when to use which valuation method?

I believe Dillution/Accretion, LBO and Comp-transactions are three prominent models. Depends on the structure of the deal. For instance, if a firm is buying another all-cash, then things aren't necessarily as complex (at least on the debt service side).

I'm not in M&A. Any thoughts from anyone else?

DCF is flawed but due to the subjectivity that is involved in that type of analysis you need to be certain that your assumptions are based on conservative market trends. Public company comparables and transaction comp analysis are the other two primary valuation methods. Transaction comp analysis usually yields the highest valuation (as it includes a takeover/control premium, with DCF being the next highest (due to optimisitc subjectivity), and comps usually yielding the lowest value.

Generally, you will use at least two of these to ensure consistency. We almost always build an initial comp set and then run a DCF.

My apologies, I meant to say that DCF is not flawed but...

The people I respect the most are the ones that understand building a model isn't necessarily to come to one single value (because even conservative/"accurate at the time" projections rarely work out), but instead to understand a range of values where you are comfortable, and to understand that if you want a certain valuation, this is the growth/CF/etc. you need. The sensitivity to a DCF is the most interesting part.

For instance, I believe 2 years ago when Google's price was over 700, the valuations called for growth of 60-70%. While they're still red hot, obviously that valuation was a litle bit innacurate.

Alphaholic's comment above hits the nail on the head. It's all about building a "football field." This gives you a range of value. It's important to note here there is no definitive answer when it comes to valuation. One can certainly be completely off base, however if you have logical and rational assumptions behind your DCF and have completely scrubbed all of your multiples in your comparable company and transaction analyses you can always make an argument for your valuation.

FYI, I typically use 3-4 valuation methods in my pitches: DCF, LBO, Comp. Transactions and Companies.

Repeat after me: accretion/(dilution) analysis is not a valuation methodology.

Thank you everybody for the replies! If someone could elaborate more on building a "football field", I would very much appreciate it.

A football field is just a rectangular grid (chart) that shows your valuation ranges. For example, it will have a bar for something like: IEV for '08 Revenue from public comps with the lower limit of the horizontal bar being at the low of your range (generally ~(15%) below median) and the upper limit at the high end of your range (~+15%). It will have bars for your diff. years of rev, ebitda, and your M&A comps or whatever you are showing in your valuation.

what do you mean by come up with a company, do you create your own imaginary firm? in that case, just run a DCF, because your growth assumptions are entirely up to you.

Highest and best use?

I think the cost approach is more applicable to specialized properties, such as certain medical office buildings or biotech offices. There is value in the build-out of the lab space and some equipment and fixtures that run with the building.

With that said, I echo what econ said above. Even if you have a highly specialized building with a lot of technology, doesn't matter how much it cost you if no one wants to lease the space.

We use the cost approach on almost everything we buy to ensure that we are not buying above replacement cost. That is it. Most teams, if large enough, will have a good costing guy or engineer that does this well and you will just need to find market prices for land. Sometimes you can back into a land value by solving for a market development yield if there are no land comparables. Otherwise there are some high level costing software and manuals available for purchase. Appraisers need to use the cost approach quite often to appraise special purpose buildings with no lease in place or an unstable lease that cannot be quantified. As an investor, all you care about is yield

They will probably ask you what NOI is. Don't confuse NOI with net cash flow or even net profit. It is purely a real estate metric that is the operating income of the property before all capital, TIs, leasing commissions, other one time expenses and debt service. Essentially only the items that happen every year. It is kind of BS metric if you ask me, but it seems to be pretty much industry standard.

Many investors will also look at the cash on cash in a particular property, especially if they are looking to use it for cashflow.

Others doing value-add investing will not look at cash on cash and will simply look at IRR over a 3-5 year holding period (shorter term) as the capital that you are injecting into the property will hammer the cash flow.

In both of these cases NOI is irrelevant.

NOI is important because it's the equivalent of EBITDA and a good proxy for a property's unlevered free cash flow. It makes sense that the LONG-RUN ability of a property to generate NOI will be the principal barometer of its STABILIZED value.

Mogul's point is well-taken, though, since many/most value-add investors are buying properties that generate little-to-no NOI, and will make little-to-none of their profit from the cash flow from NOI yield (or its levered derivative, the cash-on-cash yield). Rather, they will try to stabilize the occupancy, so that the property is ready to spit off NOI, and then sell it to a core investor who will buy it for its ability to produce NOI.

As to the cash-on-cash return, the NOI is still fairly relevant since the cash-on-cash return will basically be a function of the cap rate (NOI yield) you can acquire the property at, affected by the amount of leverage you can add and the cost of leverage.

I got this one a couple times and just gave the bassssssic inputs, proccess and output for each. Like literally about 4 minutes at most. The more you give them, the more they grill you and such. Be accurate, concise, and give them the big picture while showing you understand some of the underlying calcs... comps and precedent should take you 30 seconds a piece I guess DCF depending on how you attack it will take a minute. I had an interviewer at a superday ask me to break down the DCF further and I did. Then she asked me more in detail about issues that can arise at arriving at WACC, etc. They will press you if they need to, don't go further than you need.

I hate saying do a search, but there was a post a few days ago about the step-by-step way to approach a DCF analysis. That should give you a good basis for the other two as well.

Search on google for "Investment Banking: Mergers and Acquisitions, Valuation and Leveraged Buyouts" and look at some info. If you have enough time I would actually buy that book. It will teach you everything about these topics in-depth and you will ace any type of technical questions with superior knowledge.

Just follow the advice of the first guy that posted.

If you feel your answer may not have been detailed enough, after you finish your explanation ask the interviewer "is there anything you would like me to discuss in more detail?" This way you don't look like a rambling fool, but you still show you're confident and know enough to answer tougher questions on the methods of valuation.

MissingNo is on the right track, but 4 min is still long as shit. Absolute maximum of 2 minutes is more like it, and just ask them if they'd like more detail. If they do, they'll ask. No one wants to hear you ramble.

I'm surprised they ask for 4. Most people here will probably only ever use 2 or 3 IMO. Those are: DCF, comps (using EV/EBITDA), and comparable transactions (though I doubt that you'd find a comparable transaction for Amazon). The only other one I can think of is residual value.

So you haven't gone over any valuation methods in class?

(1) DCF

(2) Comps

(3) precedent transactions

(4) LBO

Dude you literally posted this exact same thread two days ago - actually word for word. Do your own work.

you asked this two days ago .. answers don't change:

http://www.wallstreetoasis.com/forums/valuation-suggestions-for-project…

i'll try it out where did you hear about this exactly?

initially i found it on youtube, The guy from the video then provided a a link to his website (i think). It didn't give the specifics of the method, but it gave free spaces to input the numbers. I punched in some figures and then I cracked the formula. From my experience this method gives me a greater value then my initial analysis, so i am little skeptical, due t my conservative approach. But let me know what do you get.

This is basically just finding the plow back rate of the company and forecasting BV. Also forecast shares outstanding and then you can find out a $ price per share. You then discount that dollar pice by the annual return you want to obtain and then you can find a buy price that satisfies your returns. Kind of what you said but this is what Buffett has been doing all along.

Yea I have used this method and in terms of personal valuations I think it is a good one at least for large stable companies. Remember though you always want a margin of safety. Buffet typically uses 15% to get his buy price, but at today's valuations I would bet it would be hard to find stocks in that range. Also use a 10 year period for projecting and discounting back. Hope that helps

have you looked at the vault guide to finance interviews?

You're close on the EBITDA multiple, but you have it mixed up. Here is how it works:

It is typically called "Comps" or Comparable Company Analysis. The idea is that if the company you were trying to value was public, what would it be worth? It is very easy to tell how much a public company is worth, all its financials are publically available. So what you do is you find a whole bunch of public companies that are similar to the one you are trying to value (this can get creative sometimes), and you look at the EBITDA multiples (you can use Rev or EBIT as well, sometimes people value Tech deals off of revenue, especially if they have a negative EBITDA). EBITDA / Enterprise Value is going to get you a number between 0 and 1, that isn't very helpful. What you want is EV / EBITDA. So if the median EV/EBITDA of your public company set (you don't use just 1!) is say, 8.0x, and your company has an EBITDA of $100m, you could say the value of your company is $800m.

Got it?

As far as picking whether to value based on P/E or EBITDA - well, typically that's an insight based on some time in the industry and experience with how these things are traded. Heck, some things like internet start-ups may even be valued off of revenue multiples. You haven't even gotten to asking whether it's valued off 1-year or 2-year forward multiples.

If you want to get into some technicalities, here's a useful wrinkle to throw into the mix: don't forget who the metric is applicable to when thinking about whether the multiple is off of aggregate (or enterprise) value or equity value. If the cash will flow to all holders in the company (ie EBITDA is before interest and taxes, and so it will eventually pay off holders of both debt and equity stakes) it should be a multiple of agg value. Thus EBIT, revenue, basically any statistic created above the Interest line is an agg multiple. EBT (earnings before taxes but after interest), net income, etc. will be, then, multiples of Equity value. Conceptually, you've already paid off debt holders as these metrics fall below the interest line item.

multiples aren't intrinsic - they are relative.

LBO sum of the parts (can be intrinsic or have relative aspects depending on how you value lines of business/assets) liquidation value (not purely intrinsic; similar to sum of the parts in that way)

Liquidation value, Comps