Depending on who you ask, mergers are:

1. Value-destructive, since buyers pay a very real premium for hypothetical synergies,

2. Value-creating, if done right, because they allow two complementary businesses to function together, or

3. Value-transferring, in that they enable industry consolidation that transfers consumer surplus to producers.

There are good case studies for all of these. AOL/Time Warner was the biggest merger of all time when it was announced, and led to the biggest GAAP loss in history a few years later when AOL's goodwill was written down. Bank mergers in 2007-8 also had a rough run, with RBS buying ABN Amro and promptly needing a bailout, and Bank of America trying and failing to catch a falling knife by buying Countrywide in early 2008, when subprime was obviously doing badly but looked contained. (They got it for $2.5bn in stock and paid an order of magnitude more in fines, put-back costs, settlements, etc. Financials, like oil futures, have no trouble going below zero.)

On the value-creating side, the #1 search engine in most markets (Google) bought what would eventually be the #2 search engine in most markets (YouTube) back when YouTube looked like it came with about the same magnitude of legal liabilities Countrywide would turn out to have. But Google (disclosure: long) correctly bet that, as a tech company that was good at scaling, they'd have a comparative advantage in the bandwidth- and storage-intensive business of serving video, and that as an ad company, they probably ought to have an ad product for video given how much ad revenue that category produced.¹ Meta (disclosure: long) was able to execute defensive acquisitions like Instagram and WhatsApp to keep their position in photo sharing and chat, and executed other crafty deals, too, like acquiring a mobile VPN company in order to create an in-house intelligence agency showing them which other apps were taking off. The company now known as Booking Holdings used to be Priceline, but Priceline's acquisition of Booking gave them a better position in Europe and some unbeatable search marketing expertise. Disney has a great track record of finding companies with their own Steamboat Willy and turning that into Mickey Mouse, which they've done with Pixar (whose film slate was five originals and one sequel at the time of the acquisition—since then they've done 13 more originals and 9 sequels).

And there are some deals that are probably close to value transfer. For a long time, US markets were basically subsidizing cheap air travel and nice union contracts in the airline sector. As Buffett used to point out, the industry had been net unprofitable²  from inception at various points in the 2000s. But once they consolidated into three big network carriers, one big discounter, and a handful of small players who the incumbents knew how to compete with.³

There's a whole taxonomy of mergers and merger approaches, and there are companies that have built up their business one acquisition at a time—someone like Constellation Software or Jack Henry don't have that many standout successes or misses, because they have an excellent playbook for doing repeatable small-scale acquisitions and letting them run fairly autonomously. In a way, they're functioning like a passive investor who happens to always buy 100% of their target: they give companies the flexibility to raise prices, because the company knows that churn isn't an existential risk the way it would be if they didn't have a backup source of capital. That's a direct wealth transfer from customers to the company, but it's also a risk transfer in the other direction: Constellation basically built a diversified portfolio that keeps its acquisition targets alive for longer, and thus encourages them to operate as if they'll be justifying the same kinds of price hikes a decade from now. End customers might be happier with an alternative arrangement, but in that sense those same end customers were free-riding on the fragmentation of the software business.

Oddly, some of the big winners in strategic deals look a lot like the big losers: opportunistic M&A can go incredibly well (John Malone basically let SiriusXM get within days of running out of cash before finally lobbing an offer that the company had no choice but to accept). Time Warner felt like it was covering a short position in digital, and while it was right that magazines were going to die and movies would eventually be disrupted, it was a decade early on that trend really starting to bite, and also came close to top-ticking the market. But the YouTube deal looked like the same thing! Google had its own video product, and could use search for distribution, but YouTube had a more stripped-down uploading process, played through Flash (which most people already had installed), had social features on the site, and managed to get product-market fit as a way to share copyrighted clips as embedded videos on other fast-growing sites (where Google could drive even more traffic). It might be the most synergistic deal in history since what YouTube really needed was infrastructure and a legal team and what Google needed was to stop losing share in video.

One way strategic deals can work very well is if the acquisition target is basically a perfect answer to the acquirer asking the question "what should we have done differently X years ago?" There are more nebulous synergies that can work, like the assumption that company A's sales team can bundle B's products and vice-versa, but these are incredibly dependent on getting the sales team's incentives right. They work well when there are natural complements (both Google and Microsoft built their office suites partly through M&A, and in Google's case it was even easier because once it was clear that Google was going after the office market, that was basically a signal to build a browser-based version of anything Microsoft had to offer. This had mixed results (Zenter was pitched as "the Gmail of slides" and got acquired to be part of Google Slides, but Kiko, an online calendar, couldn't get to critical mass before Gmail launched its own calendar tool, though the founders turned out okay—they went on to do Twitch). 

On the cost side, there are some somewhat boring synergies that basically take the form of a short-term investment in a control premium plus integration costs, followed by a higher long-term return from replacing two CFOs, CMOs, etc. with just one. But these also carry a cost in terms of morale: the deals work if the actual redundancy was that the structure of the industry didn't call for two companies, not that the two companies each had people with the same title and responsibility. Managing twice the balance sheet is not quite twice the work, or company size would be capped by CFO talent, but it's still more work. And if a company repeatedly does this merge-and-purge maneuver, they need to be very good at assessing talent, and good at balancing the information advantage they have with their existing team against the social capital that existing team members will use to hold on to their job if the person at the acquired company is clearly better at it.

As a general rule in corporate America, bigger really is better—bigger companies can pay more and still produce higher profit margins, despite all the bureaucracy and inefficiency necessary to scale. But getting bigger is expensive, and buying scale suffers from adverse selection. Meanwhile, complementarity and cost synergies are very hard to assess from the outside: a striking pattern is that many of the best strategic acquisitions were still managed mostly independently for years after the acquisition, and even when deeper integration happened years after the fact it led to conflicts. You can buy assets, brands, and talent, but you never know what problems you bought along with them until after the deal is closed and the funds are wired.

Disclosure: Long META, Long GOOGL

The Diff has looked at many acquisition-driven business models and turnarounds, like:

• We looked at Transdigm, an acquisition-driven conglomerate that pushes prices as far as they can ($).

• Plaid ($) would have been a great example of a strategic merger, one way or another, if it had gone through.

• Figma's also a good what-if ($).

• This older piece on bubbles talks about the conglomerate boom of the 60s and the PE boom of the 80s as bets against synergy.

• Reddit didn't quite make sense as an independent company early on, but made much less sense as a subsidiary of a magazine publisher ($).

Join the discussion!

1  At its peak, ad spending on US linear TV was >$70 billion/ year.

2  Airlines tend to be the least value creating portion from a ROIC perspective in the aviation value chain with fuel production businesses and freight forwards capturing the majority of the outsized return in the industry over time. Even within carriers, the tendency is for value creation to happen through either loyalty programs at mature carriers, or through small ones growing fast when their younger fleet and workforce has a low marginal cost and then stalling once maintenance and seniority-based pay start cutting into their margins.

3  As it turns out, you can hide a discount carrier inside of a full-service one through basic economy fares, at which point you have the cross-subsidy of carrying passengers at every price point, and can use that subsidy to have connecting flights that compete on price with direct flights, and have much higher density of scheduled flights.