There are two ways companies capture new markets, either through in-house innovation or through acquisitions. It is a very conscious and thought-through decision that managements need to make whether to build or buy. Historically, there have been very successful and positive outcomes for both approaches. Consider the following examples:

• Apple innovated and built the iPhone.

• Google bought Android for $50 million, capturing a large market share.

When done right, acquisitions can act as a big boost to the company as has been the case with Google. However, one needs to remember that statistically acquisitions have been more value-destructive than value-accretive.

One should be wary of companies that are serial acquirers. Take the example of Valeant. Since 2009 Valeant went into an acquisition spree all the while cutting its R&D spending on drug discovery. It was a classic case of buy versus build. However, to keep its streak of heady growth, Valeant went on to buy companies with questionable operations and accounting. One may be aware of the whole, very public, saga of Valeant, Allergan, and Bill Ackman.

Bill Ackman (Pershing Square), a renowned hedge fund manager, had taken a large stake in Allergan while Valeant was pursuing it. Allergan repeatedly shunned and rebuffed Valeant’s overtures even though Ackman was pressuring them to accept Valeant’s buyout offer. Long story short, the acquisition never happened, Valeant went on to lose 99% of its market value, and class action lawsuits followed. Valeant and Ackman were fined.

The red flags were there for all to see, however, investors cheered every acquisition Valeant made and kept driving up its stock. If it weren’t for the whole Allergan saga, Valeant would have had to face reality eventually as at some point they would have run out of acquisition targets.

Acquisition Accounting Jugglary

One common theme one would have noticed is that upon completion of an acquisition, the acquired company adds an outsized chunk of earnings, more than it was previously earning, to the acquirer. One should be wary of this as we shall see how this is a one-off boost to earnings and certainly not sustainable. So what happens?

Right before the acquisition closes, the target’s revenue growth stalls, expenses balloon, and some even report losses. All of this is explained away by the management as, “restructuring operations to build synergies”. In reality, although not always, the finance guys from the acquirer walk in and have a conversation, something in the following lines.

So, what would this achieve? Revenue will be recorded after the acquisition is completed. Also, by paying the bills early, the expenses would be recorded before the acquisition is completed. Thereby, boosting the revenue, earnings, and cash from operations of the combined company, more than analysts and investors would have expected. The CEO can bombastically announce on the earnings call the great job they are doing in integrating and how the purported synergies are paying off.

These are the simplest of simplest acquisition tricks that can be or are played. Also, watch out for revenue recognition practices being changed. For instance, the percentage-of-completion-of-work-based revenue recognition could be accelerated. Instead of charging a one-time sale fee, the management may decide to pivot to a subscription-based model (exactly what Broadcom has done). The useful life of PPE (Property Plant Equipment) can be extended bringing down depreciation and amortization costs.

The games played with the useful life of PPE are pretty common even outside of acquisitions. Nvidia pulled this trick when it made a subtle change in 2023.

A $0.05 per share boost in the context of $1.20 EPS is not that small. If not for the accounting methodology change, the EPS would have been $1.15. That is a 4.5% boost. Also, as most of you may have seen, a company’s stock crashes if it misses even by a penny. Hence, such changes are not insignificant.

A Recent Example - Broadcom

On November 22, 2023, Broadcom completed the acquisition of VMWare in a cash-and-stock transaction. The preliminary total cost of the acquisition was approximately $86.3 billion of which $30.4 billion was paid in cash and the rest was funded by 54.4 million shares. Quite a huge deal.

Now, let’s look at the highlights from the quarterly income statements of VMWare before the acquisition closed.

Two things stand out:

1. See how the revenue has fallen off in the last two reported quarters?

2. Look at the reported months. April became May, and July became August. Quite simply could have been in preparation for the acquisition or could be something nefarious like “Calender Games”. We will get to it a little later.

Remember the discussion we had earlier about pushing out revenue collection and pulling forward expenses? Let’s look at the last two reported quarters before the acquisition closed.

Historically, VMWare has managed its accounts payable well and the accounts payable has always remained on the low side. What stands out is the approximately $1.2 billion jump in accounts receivable and the approximately $1.3 billion drop in cash from operations caused mostly by, you guessed it, change in accounts receivables. Also, note the discrepancy and swing as compared to the net income. Interesting, right?

So, logically, the $3.73 billion in accounts receivable is what Broadcom would inherit, and given the large accounts receivables VMWare has historically carried, the receivables should show up in Broadcom’s books after the acquisition is completed. Did it? Judge for yourself.

So, Broadcom’s receivables increased by $1.8 billion while it most probably inherited $3.73 billion of receivables from VMWare and that is not counting the receivables that might have accrued till November 22, 2023. An astute investor would have seen through all of this and expected a huge contribution to Broadcom’s revenues and earnings from VMWare or used to mask the weakness in VMWare when it reported the results for the period ending right after the acquisition was completed. Also, the investor would have known that all of this was a one-off event and that reality would show up in the next quarter.

Here are the excerpts from the press releases after Broadcom reported earnings. First from Q1 2024 (March 7, 2024).

One can infer two things here. Either Broadcom bought a dud or used quite a lot of VMWare’s revenues to juice up its standalone numbers. Which is it? I don’t know for sure, however, it doesn’t look so good.

Now let’s do a full quarter of the consolidated entity, i.e. Q2 2024.

Again, let’s reconcile the numbers as we did for Q1 FY2024.

If this doesn’t scare the daylights out of investors, not sure, what else will. There is a clear and marked deterioration happening in the VMWare business even after all the carried-over receivables. Were there warning signs? Here is what Broadcom’s CEO wrote in a blog post on March 14, 2024.

Here is an interesting story (excerpts) reported by The Register on May 22, 2024.

Finally, coming to the change in calendar days or weeks as was highlighted earlier. Calendar games were made famous and quite nefariously used by CA (Computer Associates) in the early 2000s. The scheme was as simple as increasing the number of days in the last month of the quarter from 30 days to 35 days. What did it achieve? It helped CA book the extra revenue from the five days added. It was obviously illegal and the CEO ended up in jail.

Now, Broadcom changed its calendar reporting date and thereby extended its Q1 2024. The intent here would be very important or it could be as innocuous as a change required due to the VMWare acquisition. However, it did one thing. It allowed Broadcom to include 10.5 weeks of VMWare revenues instead of 10 weeks and it potentially included some of its own revenue it also recorded during this added half week.

The intention here is not to rain on Broadcom’s AI-driven hype and parade. Consider this an educational article highlighting some of the possibilities, mostly legal even if borderline, of what companies can do with accounting when acquiring companies.

One parting thought about Broadcom. Its revenue source is highly concentrated. All right, it can happen in the industry they are in. Now consider this statement from its 10Q for Q2 FY2024 (pdf page 33 - emphasis added).

Based on how Broadcom defines its revenue recognition policies, everything seems to be in order. However, the above language used to define a distributor as a customer should raise some eyebrows.

Until next time. Have fun.