9/27/07

Can anyone clarify why would you use one over the other? Advantages/Disadvantages? I know it's a recurring question asked in interviews and I just want to make sure I know it well.

Thanks.

Comments (105)

9/27/07

It depends on the industry. In industries that focus on cash flow such as energy, EBITDA (a proxy for cash) is king. Other industries that are highly levered such as ulilities focus on EBITDA because if a company has positive earnings but negative cash flow and cannot make it's debt payments, it goes belly-up.

It also depends on who you are talking to. Many investors only care about P/E so that is all they look at. Many accounting guys will tell you NI is worthless because companies manipulate it to be whatever it needs to be to make the investors happy. Everyone knows this is true (earnings cannot possibly be as consistent as they appear to be in SEC documents) but investors seem oblivious to that fact and only want to know whether the EPS is above or below consesus.

The WSO Advantage - Investment Banking

Financial Modeling Training

IB Templates, M&A, LBO, Valuation +

IB Interview Prep Pack

30,000+ sold & REAL questions.

Resume Help from Actual IB Pros

Land More IB Interviews.

Find Your Perfect IB Mentor

Realistic IB Mock Interviews.

9/27/07

P/E and EV/EBITDA are completely different ratios. Keep in mind where each of them are derived from.

Enterprise Value is a figure that takes into account the entire firm - remember, that's market cap PLUS debt (as well as some other stuff but forget about it for simplicity's sake). Similarly, EBITDA is an earnings number that ignores capital structure. It comes from the income statement, starting at EBIT, before you account for interest expense. That means it doesn't discriminate between equity holders or debt holders, also accounting for the entire firm.

Net income, on the other hand, is different. It's the bottom line of the income statement. Always remember, in capital structure, equity holders are the LAST people to claim any income, or assets in the event of bankrupcy. P/E, EPS, or Market Cap / Net Income are all ratios that deal with only equity, since it's taking from the net income number, which is what equity holders receive.

There aren't any "advantages or disadvantages" - they're just two different metrics that allow you to compare apples to apples, not apples to oranges.

As for when you use them - EV/EBITDA is used most usually as a multiple for transaction comps, or trading comps. You want to see the cash-generating power of the entire firm, and you don't care whether it's equity or debt financing this cash-generating operation. That's why EV/EBITDA is used for pure valuation.

EPS and P/E, being equity multiples, are usually ways to compare companies by looking at stock price. These are never used for straight-up valuation (although P/E or stock price might appear at the end of a DCF, to be what the model calculates it to be).

This isn't the most perfect definition - I hope this helped at least a little bit.

9/27/07

werdwerd is correct P/E will give you an equity value but EV/EBITDA will give you an implied enterprise value. P/E multiples aren't used too often in valuation since net income is more volatile and more easily manipulated by Management. Additionally, net income will include extraordinary items, discontinued operations, effects from accounting changes...that will need to be excluded. Also, P/E multiples reflect future growth potential of a company which could vary wildly in comp set. That being said...P/E and P/B are the multiples that are used to value financial services companies like banks since they have little to no "capital structure" related debt and interest expense is an normal operating expense.

9/27/07

right, well said, by uscwiseguy. Banks sometimes ask you how certain calculations are different if you're looking at a bank - good to know.

9/27/07

There was a thread from a couple of days ago where these questions were answered. Do a search for EV/EBITDA.

9/27/07

Well for starters, EV/EBITDA is capital structure neutral whereas P/E is not. A metric like EBITDA would probably be pretty damn useless for valuing banks and the like given that it excludes interest.

9/27/07

one example, EV/EBITDA is not used in FIG because debt/interest is a huge part of banks' operations

ROIC excludes non-operating assets and can be used for FCF and economic profit calculations

9/27/07

One use is to determine exit value of an LBO if you're assuming an IPO exit.

9/27/07

another use would be to determine if a merger would be Accretive or dilutive

9/27/07

Do you mean P/E or EV/EBITDA? two different metrics..

P/E is more or less based on the market's perception of company future earnings. EV/EBITDA has more to do with the company's capital structure.

9/27/07

Capital structure has no real effect on Price / Earnings other than as a result of investor belief and share price. Debt has no direct effect on P/E. P/E is a very simple calculation which is just share price divided by EPS. Debt could have an indirect effect on P/E in the following scenarios:

  • In the future, interest payments on debt might reduce EPS
  • Debt issuance can give a signal to investors which will cause the price to change

EV is very different as it does actually take debt into account. The issuance of debt (all other things being equal) would reduce the EV of a company and if EBITDA was unchanged, this would lead to a lower EV / EBITDA multiple.

9/27/07
Asatar:

EV is very different as it does actually take debt into account. The issuance of debt (all other things being equal) would reduce the EV of a company and if EBITDA was unchanged, this would lead to a lower EV / EBITDA multiple.

EV is net debt plus equity. Issuing debt would raise the EV, and lead to a higher EV / EBITDA multiple.

Consultant to a Fortune 50 Company

9/27/07
alexpasch:
Asatar:

EV is very different as it does actually take debt into account. The issuance of debt (all other things being equal) would reduce the EV of a company and if EBITDA was unchanged, this would lead to a lower EV / EBITDA multiple.

EV is net debt plus equity. Issuing debt would raise the EV, and lead to a higher EV / EBITDA multiple.

yea totally, just before PE firms sell their companies they load them up with debt to get massive exits, because equity never changes, right? and the "net" in net debt is just for decoration...

"After you work on Wall Street it's a choice, would you rather work at McDonalds or on the sell-side? I would choose McDonalds over the sell-side." - David Tepper

9/27/07
Oreos:
alexpasch:
Asatar:

EV is very different as it does actually take debt into account. The issuance of debt (all other things being equal) would reduce the EV of a company and if EBITDA was unchanged, this would lead to a lower EV / EBITDA multiple.

EV is net debt plus equity. Issuing debt would raise the EV, and lead to a higher EV / EBITDA multiple.

yea totally, just before PE firms sell their companies they load them up with debt to get massive exits, because equity never changes, right? and the "net" in net debt is just for decoration...

See my prior post. I am assuming the debt gets invested and you're not just grossing up the balance sheet. The EV would be higher, because you have more assets due to whatever you buy with the debt. Whether this would be good or bad for equity value depends on the perceived value of the asset bought with the debt relative to the value of the debt. If we're talking immediately after the investment, equity value will be unchanged.

Let's say you have a house you rent out worth $100K. You borrow $50K to build an addition. Once the addition is built, your prior rents collected wont show the new rent potential of the addition (EV/EBITDA multiple is higher). Your house is now worth $150K (enterprise value is higher). equity value in theory remains unchanged (though depending on whether the addition was a good investment, equity value may change).

Where in my earlier post did I say that grossing up the balance sheet was always (or even ever) a good idea, let alone that private equity firms do this to get massive exits? I said the multiple goes up (referring to ttm multiple), I did not say wealth is created out of thin air...

Consultant to a Fortune 50 Company

9/27/07

On #1, if you raise money via debt you have fewer shares outstanding, so EPS is higher, thus P/E is lower.

I think the answer to #2 is that, upon issuing new shares, EPS goes down as there are more shares but obviously the same amount of earnings. EPS is the denominator of the P/E ratio, so if the denominator decreases, the ratio increases. This is assuming price does not change.

9/27/07

It has to do with the expected return on debt versus equity and the impact that has on valuation multiples.

  1. In lay man's terms, they have a lower PE because they are now more levered and the equity is riskier. Just model it out and you'll get an idea of the math. The debt has to be paid back and is senior to the equity.
  2. The company has a theoretical "return on total assets". In a levered company, the "return on debt" is less than the "return on total assets" which is in turn less than the "return on equity" (look up WACC). By selling equity to retire debt, you are reducing the risk level to the equity component by removing the debt component. Because the return on assets stays constant, the expected rate of return on the debt goes down, and thus the P/E multiple rises.

Honestly, all this financial engineering is just such total bullshit. Everyone's trying to pick up nickels by trying to figure out whether they should do 10% debt or 30% debt instead of figuring out how to run a fucking company. Apple has zero debt. I'm sure it bugs some financial "wizards" that they do. Talk about missing the forest for the trees. (Yes, I know Modigliani Miller is not reality; but finance guys making 8 figures because somehow they are creating that value by changing the cap structure of a company is lunacy; it's just more fooled by randomness crap).

Consultant to a Fortune 50 Company

9/27/07
HezBalla:

P/E ratios are impacted by a company's choice of capital structure - companies which raise money via debt will have lower P/Es (and therefore look cheaper) than companies that raise an equivalent amount of money by issuing shares, even though the two companies might have equivalent enterprise values (For example, if a company with debt were to raise money by issuing shares of stock, and then used the money to pay off the debt, this company's P/E ratio would shoot up because of the increased number of shares - although nothing about the fundamental value of the business has changed).

1) Why will companies that raise money via debt have a lower p/e? What's the math here? Where does debt fit into the p/e equation?

  1. "For example, if a company with debt were to raise money by issuing shares of stock, and then used the money to pay off the debt, this company's P/E ratio would shoot up because of the increased number of shares ". why is this? Where does money raised by issuing shares fit into the p/e equation?

Where are you getting this from? Companies with debt do not necessarily have lower P/E ratios.

The Modigliani/Miller perspective is that the firm value is unaffected by the capital structure. EV = Debt + Equity. If you increase debt, there is a dollar for dollar decrease in equity.

This isn't quite complete however because the cost of debt and equity are affected by the amount of leverage, because it changes the riskiness of the firm. Financing entirely with debt would result in a prohibitively high interest rate. Financing entirely with expensive equity is also inefficient. If you have an idea of the market rates for debt and equity at various levels of leverage, you can calculate a debt/equity combination that minimizes the weighted average cost of capital and therefore maximizes the value of the firm. So depending on where you are relative to the optimal capital structure, you can increase or decrease firm value by issuing debt or equity.

I am wise because I know that I know nothing -Socrates

9/27/07

So many dumb replies in this thread. Mother of god. I'll drop a preftigious response soon, brothers.

9/27/07

Holy shit.. Never seen so many bad replies. Issuing debt doesnt affect P/E? Lol what about interest expense

9/27/07

EV unaffected by debt; 500mm of debt nets out with 500mm of cash proceeds.

Wtf is going on here

9/27/07
mrb87:

EV unaffected by debt; 500mm of debt nets out with 500mm of cash proceeds.

Wtf is going on here

Obviously when you issue debt you get cash, but who the hell just sits on the cash of a debt issuance? For example if you invest it in PPE the ttm EBITDA obviously won't reflect that. Enterprise value will be higher, but EBITDA won't.

Consultant to a Fortune 50 Company

9/27/07
alexpasch:
mrb87:

EV unaffected by debt; 500mm of debt nets out with 500mm of cash proceeds.

Wtf is going on here

Obviously when you issue debt you get cash, but who the hell just sits on the cash of a debt issuance? For example if you invest it in PPE the ttm EBITDA obviously won't reflect that. Enterprise value will be higher, but EBITDA won't.

So the assets purchased don't earn any income? Your EBITDA in this case will be adjusted pro-forma for the earning power of those assets you purchased. Is the company buying gold and locking it up in a vault?

9/27/07
alexpasch:
mrb87:

EV unaffected by debt; 500mm of debt nets out with 500mm of cash proceeds.

Wtf is going on here

Obviously when you issue debt you get cash, but who the hell just sits on the cash of a debt issuance? For example if you invest it in PPE the ttm EBITDA obviously won't reflect that. Enterprise value will be higher, but EBITDA won't.

EV is capital structure-neutral........................................................

9/27/07

"Kapital" is that like Karl Marx's version Investment Banking?

"After you work on Wall Street it's a choice, would you rather work at McDonalds or on the sell-side? I would choose McDonalds over the sell-side." - David Tepper

9/27/07

Oreos:
"Kapital" is that like Karl Marx's version Investment Banking?

HAHAHAAHa, yes sorry. I am from Germany and sometimes words get mixed up :)

I am not going to edit it - just so that people will understand your comment :)

but great response !

9/27/07
9/27/07

thepman:
See here: http://www.wallstreetoasis.com/forums/pe-vs-evebit...

thanks but this realy does not solve my problem:

Enterprise Value is a figure that takes into account the entire firm - remember, that's market cap PLUS debt (as well as some other stuff but forget about it for simplicity's sake). Similarly, EBITDA is an earnings number that ignores capital structure. It comes from the income statement, starting at EBIT, before you account for interest expense. That means it doesn't discriminate between equity holders or debt holders, also accounting for the entire firm.

Why is it capital structure neutral?
Shouldn't the capital structure be faktored in in the share price???

Thanks

9/27/07

N96k2q2NVy:
thepman:
See here: http://www.wallstreetoasis.com/forums/pe-vs-evebit...

thanks but this realy does not solve my problem:

Enterprise Value is a figure that takes into account the entire firm - remember, that's market cap PLUS debt (as well as some other stuff but forget about it for simplicity's sake). Similarly, EBITDA is an earnings number that ignores capital structure. It comes from the income statement, starting at EBIT, before you account for interest expense. That means it doesn't discriminate between equity holders or debt holders, also accounting for the entire firm.

Why is it capital structure neutral?
Shouldn't the capital structure be faktored in in the share price???

Thanks

Share price is the market value of the equity only and is derived from Net Income (Earnings). It does not exclude significant one-time gains or expenses, interest and D&A.

EBITDA itself is a much better measure and is a close representation of operating income and cash flow. That alone makes EBITDA multiples more accurate and reliable than P/E.

Enterprise Value includes the market value of debt and equity, meaning it is capital structure neutral. P/E can be distorted by higher or lower leverage levels b/c only the market value of equity is being taken into account.

9/27/07

anyone?

9/27/07

EV = market cap of debt plus equity. This will be cap structure neutral if MC of equity is based on BOOK value (because, for example with a share buyback, an increase in debt will be offset by an equal decrease in equity). However, if MC of equity is calculated as P*NOSH, and the price changes because of changes in the cap structure (as in PE) then surely the EV/EBITDA will also change? (Ie. with the same share buyback an increase in debt will be offset by an equal decrease in equity BUT THEN the more levered company is considered more risky and the share price drops, so EV drops).

What am I missing?

9/27/07

I'm a little confused, too. I'm going through the BIWS guide, and this is what I see:
http://cl.ly/image/3V193h232g0R

"... distorted by... capital structure..."

So according to this guide, P/E is distorted by capital structure.

9/27/07

There's no denying P/E is distorted by the cap structure becasue the P and the E will change (a more levered comany willl be priced differently to a less levered one and through interest it's earning will differ). My question is why this doesn't flow through to EV/EBITDA, if you're using P*NOSH to get EV.

9/27/07

musto430:
There's no denying P/E is distorted by the cap structure becasue the P and the E will change (a more levered comany willl be priced differently to a less levered one and through interest it's earning will differ). My question is why this doesn't flow through to EV/EBITDA, if you're using P*NOSH to get EV.

supplant the EV equation for multiples approach.

EV isn't a discrete figure. So when people say it is capital structure neutral this isn't literal. EV is (generally) far less affected by P and it is generally considered that the changes in P are picked up else where such as changed in debt quantum. but remember, we're trying to value to cash flows of a business to both debt and equity going forward, this is the fundamental tenant which flows through into EV and which is picked up in P (as a potential cash flow).

please bear in mind this is a very condensed analysis, and needs greater explanation / effort.

"After you work on Wall Street it's a choice, would you rather work at McDonalds or on the sell-side? I would choose McDonalds over the sell-side." - David Tepper

9/27/07

So the commonly held belief that EV/EBITDA is cap structure neutral is not true. It's just LESS affected than P/E. Fair?

9/27/07

musto430:
So the commonly held belief that EV/EBITDA is cap structure neutral is not true. It's just LESS affected than P/E. Fair?

no. you're jumping ahead, still looking for one figure for EV and ignoring my reference to multiples. the effects on EV of cap structure are minutiae in comparison to P/E's delta to cap structure.

"After you work on Wall Street it's a choice, would you rather work at McDonalds or on the sell-side? I would choose McDonalds over the sell-side." - David Tepper

9/27/07

Oreos:
the effects on EV of cap structure are minutiae in comparison to P/E's delta to cap structure.

Isn't that what I just said?

On a side note, I assume you always use market cap for a public company and book equity for a provate company - correct?

9/27/07

musto430:
Oreos:
the effects on EV of cap structure are minutiae in comparison to P/E's delta to cap structure.

Isn't that what I just said?

On a side note, I assume you always use market cap for a public company and book equity for a provate company - correct?

no. multiples god damn it.

"After you work on Wall Street it's a choice, would you rather work at McDonalds or on the sell-side? I would choose McDonalds over the sell-side." - David Tepper

The WSO Advantage - Investment Banking

Financial Modeling Training

IB Templates, M&A, LBO, Valuation +

IB Interview Prep Pack

30,000+ sold & REAL questions.

Resume Help from Actual IB Pros

Land More IB Interviews.

Find Your Perfect IB Mentor

Realistic IB Mock Interviews.

9/27/07

EBITDA excludes depreciation so in your scenario both companies would have the same EBITDA. Keep in mind that EV is reduced by cash on the balance sheet so perhaps the EVs are different b/w the two companies.

9/27/07

Depends on the industry/company. You wouldn't use EV/EBITDA for financial companies because they usually depend on loans & interest for their operations. On the other hand, you wouldn't use P/E for a company with negative earnings, cause you can't compute P/E that way.

Two companies can't have same operating expenses if one has higher depreciation, unless one of them has higher depreciation and lower of something else. In that case, EBITDA for the second company should be higher.

hope this helps my br0

9/27/07

Dog:

Depends on the industry/company. You wouldn't use EV/EBITDA for financial companies because they usually depend on loans & interest for their operations. On the other hand, you wouldn't use P/E for a company with negative earnings, cause you can't compute P/E that way.

Two companies can't have same operating expenses if one has higher depreciation, unless one of them has higher depreciation and lower of something else. In that case, EBITDA for the second company should be higher.

hope this helps my br0

Ya the two companies (A,B) would not have same operating expenses ...my mistake... I meant PnL items are same ( Line by line item ) just the depreciation of one (say B) is higher .. How will this impact the EBITDA of one to be higher ? and what impact will it have on the three financials ? Please elaborate

9/27/07

1.If EV/EBITDA and P/E are common metric in the industry, good analysts tend to prefer EV/EBITDA over P/E. EPS doesn't matter at all, earning is not CASH and Ebitda is a better proxy for cash! P/E multiple are not a determinant of value, but rather a function of value.

2.You mean EBITDA Multiple or EBITDA?

EBITDA will be the same, but the EBITDA multiple will be different, because, the one with high depreciation >> high capex will have to reinvest part of the CFO to growth the business and therefore trades at lower EBITDA Multiple. Growth doesn't come for free!

9/27/07

jameshunt:

EBITDA excludes depreciation so in your scenario both companies would have the same EBITDA. Keep in mind that EV is reduced by cash on the balance sheet so perhaps the EVs are different b/w the two companies.

Did not understand your second line ? How are EVs different ?

9/27/07

Ok if operating expenses are different, then EBITDA stays the same.

Company 1:

Rev: 100
COGS: 10

office expense: 20
Depreciation 20
Taxes & no debt as per your example, operating expense is 40 (20+20), but EBITDA is 70 (100 - 20OFFICE - 10COGS) Cool?

That should answer your question in the original post. EBITDA would be different if your operating expenses are the same while depreciation was higher for a second company. For example, if depreciation changed to 30, office expense must decrease to 10 in order to keep the 40 operating expense, correct? In that case, EBITDA is 80 now.

9/27/07

Thanks for such a detailed example !

9/27/07

crushkiller:

jameshunt:

EBITDA excludes depreciation so in your scenario both companies would have the same EBITDA. Keep in mind that EV is reduced by cash on the balance sheet so perhaps the EVs are different b/w the two companies.

Did not understand your second line ? How are EVs different ?

The 'price' in price to earnings is market capitalization which equals share prices X number of shares outstanding.

Enterprise Value (EV) is Market Capitalization (what we calculated above), plus minority interest and preferred stock, but minus cash. If two companies have the same market cap (and there is no minority interest/preferred stock involved) and one has a higher cash balance, the one with the higher cash balance will have a lower Enterprise Value.

9/27/07

Correct me if i am wrong here, but this should be very simple. All else equal, the company with higher depreciation would have a lower multiple.

Let's say Company A has an EV of $400 Million, and Company B also has an EV of $400 Million.

Let's also say Company A has EBIT of $150 Million, and Company B also has EBIT of $150 Million.

Company B, however, has $20 million of depreciation while Company A only has $5 million of depreciation.

Company A has an EV/EBITDA of 2.58x and Company B has an EV/EBITDA of 2.35x.

Think of it as a fraction really. If one has higher D&A than the other, and all else is equal, than that company will have a higher denominator, thus making the multiple smaller.

"An investment in knowledge pays the best interest." - Benjamin Franklin

9/27/07

Understood your rationale for the multiples in terms of EV/EBITDA ..
Could you please let me know which Multiple should I prefer for valuation - EV/EBITDA or P/E ...if company has not debt and taxes .. the tax rates and interest component would be out of picture ... so which multiple should one use to value a company ... in terms of preference and why ?

9/27/07

crushkiller:

Understood your rationale for the multiples in terms of EV/EBITDA ..

Could you please let me know which Multiple should I prefer for valuation - EV/EBITDA or P/E ...if company has not debt and taxes .. the tax rates and interest component would be out of picture ... so which multiple should one use to value a company ... in terms of preference and why ?

I would say still use EV/EBITDA. Since the companies do have D&A, it needs to be added back to reflect the "cash potential" from its operations. If you back out the D&A, which you would be doing with P/E, than your skewing your multiples a bit because non-cash charges should be added back for comparability purposes.

"An investment in knowledge pays the best interest." - Benjamin Franklin

9/27/07

If there's no debt and no taxes then EBITDA and the "E" from the P/E equation are the same with the exception of D&A. If the D&A includes a lot of funky amortization of intangibles and odd depreciation charges then you'd probably be better off with EV/EBITDA. On the other hand, if D&A is just straight depreciation for PP&E then P/E might be a better metric as it captures the capital cost of the business.

Keep in mind valuation is a relative thing. Regardless of the metric you choose, you need to make sure you can compare it to truly comparable metrics for other similar companies.

9/27/07

hildozang:

good analysts tend to prefer EV/EBITDA over P/E

Yeah, way to generalise.

Feeling ranty after a not so great earnings call, so sorry in advance. Anyway...

There are going to be a ton of people who disagree with this, but I think EV/EBITDA is only useful as a relative metric (and sometimes not at all), and fair enough, valuation is relative but you can't just use EV/EBITDA on its own. EBITDA is helpful when other measures of earnings further down the P&L are going to be obscured by inconsistent D&A across the companies you are looking at. An example of this is in mining, where capex tends to be very uneven over time, and as such, so does depreciation. Or when you have software companies that have different accounting policies with regard to capitalising development costs. Etc.

The whole 'cash potential' concept is completely retarded from an equity investor's point of view. If you're a debt guy and just care about how much cash you could potentially squeeze out within [x] time period if you need to, then fine. Screw capex, let the business decline, you don't care since once you're repaid at par, you're out.

But as an equity holder, you think about the long term prospects of the business - and there aren't many businesses where you can take out all of EBITDA every year as cash, and still maximise the long term value of the business. If you care about EBITDA, that implies you have D&A to add back, i.e. the company has depreciable assets, which, oh yeah, cost money to buy in the first place, and will cost you money to replace. EBITDA also ignores taxes, and again from an equity perspective, you probably want your company to be profitable, which means... it's going to pay taxes. EBITDA is just super misleading. Or as Charlie Munger calls it, bullshit earnings.

Not saying that P/E is perfect, far from it - and if you have a bunch of exceptionals / arbitrary non-cash charges
e.g. amortisation of acquired intangibles / etc, you should be adjusting for those things. The problem is that EBITDA just ignores a ton of pretty important stuff.

9/27/07

EV/EBITDA is a more clear-cut way to value a company because it doesn't include anomalies such as tax rates. P/E depends on your net income which can be manipulated and therefore may not provide a concrete base for a comparison.

"I like Ackman," Mr. Icahn said. "I'll tell you why I like him. Anyone that makes me a quarter of a billion I like."

9/27/07

These are only two in the grand scheme of things which I am sure you are already aware of. EBITDA is only one proxy of cash flow, meaning I hope you are taking others such as working cap, CAPEX, etc into consideration. If we are just looking solely at these two, then EV/EBITDA will paint you a much more complete picture of the company you are evaluating. (definitely used more in IB) In valuation terms, with comparable analysis you would typically use EV/EBITDA as one of your multiples, and in most cases its EV/OCF, EV/REV, EV/NI, EV/Assets, etc. Basically, do take P/E into account and for what its worth, but your go-to between these two measures is the EV/EBTIDA. There is definitely a lot more to it; I would recommend researching yourself a bit. As of right now, I am using multiples analysis and valuing a company with measurements such as EV/EBITDA...that's why I thought I'd try to help you answer your question. PM me if you'd like to see some spreadsheets with these multiples.

9/27/07

They are both most common valuation tools.
P/E is most useful for non-cyclical Industries,and is limited for companies with little or no earnings. Furthermore because net income depends on interest expense, tax regime, P/E can derive different results among companies of similar size and operating magin.
EV/EBITDA serves as a valuation standard for most sectors, and suits capital intensive industries and companies of quasi-monopoly. It is not fit to companies with fast change on fixed asset. By the way, it is independent of captial structure and taxes, as well as tortions in D&A.

9/27/07

Thanks guys for your comments!

Some good interesting points here by all of you. As Waymon3x6 and sigmahatsquared suggested, it appears that EV / EBITDA would be a more accurate way to value a company whose bottom-line could be distorted by depreciation, interest expense, or taxes. I presume the true operational performance would be evident by EBITDA. However, using this multiple, won't we miss the fact that a company doing higher capital expenditure may not always be optimal? Or for the fact that as an investor, he / she will only be concerned with net income and so interest & tax expense are vital information needed to make a decision? Shouldn't P / E then be a better alternative?

I have seen many research reports relying on P / E multiple valuation. I am not sure what is happening behind the scenes, but on those reports, the analyst would assign a target multiple for the company, and then come up with a 1 year Target Price. It doesn't appear that any other multiple was used at all during valuation. Perhaps, they were used only on a relative valuation basis, but not when coming up with a target price.

So, I'm curious. As an investor, it makes sense to look at Earnings Per Share and P / E. But maybe if there is an abnormal tax expense, or higher interest expense due to short-term debt, or maybe an exceptional expense, perhaps EV / EBITDA would give us a better picture when coming to a target value. But then again, at least EPS can be adjusted so that only recurring components are part of EPS. And then a P / E based valuation can be conducted.

I like the point that Aris_Joe made about cyclicality and that it suits an asset heavy industry. But I still think that ignoring depreciation (despite it being a non-cash expense) is not the right way to go about it. Perhaps EV / NI would be helpful? But then I haven't seen many reports using that metric at all.

Let me know your thoughts! Thanks again for your responses!

9/27/07

KoolSIM:

Thanks guys for your comments!

Some good interesting points here by all of you. As Waymon3x6 and sigmahatsquared suggested, it appears that EV / EBITDA would be a more accurate way to value a company whose bottom-line could be distorted by depreciation, interest expense, or taxes. I presume the true operational performance would be evident by EBITDA. However, using this multiple, won't we miss the fact that a company doing higher capital expenditure may not always be optimal? Or for the fact that as an investor, he / she will only be concerned with net income and so interest & tax expense are vital information needed to make a decision? Shouldn't P / E then be a better alternative?

I have seen many research reports relying on P / E multiple valuation. I am not sure what is happening behind the scenes, but on those reports, the analyst would assign a target multiple for the company, and then come up with a 1 year Target Price. It doesn't appear that any other multiple was used at all during valuation. Perhaps, they were used only on a relative valuation basis, but not when coming up with a target price.

So, I'm curious. As an investor, it makes sense to look at Earnings Per Share and P / E. But maybe if there is an abnormal tax expense, or higher interest expense due to short-term debt, or maybe an exceptional expense, perhaps EV / EBITDA would give us a better picture when coming to a target value. But then again, at least EPS can be adjusted so that only recurring components are part of EPS. And then a P / E based valuation can be conducted.

I like the point that Aris_Joe made about cyclicality and that it suits an asset heavy industry. But I still think that ignoring depreciation (despite it being a non-cash expense) is not the right way to go about it. Perhaps EV / NI would be helpful? But then I haven't seen many reports using that metric at all.

Let me know your thoughts! Thanks again for your responses!

I think your read of what was said is good, and I don't think that people were outright wrong, but what you have to understand is that there's no "best" metric for valuing a company. You have to look at the company itself, and you have to look at what you're buying. Like, there's a huge difference between an LBO and granny calling up her broker to buy a couple shares of the security, and you're gonna look at different metrics in each case.

PE guys and bankers care about EBITDA because its an unlevered cash number, right? It's a proxy for your ability to service debt and therefore your value as an LBO target. If you're buying out a company, EBITDA matters because that's what your lending syndicate is going to look to when they finance the buyout.

But if you're just trading the equity in retail sizes, why would you give a shit about EBITDA? All you care about is dividends/discounted future dividends, which are levered, post tax numbers. P/E multiple makes more sense.

Plus, depreciation matters. It matters from a GAAP perspective and it matters from a real economic perspective. I would counter your point that we should ignore depreciation with the point that that implies that all income producing hard assets are perpetuities, which is not correct in my opinion.

So idk, there's no "best." You use the right tool for the job.

9/27/07

Thanks for your input NYCBandar!

Well I am looking at an investor point of view, or a sell-side research point of view where the end user of that research are actual investors (institutional and retail) and not corporates looking for M&A.

I see your point where EBITDA wouldn't matter to the end equity investor. I have the same argument because I feel that at the end, investors are more concerned about money they earn out of their investment and EPS is a good indicator of that.

I suppose FCFE would be the best indicator, but I understand that not all companies would have positive FCFEs, especially growing companies. So using that to value a company may not be the right approach. So the closest we come to for an investor's perspective is EPS, or adjusted EPS taking into account any one-time expenses like restructuring or impairment.

Perhaps EBITDA is a good indicator for a very long-term investor? As EBITDA is considered a decent proxy for operating cash flows, more OCFs could perhaps mean a higher growth rate for the company going forward and that could trickle down to better EPS growth later on. But I don't see why I would use EV / EBITDA even in this case to come up with a target fair value 1 year down the line. Maybe I am stuck with my perspective. Some more thoughts could perhaps clear that out.

PS: I never made a point against depreciation. Especially I would keep an eye on it as it would give a good indicator whether the company is growing capex too fast? Or is there a chance of future increased revenues and the capex is increasing earlier? Though, it doesn't affect cash, I would still keep an eye on it.

9/27/07

EV/EBITDA:
Widely used ratio in comps to get the company value based on similar companies (you wouldn't use the average EV/EBITDA of a handful of tech companies as a proxy to value a utilities company).

Example of use: Exit multiple in a model. You forecast cashflows for 5-10 years in your DCF and then use an exit-multiple such as EV/EBITDA for your last year, which represents an amount you can reasonably expect in the future, given your projected EBITDA and what similar companies (size, industry, risk, etc) are worth today in relation to their own EBIDTA.

P/E:
Share price over EPS. Basically, you can see this metric as how much investors are currently willing to pay for a dollar of earnings. Different industries have different thresholds as to what represents a high PE and what is the norm. (Tech companies have higher PEs because of anticipated growth, which means that investors are willing to pay more per share because they hope that the EPS will grow at a high rate in the future.)

Hope this helps

"Those who know don't tell and those who tell don't know." - Michael Lewis

9/27/07

Equities In Dallas:

EV/EBITDA:

Widely used ratio in comps to get the company value based on similar companies (you wouldn't use the average EV/EBITDA of a handful of tech companies as a proxy to value a utilities company).

Example of use: Exit multiple in a model. You forecast cashflows for 5-10 years in your DCF and then use an exit-multiple such as EV/EBITDA for your last year, which represents an amount you can reasonably expect in the future, given your projected EBITDA and what similar companies (size, industry, risk, etc) are worth today in relation to their own EBIDTA.

P/E:

Share price over EPS. Basically, you can see this metric as how much investors are currently willing to pay for a dollar of earnings. Different industries have different thresholds as to what represents a high PE and what is the norm. (Tech companies have higher PEs because of anticipated growth, which means that investors are willing to pay more per share because they hope that the EPS will grow at a high rate in the future.)

Hope this helps

This is awesome. But I do have a question...can't we calculate EV in itself on the company we are valuing? What does the multiple EV/EBITDA tell us in this case? In other words...since we can calculate EV in itself without using multiples for our target company, why do we look at multiples? To tell us how many times (in EBITDA) the company overpaid for the company? Thanks!

9/27/07

Typically you project out financials for a few quarters / years and get an expected EBITDA. You then apply an EV / EBITDA multiple based on comps or any other kind of expectation and this gives you an expected EV. Using your FS projections, adjust the EV to get back to equity value and you get derive an implied share price. The multiple is used because it's easier to justify your valuation based on an income statement and balance sheet with growth percentages rather than just plucking an EV number out of thin air.

In reality - you will have a share price you want to justify and tweak your assumptions working backwards to prove that this is reasonable.

9/27/07

Asatar:

Typically you project out financials for a few quarters / years and get an expected EBITDA. You then apply an EV / EBITDA multiple based on comps or any other kind of expectation and this gives you an expected EV. Using your FS projections, adjust the EV to get back to equity value and you get derive an implied share price. The multiple is used because it's easier to justify your valuation based on an income statement and balance sheet with growth percentages rather than just plucking an EV number out of thin air.

In reality - you will have a share price you want to justify and tweak your assumptions working backwards to prove that this is reasonable.

I guess my question was more along the lines of..we can calculate EBITDA for a company, and we can certainly project these values out for the next year (or the company can, and give us the financials.) Can't we also calculate EV for a company? Can't we also project out EV using financials the company gave us or we calculated? Or is this kind of impossible to calculate for even today + forward? Because my issue was...if we can calculate the EV now/forward..why are we using multiples based on other companies to tell us what EV is when we can calculate it exactly ourselves? Thanks!

9/27/07

Think more conceptually and less academically. What is EV? It's simply the market cap of a company, adjusted for net debt (and other LT liabilities). How can we project this if we don't know the future market cap (which requires us to know the share price)? Instead, we assume that the value of the company is in some way related to an underlying metric (EBITDA in this case as we are being capital-structure neutral) and THEN adjust back.

Consider this. If I give you the expected financials for a company in 2 years time, how can you determine the EV? What would you use? EV is just (Shares x Share Price) + Adjustments. You know shares, you know adjustments but you don't know share price. We use a multiple as a way to back into an implied share price.

Try it yourself. Take any company you want, look at their financials and try and tell me their EV without using the share price.

9/27/07

Asatar:

Think more conceptually and less academically. What is EV? It's simply the market cap of a company, adjusted for net debt (and other LT liabilities). How can we project this if we don't know the future market cap (which requires us to know the share price)? Instead, we assume that the value of the company is in some way related to an underlying metric (EBITDA in this case as we are being capital-structure neutral) and THEN adjust back.

Consider this. If I give you the expected financials for a company in 2 years time, how can you determine the EV? What would you use? EV is just (Shares x Share Price) + Adjustments. You know shares, you know adjustments but you don't know share price. We use a multiple as a way to back into an implied share price.

Try it yourself. Take any company you want, look at their financials and try and tell me their EV without using the share price.

This is making more sense..but what about for consideration of a private company? And what about in times when we are valuing in the present sense and not forward? (Or do we ALWAYS/MOST OF THE TIME project out financials and use trading comps to calculate multiples for future earnings? I wasn't sure if this happened most of the time or we tried to use multiples to value a company now.)

Side note: Would you subtract net debt in all these calculations and multiples to come to a figure of "equity value" and would that be useful in determining a range of how much the equity in the company is worth to investors in a potential recapitalization?

Thanks!

Best Response
9/27/07

Private Company
Same concept except instead of trying to work out a per-share value, you're looking for the value of the company. All you do is avoid the last step which is market cap / NoSh, and this gives you the value of the company. Investors in private companies typically buy % equity stakes so if your private co (say Dropbox) is valued at $10bn based on EV/EBITDA and some private equity company comes along wanting to buy 5%, 0.05 x 10bn = $500m.

comps vs Multiples
I don't think you've quite nailed the concepts here. All valuations (pretty much) are done on multiples, the only reason we use comps is to establish a range of reasonable multiples for a similar sized company in a similar industry. You can do this going forwards or in the present, it doesn't matter. All you're doing is applying the reasonable multiple to an underlying fundamental factor of the company.

Recap
Yes.

Basics
For your understanding I think it might be helpful to do a very quick walkthrough of how the valuation process would be done.

Lets say you want to try and value AAPL both currently and with an 2014E share price (the numbers I will be using are not even close and purely illustrative). Picking numbers out the air and saying "I reckon it will trade at $600" is ridiculous, you might as well throw leaves into the wind. Instead we can look at the fundamentals and use a sensible estimate.

Now you go and build out your operating model for the past few years and project forwards to the end of 2014 using reasonable assumptions derived from discussions with management, supply chain, etc. For the sake of argument lets say you get to 2013 EBITDA of $100, 2014 EBITDA of $120, 2013 net cash of $150 and 2014 net cash of $200.

Next, we need to see what a reasonable multiple to trade at is. We've decided to use EV/EBITDA. To do this, we can look at some competitors and try and find a reasonable range of EV/EBITDA multiples. We choose Samsung, Google, Sony and Microsoft (imperfect list I know). The average of these EV/EBITDA multiples is 10x with a lower range of 8x and higher range of 12x.

Now, we can apply these multiples to the EBITDA of AAPL to come up with estimated figures - using averages we get an EV of $1,000 for 2013 and $1,200 for 2014. Now we adjust for net cash and come with a market cap of $1,150 for 2013 and $1,400 for 2014. Divide by shares outstanding (200 in this scenario) to get 2013 share price of $5.75 and 2014 share price of $7. We can then create a range of share prices using our lower and higher EV/EBITDA multiple ranges.

Hopefully this lets you see how we actually use multiples and comps to get an EV and then reverse back to share price. You can't just project out financials to get an EV, you have to have some kind of market-related metric to apply or your numbers are fairly meaningless. The art and skill in valuation is not creating models but in making the numbers / projections realistic and relevant.

9/27/07

Asatar:

Private Company

Same concept except instead of trying to work out a per-share value, you're looking for the value of the company. All you do is avoid the last step which is market cap / NoSh, and this gives you the value of the company. Investors in private companies typically buy % equity stakes so if your private co (say Dropbox) is valued at $10bn based on EV/EBITDA and some private equity company comes along wanting to buy 5%, 0.05 x 10bn = $500m.

comps vs Multiples

I don't think you've quite nailed the concepts here. All valuations (pretty much) are done on multiples, the only reason we use comps is to establish a range of reasonable multiples for a similar sized company in a similar industry. You can do this going forwards or in the present, it doesn't matter. All you're doing is applying the reasonable multiple to an underlying fundamental factor of the company.

Recap

Yes.

Basics

For your understanding I think it might be helpful to do a very quick walkthrough of how the valuation process would be done.

Lets say you want to try and value AAPL both currently and with an 2014E share price (the numbers I will be using are not even close and purely illustrative). Picking numbers out the air and saying "I reckon it will trade at $600" is ridiculous, you might as well throw leaves into the wind. Instead we can look at the fundamentals and use a sensible estimate.

Now you go and build out your operating model for the past few years and project forwards to the end of 2014 using reasonable assumptions derived from discussions with management, supply chain, etc. For the sake of argument lets say you get to 2013 EBITDA of $100, 2014 EBITDA of $120, 2013 net cash of $150 and 2014 net cash of $200.

Next, we need to see what a reasonable multiple to trade at is. We've decided to use EV/EBITDA. To do this, we can look at some competitors and try and find a reasonable range of EV/EBITDA multiples. We choose Samsung, Google, Sony and Microsoft (imperfect list I know). The average of these EV/EBITDA multiples is 10x with a lower range of 8x and higher range of 12x.

Now, we can apply these multiples to the EBITDA of AAPL to come up with estimated figures - using averages we get an EV of $1,000 for 2013 and $1,200 for 2014. Now we adjust for net cash and come with a market cap of $1,150 for 2013 and $1,400 for 2014. Divide by shares outstanding (200 in this scenario) to get 2013 share price of $5.75 and 2014 share price of $7. We can then create a range of share prices using our lower and higher EV/EBITDA multiple ranges.

Hopefully this lets you see how we actually use multiples and comps to get an EV and then reverse back to share price. You can't just project out financials to get an EV, you have to have some kind of market-related metric to apply or your numbers are fairly meaningless. The art and skill in valuation is not creating models but in making the numbers / projections realistic and relevant.

Definitely puts things into perspective. So would you say the end value - after we've applied the multiples to AAPL to get market cap + lower-high share prices, would this # (market cap) be the predicted underlying cost to acquire the company? I'm assuming the share price would just tell you the same thing/value as you'd just multiple share price * outstanding shares to get a pretty good estimate of what the company is worth? As in, the same or similar value would be had from market cap using multiples and share price using multiples..and this would tell you how much you should pay for company at the most basic level? Thanks!

9/27/07

Enterprise value is the true cost of acquiring a company. Market cap is simply the amount it would cost (in theory) to buy 100% of traded shares and therefore own 100% equity in the company (ignoring issues like treasury stock etc.). In theory, assuming no transaction premium, you acquire a company for the market cap but you take on their balance sheet so you add on net debt (i.e. any net cash will reduce the final cost and any net debt will increase it).

I think you're trying too make this into too much of a financial / theoretical concept and ignoring the big picture / basics of what these terms actually mean. 'Multiples' is just basic maths, X/Y, its not some arcane divine financial concept.

I'm not entirely sure what your final question is.... Market cap using multiples and share price using multiples? These are basically the exact same number, one is just divided by number of shares. Consider P/E multiples. If you have earnings of $500 and shares outstanding of 100, your EPS is $5. Applying a 10x P/E multiple to either earnings or EPS will give you exactly the same outcome - a market cap of $5,000 or share price of $50.

9/27/07

Not necessarily overpaid, but ebitda multiples tells you how much the company is valued. You wouldn't say you buy a company for 2x EV because that means you double count certain factors and that would make no sense, such as minority interest, cash, etc.

9/27/07

explosions09:

Not necessarily overpaid, but ebitda multiples tells you how much the company is valued. You wouldn't say you buy a company for 2x EV because that means you double count certain factors and that would make no sense, such as minority interest, cash, etc.

Tell us how much the company is valued at in terms of EBITDAx(Multiple) = Implied EV? My issue is..this is kind of like saying I have a company that's sole asset is a $100 bill. No offices, no employees, just a $100 bill. Let's say the $100 bill makes money..$1 a year. EBITDA = $1, and EV = $100. EV/EBITDA = $100. In other words..why would I go look at other companies comprised of $50, $100, $500 bills for example..and use their multiples instead of just looking at the true value of what my company is worth?

9/27/07

you should be very thankful the fine folks on this board has taken the time to answer this question.

by you asking, it tells me 2 things:

1. you are not self learner
2. you are dumb enough to ask this question

all of your questions can be answered within 10 minutes of googling, or 15 if you are slow.

9/27/07

Whiskey5:

you should be very thankful the fine folks on this board has taken the time to answer this question.

by you asking, it tells me 2 things:

1. you are not self learner

2. you are dumb enough to ask this question

all of your questions can be answered within 10 minutes of googling, or 15 if you are slow.

Obviously I'm dumb enough to not know the answer, so I asked. But, it's all relative isn't it? I mean there's kids sitting in their room not even studying for these interviews..I've at least taken the initiative to ask and try to learn. The information we learn in school is readily available within 10 min. of googling as well, but there seems to be a consensus that having someone explain these things to you is attractive to some people who still pay to go to school....

9/27/07

.

"I do not think that there is any other quality so essential to success of any kind as the quality of perseverance. It overcomes almost everything, even nature."

9/27/07

1. If the stake increases, EBITDA multiple goes down and vice versa.
2. No impact.

"I do not think that there is any other quality so essential to success of any kind as the quality of perseverance. It overcomes almost everything, even nature."

9/27/07

thanks
1. i am not talking about increasing/decreasing. i am talking about at a specific moment in time. what do you mean by that?
2. why no impact? i thought your net income includes the consolidated part that you won't get.

9/27/07

2. I am talking from the perspective of increase/decrease again.

"I do not think that there is any other quality so essential to success of any kind as the quality of perseverance. It overcomes almost everything, even nature."

9/27/07

Q1. You could, but we do not subtract the minority from EBITDA. We simply add it to EV for the sake of convenience and consistency.

"I do not think that there is any other quality so essential to success of any kind as the quality of perseverance. It overcomes almost everything, even nature."

9/27/07

well, speaking of convenience, you just go to CapIQ or pull it with the excel plug in..
what i am asking is when you are at much later stage in the process and require a precise and correct number. but i guess i am not making myself clear enough since i don't understand your point about increasing or decreasing.

9/27/07

found it very well explained at another side

9/27/07

P/E because the difference between EBITDA and Earnings is bigger than the difference between EV and Market Value.

9/27/07

also.. not every company give out dividends and/or appreciates in stock value (=> infinte PE) but pretty much every company has some EBITDA (=>finite EV/EBITDA)

9/27/07

...

9/27/07

PE is higher than EV / EBITDA.

Think of a company that has no net debt. Market cap = EV
Earnings = EBITDA - D&A - Interest - Tax

(Above = quick interview answer, below more in depth interview answer - notice numbers selectively chosen for simplicity)

Introduce Debt @ 2.0x EBITDA at 10%
Assume the following:
EV = 100
EBITDA = 20
D&A = 5
Tax = 33%

Leverage = 40
Interest Expense = 4

EBT = (20-5)-4 = 12
Tax = 4
Earnings = 8

Market cap = 100 - 40 = 60

P/E = 60 / 8 = 7.5x

Versus previous P/E of 10x (market cap of 100, earnings of 10) (implied Ke of 10%, vs. kd (1-t) of 7%)

Increasing leverage decreases quality of earnings reducing PE. Another way to think about it is introducing debt implicitly reduces the blended multiple vs. the prior P/E multiple with low cost of capital. Not a clean analogy (because there is other crap: D&A, tax in there), but levered multiples are generally lower.

9/27/07

lol did u really dedicate a thread to express your disdain for IBD only to come back to ask a question... Either way this is awkward.

"Why Investment Banking? Honestly i really cannot see the appeal in this field. Lets say you do make it through the ropes of finance, you work your ass off for 20 years, retire at around 40, but then what? Go to your vacation home in the Caribbean and just sit there with your trophy wife. Not to mention, the job itself will be extremely boring for 99% of you. I'm sure when you guys were little kids/teenagers, you didn't imagine yourself working in such a boring field that has no contribution to society."

You speak in in varying levels of verbosity.You often adopt the typing quirks of others as you find it boring to settle on styles.

9/27/07

Net Debt + Equity (market value) over
Earnings + Taxes + Interest + Depreciation + Amortization

is growing, while

Equity (market value) over
Earnings

is shrinking.

When you break it out that way, it should be a little easy to come up with some possible ideas.

9/27/07

Possible explanations: low debt, high cash, and capital intensive. That's a very rare situation. Would be interested to hear the context on this company?

9/27/07

What do the earnings look like? Are they depressed? Is the P/E high because the E is low? Otherwise, as mentioned above, seems like its a biz that seems rare (low debt, high cash, capital intensive - lots of depreciation?)

Hard to tell without knowing more about the specific situation. You should dig deeper into the financial statements to understand what is driving these valuation metrics and why the valuations are the way they are

9/27/07

Possible explanations: low debt, high cash, and capital intensive. That's a very rare situation.

Actually it's not that rare at all. This is a very common situation for companies that have significant financial assets, i.e. cash or holdings in other companies. For example, Yahoo owns part of Ali Baba and thus has a large market cap. However, Yahoo's core business is worth very little, possible even negative, so Yahoo's EV is quite small.

A simple way to think about this: Market cap = EV + net financial assets

Yahoo market cap = core Yahoo business + Ali Baba stake

Since the core Yahoo business produces virtually no earnings, the P/E ratio for Yahoo is quite high, in fact >100x. However, Yahoo's EV/EBIT is low once you adjust for the Ali Baba stake. This is one of the reasons that P/E is not a great metric.

9/27/07

Very true, great point. I would argue that Yahoo is a special situation though, as most of these types of scenarios would be. Its primarily limited to holding cos. and special cases where non-core assets are a large chunk of EV, like you said. Most typical operating companies will not show these characteristics, which I think was the point of saying "rare" (at least, that was the intention in my use of it.. I could have worded that more accurately)

Not saying these can't be good investment opportunities, just thats its not your run of the mill operating company. Usually when two metrics are showing opposite indications, you need to look deeper into the statements to figure out whats going on

9/27/07

Not sure I follow the Market cap = EV + net financial assets?

Market cap is just referring to equity value no?

9/27/07
9/27/07
9/27/07
9/27/07
9/27/07
Add a Comment
WallStreet Prep Master Financial Modeling