The closest accretion/dilution comes to taking into account changes in the cost of capital is through projected changes in interest expenses, but as far as I can tell it doesn't have any way to factor the changes in the cost of equity capital. I'm also not aware that analyst EPS estimates factor cost of capital at all – they may forecast “risk” of some sort through probability weighting multiple scenarios, but this is not the same as Beta-based costs of capital under the CAPM. Additionally, relying on the availability or accuracy of analyst EPS estimates isn't really relevant to the question, since Sebasm is essential asking about the change to the *value* per share, not the earnings. To the extent that the target's business differs from that of the acquirer, it could have a significant change on the beta used to determine the acquirer's cost of capital. Put in less theoretical terms, it could significantly change the peer set against which the acquirer is compared. One example is the United Online acquisition of FTD (http://paidcontent.org/article/419-united-onlin...). Yahoo Finance may still be categorizing UNTD as an ISP, but they get 60% of their business from selling flowers.

Sebasm: I'm doubly impressed with the question coming from a first-year undergrad. I couldn't find your other comment, but feel free to reach out if you want to discuss any other related topics. With my full name I'm pretty easy to find on the internet.

The one thing to keep in mind in private equity is that good returns come from selling an investment for more than you paid. That increase in price doesn't always means there's an increase in value. Buffett may say that “price is what you pay, value is what you get”, but he's talking about long-term investing. In private equity, “price is what you pay, and price is what you get.” Value helps, but it's far from the decider of returns.

I agree, theoretically DCF takes more into account, but multiples and DCF suffer from the same problem, which is “garbage in, garbage out”. The discount rate in a DCF affects the value greatly, yet it's about as the most arbitrary as all financial metrics.

This is going to sound somewhat cynical, but the calculated financial value of a business is pretty trivial in private equity. It's based on past metrics and any attempt to include the future is futile. In university you want to believe it's as formulaic as the professors suggest, but it just isn't.

The real-life drives of good returns in private equity are purchase price, sale price and leverage. Sales growth matters, and so does profit growth, but we're mostly talking about legacy businesses in PE. Buying in at 4 times compared to 8 times and selling at 12 times compared to 6 times makes a big difference. All of the financial models and DCF calculations in the world won't predict the manipulation of a great purchase/sale price.

All the best S – just my opinion

I think Accretion / Dilution is definitely not even close to be a indication of value creation (or destruction) because it leaves out the principle of risk-adjusted and time-weighted (discounted) returns necessary to measure value creation or destruction.

So, for instance, lets say (to cite your example) 3M buys company ABC and earnings get around 1% accretive, but bought a Colombian company, so the risk of those pro-forma earnings also carry a lot more risk, and for that 1% increment in EPS 3M paid a 15x multiple, there would be clearly value destruction.

Sorry if I didn't get your point, as I stated before I'm a first year undergrad in Colombia and I don't have the knowledge of you wall st guys.

Now, about the topic… I find the multiple approach a very dangerous approach because of the reasons it's actually used. I should point out also that a DCF valuation, if done right, should yield the same value that the multiple approach and vice-versa.

One of the reasons the multiple approach is so widely used, which in my opinion make the multiple approach dangerous is “because it's easy” and don't have to calculate all the inputs the DCF have. Another is the misuse of the multiples, for example, the P/Sales or P/EBITDA used when earnings are negative is a mistake because price is an equity value and sales is a pre-debt firm value, there's no possible sensible interpretation possible because I'm leaving out all debt incidence in value.

Another is the argument for those comparable companies, because to have the same multiple, the company must have the same amount of debt, same growth, same revenues, etc… than the average.

So the conclusion for me would be, if multiples are to be used… then it'd be good to make a DCF valuation also just to support the multiple, know what am I selling and make adjustments to the company Vs the comparables and arrive to a fairer value.

Again congrats and thanks for the blog…
1st year undergrad from Colombia.