How to correctly calculate Enterprise Value
After all that heavy macro stuff, back to the nitty-gritty world of fundamental analysis.
Let’s have a look at Enterprise Value, which as concept is gaining more and more attention, among others famous “Screening guru” O’Shaughnessy has identified Enterprise value as the most dominant single factor in his new book. Also a lot of the best Bloggers like Geoff Gannon and Greenbackd prefer EV/EBITDA
Interestingly many people seem to just use and accept the “standard” Enterprise value calculation.
How to calculate standard Enterprise Value
Investopedia has the “normal” definition of Enterprise Value:
Definition of ‘Enterprise Value – EV’
A measure of a company’s value, often used as an alternative to straightforward market capitalization. Enterprise value is calculated as market cap plus debt, minority interest and preferred shares, minus total cash and cash equivalents.
Investopedia also offers an interpretation
Investopedia explains ‘Enterprise Value – EV’
Think of enterprise value as the theoretical takeover price. In the event of a buyout, an acquirer would have to take on the company’s debt, but would pocket its cash. EV differs significantly from simple market capitalization in several ways, and many consider it to be a more accurate representation of a firm’s value. The value of a firm’s debt, for example, would need to be paid by the buyer when taking over a company, thus EV provides a much more accurate takeover valuation because it includes debt in its value calculation.
So this is a good hint how to understand Enterprise Value: It originates from take-over valuation, most prominently from Private Equity investors or “old style” corporate raiders.
How to UNDERSTAND Enterprise Value
The private equity / Raider business in principle is relatively easy: You buy a company (or achieve full control) and then in a first step you extract all existing cash and/or assetswhich are not necessary to run the business from the company. In a second step, the corporate raider will then put as much debt onto the target company’s balance sheet and let it pay out as a dividend or capital reduction.
The more the Raider can get out quickly either as excess cash or as a “dividend recap” (short form for a debt financed dividend) the higher the return on investment.
The first important aspect: Excess cash OR excess assets
As I have written above, a raider of course likes best plain cash lying around. On the other hand, the raider will happily sell anything which is not really required to run a business and pocket this cash as well. However mechanical screeners will only capture cash on the balance sheet, not any “extra assets”.
A good example is my portfolio company SIAS. Their EV/EBITDA decreased strongly because the “exchanged” their extra asset in the form of a South American minority stake into cash. Another “extra Asset” company would be EVN with its Verbund stake.
Including the Verbund stake, EVN looks quite expensive at EV/EBITDA 8.3 (EV ~ 4 bn, EBITDA ~ 500 mn) against 5-6 EV/EBITDA at RWE and EON. However if we deduct the “extra asset” of 25% Verbund (~1.6 bn EUR) from the 4 bn EV, we suddenly end up with an EV/EBITDA of <5, a lot cheaper for this very conservatively run utility company.
In my experience, it is much more interesting to find companies with extra assets which don't show up as cash on the balance sheet. This was mentioned before as favorite technique of value legend Peter Cundill.
Next step: What to add to Enterprise Value
So its pretty clear that the less debt a company has the more a PE/raider will be willing to pay.
But it is also important to understand, how the capacity to put debt into a company is determined. Especially in the US, the debt will be put into the target in the form of corporate bonds. In order to sell them, you need to have a rating.
The lower the rating the more expensive the debt. In practice, raiders will try to achieve a BB rating as this is usually the “sweet spot” before bond spreads go up dramatically.
Rating agencies have relatively simple ratios to determine maximum debt loads within a certain rating category, however the most important point is this one:
Rating companies add additional items to determine debt capacity which are:
– pension deficits or unfunded pension liabilities
– financial and operating leases (capitalised)
– any other known fixed payment obligations (cartel fines, guarantees etc.).
Economically, those items are very similar to financial debt which is usually included in the EV calculations, as they represent fixed payment obligations which sometimes (like pensions) even rank more senior than debt.
It is therefore no wonder that with a “standard” EV/EBITDA screener, often UK retail companies with huge (underwater) operating lease and pension commitments show up as “cheap” and then people are surprised that they go bankrupt soon (Game Group anyone ?).
Special case: prepayments
Prepayments are an interesting feature of some business models, among other for instance at Dart Group.
Normally, a company produces its goods first and then sells them again receivables until cash is then finally collected. In the case of prepayments, cash comes first in against a payable and the good gets produced at a later stage and then delivered with no further cash inflow to the customer. If the prepayments do not carry any formal restrictions, the company in theory can use the cash for whatever it wants. So for instance if a company can finance inventory out of payables, the prepayment cash could be used to finance even machinery or to reduce financial debt. So to make a long story short: cash from prepayments without formal restrictions should be considered “free cash” and deducted from enterprise value.
How to calculate Enterprise Value correctly:
So now we have all ingredients to correctly calculate Enterprise Value:
a. Equity Market cap
b. Financial debt (long + short term)
c. minorities, preferred
d. financial leases and operating leases
e. pension deficit or unfunded pension liabilities
f. any other fixed liability which has to be repaid independently of the business success
g. cash or cash equivalents
h. “extra assets”, assets not required to run the business
Of course, EBITDA has to be adjusted as well in order to make usefull comparisons.
Basically we have to add back leasing expenses and pension expenses to EBITDA in order to compare the ratio against other companies.
Standard screening EV/EBITDA does omit various relevant elements of an “economical” Enterprise value. Adjusting it for relevant items will prevent an investor to end up with relativ obvious value traps.
I am willing to bet that a back test on the adjusted EV/EBITDA ratios would generate even better results than the “standard” EV/EBITDA calculations.
What implication will be there if I don’t add Minority Interest in Enterprise Value? Why do we add Minority interest in EV? Do I need to adjust EBITDA or EBIT if I don’t add Minority Interest? If so, why?
Thank you for the great article but I have a question. I can understand why we need to add back operating lease as it would not show up on the balance sheet but why do we have to add back financial lease? Isn’t financial lease already included in debt in forms of lease obligation? Also when adding back operating lease would we add future lease payment’s lump sum?
depending on the accunting standardm financial leases might be included in financial debt or not. If they are already inlcuded, then there is no need to double count them. For ap leases, you can either just use nominal cashflow or try to discount them at the market rate of corporate debt for the respective company. With currently low rates, the difference will be quite small unless we are talking about High Yield companies.
please correct me on this…. isnt the Market Cap an arbitrary number to try to use as a numerator? MC is price of the shares X the amount of shares outstanding. The # of shares outstanding is NOT a scientific or strategic amount (or is it).
So is it good to compare companies within the same peer group that have different shares outstanding which changes the multiples dramatically?? the peers will will both share ~roughly same industry growth, economic headwinds, macro conditions, etc? please reply so I can get past this issue… thanks
Sorry, I am not getting your question fully. Enterprise value should capture both, the market debt of the equity as well as the debt. You always want to compare the valuation of the company so you need the market cap for all peers at the same time. Market cap is always shares outstanding times share price. The actual number of shares doesn’t matter at all as the result is a ratio.
What about the stock with ticker “WAKE”? Its a small bank. Why is its “Enterprise Value” negative? It is because EV fails to include loans?
EV doesn’t work well for financial companies. Without knowing WAKE, I assume that client deposits are not counted as debt (which they should).
Shouldn’t the cash advance from customer be considered as customer funded debt?
Working capital should be included while calculating EV?
not sure what you mean
Pingback: Yet another airline stock: Transat AT | Reminiscences of a Stockblogger
“Intuitively I would say that the value of the tax shield of the exsiting debt should be included already in the market cap.”
I think that if that were always so, the private equity sector wouldn’t exist. They make their living almost exclusively off the tax shield.
I think it’s helpful to remember that there are two different things that, unhelpfully, are both commonly called “Enterprise Value”: the first is essentially a more sophisticated version of price; and the second is essentially a more spohisticated version of value.
“Price” EV uses market cap + market value of debt – (excess) cash
“Value” EV uses (ROIC/Cost of Capital) x invested Capital or, in plain English, the value of the business itself, plus the other items listed above (debt equivalents net of tax shield, minority interest etc)
As we all know, EV1 and EV2 are going to be equal, approximately and most of the time. If EV2 > EV1 by a substantial amount, however, you have a value opportunity.
And the value investor’s task is to unpick each component of EV2 and see if is correctly estimated by the market prices embedded in EV1
i still don’t think that one should deduct the Tax shield o existing debt from the enterprise value.
You cannot “cash out” this item.
Howver it is clear that up to a certain extent, a higher debtload will increase After Tax profit, but I thin it is important not to “doublecount” the cureent capital structure.
I understand your concern about double counting but the numerator is unlevered after-tax operating profit.
Interest — and that includes debt interest, implicit interest embedded in operating leases, and interest expense on pension liabilities — are added back to EBT to give EBIT. The tax expense is also adjusted to remove the tax benefits on those financing items — i.e. “income taxes becomes “taxes on operating income”. That gives you unlevered after tax operating profit.
EV / unlevered after-tax profit is the “P/E” multiple that allows one to directly compare any business with any other,
interest, leases, preferred, pension financing, minority interests, excess cash etc — these are financing choices; they have nothing to do with the business as such. They should therefore be dealt with separately, in the “non-operating items” column. And if interest expenses are in that column, tax shields should be there too.
A number of things missing from the EV formula above, the most important of which is the value of the tax shield on interest and quasi-interest expense:
–debt & capital leases
– capitalized operating leases
– underfunded defined benefit plans
multiply this total by the prevailing statutory tax rate. That’s the tax shield. It has value since the interest expense on these is tax deductible; that’s why firms take on debt (or lease, rather than buy, equipment) in the first place. This tax shield should be added back thus:
– market cap
– debt & debt equivalents
– tax shield on debt & debt equivalents
– (excess) cash and cash equivalents.
That’s EV. From EV, we can move to he value of equity by subtracting claims on equity:
— minority interest
— market value of preferreds
— net deferred tax liabilities
Preferreds are a hybrids, of course, and therefore can be treated as debt equivalents or equity equivalents according to the context and one’s own preference.
On the numerator, I’ve never understood the attraction of EBITDA rather than adjusted EBITA or, even better, after-tax operating profit. “ATOP”/EV is a very accurate earnings yield figure, especially when “ATOP” is normalized across several years. And “ATOP”/Net Tangible Assets is a highly reliable indicator of business quality. EBITDA, however, is often highly misleading, even for comparisons of direct competitors.
Anyway, thanks for your always interesting and informative blog.
thank you for the comments. I have to “digest” this first.
Intuitively I would say that the value of the tax shield of the exsiting debt should be included already in the market cap.
However, the potential delta in debt for a control buyer could indeed lower the EV.
I do think that the length of the leases really influences the leverage and value of the company.
If for example at Praktiker business is bad, you are screwed with 10 year leases because you are not able to get out without paying high breakage fees or you loose a lot of money through cheaper subleases if they are possible at all.
With short leases, the leverage is a lot lower. Don’t forget that Ev is nt necessarily an equity valuation tool but more a debt capacity measure.
I don’t like the factor approach too much because than you cannot differentiate between high leverage and low leverage.
by the way, from 2013 or so, operating leases will have to be capitalised under IFRS anyway.
Hi MMI. Do you know how the leases will be capitalised under the new rules?
not really sure but i guess it will be similar to financial leases, i.e. discounting the lease obligation by the implicit or explicit interest rate.
For instance here:
how do you calculate e. pension deficit or unfunded pension liabilities?
The problem is, we don’t know when pension receiving people will die or when paying a certain pension to an employee will start. Also future interest rates are unknown. I normally stay clear of companies with significant defined benefit pension plan.
In the US they may alter the laws allowing smoothing of interest rates for a longer period which would result in lower pension accruals in the current low interest environment.
you normally start with what is on the balance sheet. Unfunded liabilities are usually recorded on the balance sheet using standard actuarial assumptions which in a first step then can simply be added to financialdebt.
Funded DBO plans are usually mentioned in the notes.Normally they have to provide the fair value of the assets and the actuarial value of liabilities. You would then record any deficit as debt.
In a second step one might want to challenge the underlying assumption,like the exampleyou mentioned and adjust liabilites.
However, staying away fromcompanies with large pension liabiliities (and life insurance companies) in general might be a good idea.
I have been doing this (adding lease obligations to EV and adding current lease payments back to EBIT) but I’m not sure about it. I often find it makes companies with huge lease obligations look even cheaper. The effect of adding back lease payments to EBIT seems to outweigh the increase in EV. I don’t really know enough about operating leases to explain it, but it might be better to use a multiple of the lease payments to calculate the obligation than the minimum future operating lease commitment in the notes to the accounts.
yes, sometimes its not so straight forward.
It depends mostly on how long the leases are. If you have only 5 year leases, than a EV/EBITDA 5 company will stay more or less constant in valuation terms. If they have howver 10 year leases, EV/something will increase.
the “real” trick with leases is to find out if a company pays a lot more than a new competitor across the street would have to pay.
Again,this difference should show up in a lower EV/EBITDA for such a competitor (all things equal) .
does that make sense ?
Hi, and thanks for the reply. So the shorter the leases the cheaper the company looks. Yes it does make sense but as a kind of short-hand method what do you think of the multiple idea. For example if lease payments in a particular year are $1m and we pretend that money was interest on a loan at, say 6% then in very crude terms we could estimate the lease obligation or debt is $16.6m.
Obviously the interest rate is a guess but if the company borrows it might be possible to find out what rate it borrows at and use that. It gets around the problem of different length leases doesn’t it?
A good portion of the cash and cash equivalents could be accumulated foreign earnings on which significant taxes would apply in the case of repatriation. After deduction of the foreign tax credit sometimes over 20%.
… a rate over 20%.
this is often a problem with US companies,less for German or European comanies.
For German companies for instance, repatriation costs only ~1%.
Some companies allow customers to rent or lease their products. I omit the corresponding financial debt in the ev calculation.
This was the case e.g. for jungheinrich which seemed too expensive with normal ev calculation. I don’t know if this is correct but I thought it’s fair to omit the bank inside the business but keeping in mind the risks.
thank you for the comment. This is actually an interesting point.
I would argue that you have to add the debt and then subtract the assets as “Extra assets”. That might sound complicated but it forces you to think about the value.
no, I meant somthing different. For instance if a company needs to pay a 100 mn EUR fine in a year, then the reserve should be included.
Or environmental liabilities for a wind park.
Working capital is different.
Of course you start always with a mechanical screener, I just wanted to give some hints at what to look at.
“any other fixed liability which has to be repaid independently of the business success”
=> That means all liabilities except preferred shares (which you already included above), since tax liabilities also have to be paid irrespective of business success and within a certain time limit. Accounts payable are also independent of business success and so on. All these liabilities have actually be paid at some point!
For me, that seems not practical at all. It would be better to start with a much simpler mechanical screener, and then to look at the top findings manually – maybe one can exclude retailers…
one additional comment: Of course all liabilities have to be paid. But some liabilites get “rolled over” under normal circumstances like payables.
At some point in time, however payables can become “EV relevant” liabilities, mostly in a distressed scenario.