IASB claims without providing the slightest bit of evidence for this claim - to carry out the adjustments that they wish to carry out?

As for claiming that it is difficult to tell the difference between an operating lease and a capital lease, well they’ve got to be joking.  What is the point of IAS 17? What about the distinction between A-type and B-type leases that this project introduces? If this new project has at least one good thing about it, it is to offer a breakdown key, which to us seems precise, between operating leases and capital leases, and could replace the current breakdown contained in IAS 17, which is more conceptual.

And if we follow the current logic of the IASB and the FASB, why stop there seeing as though they think they’re on the right track?  Why not record an employee utilisation right on the asset side of the balance sheet (as we all owe our employers a notice period), with a debt on the liabilities side (the salaries due during the notice period).  The same goes for contracts with suppliers providing for a volume of goods and services to be purchased over several periods.  Like with all things, a line should be drawn that clearly separates that which falls under finance and that which falls under flexibility (operating leases).  In our view, that line is where it should be now, and not where the IASB and the FASB intend to put it.

Let’s hope that common sense will prevail, that IAS 17 will simply be made more precise with regard to the distinction between capital and operating leases and that the current way in which they are booked will remain in place, as neither the IASB nor the FASB has shown that their joint project would be an improvement on the current situation.  We’re convinced that the opposite would in fact be the case. 

Statistics : ROE and ROCE for the largest European listed companies

Return on capital employed

Source: computed from Exane BNP Paribas data

The average 2012 ROCE is 13.2%. In 29% of the cases, ROCE is less than 9% which is the order of magnitude of the cost of capital.

Return on equity

The average 2012 ROE is 14.8 %. Unsurprisingly it is higher than the average ROCE of 13.2% because of the leverage effect[1].

The leverage effect increases the return on equity of groups with a high ROCE. Only 16% of the groups have a ROCE above 20% but they are 22% to register a ROE above this level. On the other side, the leverage effect reduces ROE for companies with a poor ROCE. Only 3% of them have a ROCE below 3% but they are 9% to register a ROE below this level.

 Source: computed from Exane BNP Paribas data



[1] For more on the leverage effect, see chapter 13 of the Vernimmen.


Research : the case of pro-active increases in debt levels

With Simon Gueguen – Lecturer-researcher at the University of Paris Dauphine

In these columns, we regularly look at the theories of capital structure and the issue of optimal debt levels. The existence of different and often contradictory theoretical models provides an excellent area for empirical studies, which either support or disprove these different models.

The article that we present this month[1] forms part of this literature, but suggests a different approach.  Its aim is not to test the ability of different parameters to explain debt levels, but to study a precise situation – pro-active increases in debt levels.  These are sudden increases in debt resulting from an increase in gearing above the expected level given the characteristics of the firm.  They are called “pro-active” because they are not simple technical decisions that are intended to result in a target gearing being met, for example following an increase in the stock market price.  Denis and McKeon got together a sample of 2,314 events of this type, involving US firms between 1971 and 1999. They then studied the reasons behind these pro-active increases in debt levels along with the subsequent change in gearing.

For the purposes of selecting their sample firms and carrying out empirical tests, Denis and McKeon first measured, for each firm and for every year, a target gearing taking various factors into account (market-to-book ratio, tangibility of assets, profitability, size, sector’s median level of gearing). They only kept debt increases resulting from gearing of at least 0.10 of the target level in their sample.  Before the increase in debt, the median gearing of their sample is close to the target level at 0.30. The proactive increase in debt results in a level of 0.55.  Accordingly, they studied very sudden and very large increases in debt.

The analysis shows that in a very large majority of cases (67%), the funds raised are used for long-term investments. For the remainder, in 23% of cases, funds were used to finance increases in working capital.  Returning cash to shareholders out of the funds raised only concerned 3% of cases.  These pro-active increases are thus intended to finance firms’ assets and not to restructure their liabilities.  Denis and McKeon also show that, in 90% of cases, firms would not have been able to pursue their operational policies without the funds obtained by these debt increases.

Another part of the study concerns the development of gearing over the seven years that follow the initial increase.  Although this gearing tends to decrease, half of the initial increase remains present after seven years.  What does happen is that firms reduce gearing as they generate cash. Firms that have problems generating cash tend to take out more debt, which runs contrary to the theories of target debt.

By intentionally selecting atypical behaviour, the article raises more questions than it answers.  It shows that in many cases, firms’ behaviour in terms of is not consistent with the main theories.  A lot of work still remains to be done on this theme, and not doubt we shall soon be looking at it again in these pages!



[1] D.J. DENIS et S.B. McKEON (2012), Debt financing and financial flexibility, evidence from pro-active leverage increases, Review of Financial Studies, vol. 25(6), p. 1897-1929


Q&A : Why do investment funds prefer to use the IRR and the initial investment multiple than the NAV?

The reader may be surprised to note that investors which are, à priori, highly sophisticated, investment funds and in particular LBO funds, use, as criteria for their investment choices, the internal rate of return or the initial investment multiple, rather than net present value, which researchers have shown is more efficient and relevant for their purposes.

The main objection to the IRR is that it results in a maximisation of a rate of return, rather than the value of the funds invested, assuming that the intermediary flows generated by the investment are reinvested at the IRR rather than the cost of capital of the project, and sometimes results in a mathematical dead-end when there is no IRR or when there are several[1].

Let’s first eliminate the latter two situations, which assume, in order to occur, more than one sign change in the unfolding of the flows, which is not the case in LBOs where there is simply more often for the equity investor an initial flow and a final flow at the time of the disinvestment. 

Since intermediary cash flows received by LBO funds are immediately returned to investors, the issue of the rate of return to at which these flows are reinvested within the fund, does not arise.

Finally, one can only confuse maximisation of a rate of return and maximisation of value in the case of a choice between investments that are mutually exclusive (like constructing a factory that produces 100,000 tons or another that produces 125,000 tons) or when investments are constrained by limited financing which means that not all value-creating investments can be made.  It is not easy to see how the first case can apply to an investment fund.  In the second case, the risk is low as a fund that succeeds raises a new fund before it had even exhausted the investment capacity of the previous one.

Yes, investment funds do not respect financial orthodoxy, but in their case, this does not have consequences, because they are in the situation where defaults on the IRR do not apply.

More practical reasons mean that the intensive use of IRRs or initial investment multiples does not turn investment funds into renegades!

Talking about a net present value of 100 on an investment does not make much sense for investors, some of which have invested 2, others 4 or 10, or even 15 in this opportunity.  It makes much more sense to talk about a multiple of 3 times the initial investment or about an internal rate of return of 20%.  This makes sense to each investor because they have all invested different amounts.

Finally, for an industrial player, an investment is often in equipment which generates cash flows every year and is kept for most of its operating life.  For an investment fund, an investment is an enterprise which generally does not pay it any intermediate cash flows and is kept for only a few years.  For an industrial player, the initial investment multiple makes no sense since the investment is progressively consumed and loses value - it is not intended to be sold. For the investment fund, this makes a lot of sense, since from the time the investment is made, we know that the outcome is the resale in a few years.  And what better way to mark success than to put a figure on this somersault?



[1] For more see chapter 17 of the Vernimmen