Why low interest rates are allowing big companies to reshape the semiconductor industry, and how this could backfire.
The mergers and acquisition activity that has reshaped the semiconductor industry over the past couple of years resembles a frenzy of acquisition activity that has occurred multiple times in previous boom years. But this round comes with a twist. It’s being driven by historically low interest rates, which means it probably won’t stop until interest rates rise and the cost of borrowing capital for those acquisitions makes them much less attractive.
In fact, you would have to go back almost to the time the transistor was first invented (1947 at Bell Labs) to find comparable interest rates.
There are some other interesting anomalies in this round. While the big are indeed getting bigger, they aren’t necessarily gaining more market share. As Wally Rhines, Mentor Graphics’ chairman and CEO, pointed out in his keynote speech at the User2User Group meeting this week, the market share of the top 50 semiconductor companies actually has dropped over the past 10 years. In fact, more of the total semiconductor industry revenue is being spread out across more companies, which points toward deconsolidation rather than consolidation. But he also noted that economics allows for much higher-profile acquisitions. So to turn a $1 billion acquisition accretive only requires $30 million in profit at 3% interest for a five-year loan.
At some point, this will change. Interest rates will rise again. Normally interest rates are tied to inflation, which is reflected in the consumer price index. The U.S. Bureau of Labor Statistics defines the CPI as “the prices paid by urban consumers for a representative basket of goods and services.” That includes housing, fuels and utilities, services, and energy. But with oil prices falling, and Saudi Arabia now looking at the world after oil, it’s clear that the basis for the consumer price index needs some adjustments.
That all takes time, however. Bureaucracies move slowly. The U.S. Federal Reserve has stated publicly that it plans to keep interest rates low to foster continued economic growth, with only modest increases in the foreseeable future. At the same time, a combination of government and private investment from China will continue pouring into the semiconductor industry, to the tune of about $120 billion. Put all of those factors together, and it appears likely that acquisition activity will continue for much longer than the usual cycles would allow.
Not all of these acquisitions or investments will pay off, of course. In fact, the odds are higher that acquisitions are bad than good. But given investors’ focus on short-term profits, no one will really think twice about it. By the time things ultimately turn sour, the CEOs and CFOs who made those deals most likely will be gone. The market will have shifted yet again. But that may be 5 or 10 years from now.
Those companies that have made good purchases will see their revenues rise enough to offset those acquisitions. Companies will be able to ride out this wave of low interest rates, but at some point they will be left holding a pile of debt. If there are enough bad purchases, those ultimately will send stock prices tumbling across the entire sector and make borrowing even harder. (Banks always charge more when the risk goes up, even if it means paying for their own risky ventures.)
But there’s an interesting twist at the end of this cycle that never existed before. All of the past acquisition frenzies ended much more abruptly, due to outside economic pressures. This one will likely last long enough to do apples-to-apples comparisons of deals and measure who got it right and who didn’t, applying the same cost basis for borrowing capital. There are no business cycles to hide behind and nothing else to blame for what goes right or wrong. And for investors clamoring for transparency, it will look as if someone has just cleaned a dirty window.