M&A Models from Scratch

People have a tendency to overcomplicate merger models, and template examples on the Internet are also very complex in order to generalize and cover every esoteric contingency there is. It is difficult to build a general M&A model because every company, every capital structure, etc. can be wildly different, which means most generalized models are too long, too complex and too cumbersome for most actual investment banks in practice, especially during live deal situations.

The bulge bracket bank I worked at had this ridiculous general M&A model that we used for pitches. The thing was probably 5 megs with nothing in it, which is unacceptable. My FactSet would crash ~40% of the time when I opened it. I then worked for an EB where models were extremely streamlined, fast, and (in my opinion) far better. But they were custom-built for each situation. I now also create custom models for each situation on the buyside.

Assuming you have a solid foundation in corporate finance (e.g. can build a DCF), there are really only two things to keep in mind when you're building an M&A model from scratch:

What's the objective? Most times, it's public buyer accretion/dilution, but this can vary (for example, if you're modeling a private-public acquisition and trying to see how this will impact your returns)

The ONLY thing that matters in each line item is "how does this impact the pro forma math?" In other words, if you're trying to determine public buyer accretion/dilution, every line can be intuitively deduced with the question of "how is this impacting accretion/dilution"?

Your pro forma items (PF revenue, EBITDA, etc.) are just acquirer + target, add synergies if you want to include those assumptions (for your first model, just make it simple and don't add any. Or just add 10% of sales if you desperately want to). Operating line items (GP, EBITDA, EBIT) are a little bit trickier - there are companies that report EBITDA differently (for example, pre- or post-stock based compensation expense, since EBITDA is a non-GAAP metric), so you'll want to make sure your target metrics have been adjusted to conform to buyer methodology (if your buyer is pre-SBC, you should adjust your target to be pre-SBC, as everything in your model should conform to the buyer's perspective). But again - just simple addition (buyer EBITDA + target EBITDA (conformed to buyer reporting methodology) + synergies = pro forma EBITDA).

When you get to non-GAAP EBIT, you should net out non-GAAP adjustments to arrive at GAAP EBIT (check your acquirer's and target's 10Ks and 8Ks to see definitions - for tech companies, this mostly amounts to whether or not companies add back stock-based compensation and amortization of intangibles). This way you can take out interest and arrive at a GAAP pre-tax income, which you can use to figure out GAAP taxes.

Your interest expense is the pro forma interest expense. People go overboard on this line item in particular but you should make it as simple as possible. What's the acquirer's existing capital structure? How is the acquisition being financed? What's the go-forward capital structure post-transaction? When you're building your first model, please don't go through the effort of creating multiple new tranches of debt (for example, this many turns of senior debt and that many turns of mezz, etc.) - that's overcomplicated. Just throw on a single new illustrative tranche of debt with no early prepayment (therefore you don't have to model optional repayments) at 5% interest and call it a day. You can add complexities as your modeling skills mature. If you don't yet know how to build an LBO model, just take existing interest expense, keep it constant and layer on the interest expense from new debt. I know that's a shortcut, but the cash flow waterfall down through debt paydown requires the most work, so avoid if you want to get a general idea of how to build an M&A model before getting more advanced technical skills. That's pro forma interest.

Take out taxes. Just use the buyer's tax rate, and assume the historical tax rate remains consistent, and (for your first model) pick a company that is profitable and pays normal taxes. Or just use 35% as an assumption. Please don't pick a company with NOLs, as that will just trip you up.

Now you have GAAP net income. Here's a tricky part - add back tax-affected non-GAAP adjustments. You should have the numbers you used to go from non-GAAP to GAAP EBIT (you subtracted them, remember?), just add them back, but post-taxation (multiply by 1 minus the tax rate).

Now you have non-GAAP net income. Divide by pro forma shares outstanding (did you issue shares to pay for the transaction?) and voila, you have PF EPS and accretion/dilution.

There's a reason a lot of banks will ask for paper accretion/dilution models in interviews. They're not difficult. You can even just start with EBIT, layer in the pro forma capital structure using some fake assumptions and arrive at an acc/dil result.

By far the most complex portion of a "full" M&A model is the debt paydown, which requires a more fundamental understanding of cash flow. Honestly, in that case, you should learn to build an LBO model before tackling this.

Again, even with LBOs - start with the very basic fundamental skeleton of a model just to understand how everything flows before adding complexity. When adding features, just ask yourself "how does this flow through?" and you'll be alright.

I know the above is convoluted and long and confusing, so if you have questions I'd be happy to answer PMs. Also feel free to send me an initial model (if you think it's too bad to send, trust me I've seen worse so don't worry about that) and I can review.