“The greatest victory is that which requires no battle.”― Sun Tzu, The Art of War
The semiconductor market cap for publicly traded chip design companies is currently at ~$3T and growing, but that is a small measure of the total industry impact. Apple alone is a ~2.3T company, commanding impressive margins on its devices, and yet the devices are useless without the chips that go into them. If we were to tabulate the value of the technology stack across industries where chips are used, and the productivity gains enabled by the chips, even a conservative estimate of total value is staggering.
And the chips cannot be made without the depth and breadth of automation and productivity provided by electronic design automation (EDA) companies. With these kinds of economics, EDA market cap hovering at $190B seems counter-intuitive.
To be fair, EDA companies have improved their game in the past five years or so. Going from a combined enterprise value of ~$23B in 2017 for Synopsys, Cadence and Mentor (now Siemens EDA) to about $135B today for Synopsys and Cadence alone shows that the companies are managing to capture some of their due worth.
It is difficult to discern for outsiders whether this increase is just a lock step increase as the chip industry grows or the EDA industry has adopted more scalable strategies that allow it to capture more of the value into its own margins. The truth is somewhere in the middle, with much left to be done for EDA companies to combine forces and reposition the industry back to its glory days of innovation and growth.
Business cycles are funny. Companies strive to build a business so it can have a strong exit value, either through IPO or acquisition, and build options along the way so the company can control the narrative of when an exit path is exercised. “Maximizing Shareholder Value” is a mantra chanted across the chain. However, these glorious exit paths, once exercised, also set in motion counter-productive management decisions for many companies. Executives trade in long-term strategic decisions for short-term earnings gains (particularly in cases of publicly traded stocks) or chase rabbit holes with overpriced acquisitions without understanding the true potential of the underlying technology or the full synergies it can realize.
This is how EDA shot itself in the foot in this oligopolistic industry where 90% of the market share is dominated by the Big 4 (Synopsys Inc., Cadence Design Systems, Siemens EDA and Ansys).
Since the early 2000s, the big, publicly traded EDA enterprises pursued maximum shareholder value by optimizing for the next earnings call and created the DNA of enterprise-wide deals that allow big chip companies to lock in deeply discounted pricing for bundled tools, where protecting pricelist for individual tools no longer matter. It would seem that this is better for the consumer, but it isn’t when one looks at the history of how EDA grew.
The heydays of EDA are characterized by bigger companies commanding the sales channels but making room for smaller companies with bright ideas to demonstrate new product and market opportunities. In a tacit, industry-wide cooperation, smaller companies had an open field to define new market spaces and expand the scope of EDA. They eventually capitalized on their investment through lucrative exits.
These days, the tacit cooperation is among the Big 4 to have a similar basket of goods in their portfolios and limit noisy competition from possible new market segments, especially if these new solutions contradict their existing flows. It allows the four to improve margins and their P/E ratios through operational efficiencies but not with breakthrough technologies. The barriers to entry into the industry have been made so high, branding EDA as a mature and aging industry, that it is simply not attractive for young entrepreneurs to try and make their mark in this space.
This is a disservice to the consumers as well as to the EDA industry itself. We are still early in the potential of how far technology can go and how much complexity chips can manage. By limiting the pace of innovation in EDA, we are also limiting the addressable market size for the industry. With the big companies lacking agility in developing out-of-the-box solution areas, coupled with the anti-competitive approach of keeping smaller companies out of marquee clients, it is the users paying the price. It is crippling the potential of semiconductor design schedule improvements and suppressing a greater risk-taking behavior in new generations of chip design. Except for a handful few, the golden era of chip design houses willing to bet on new, disruptive and enabling technologies has been nearly wiped out in the name of a “mature” industry.
Geopolitical pressures have created new impetus. With greater push to regain thought leadership and sustainable competitive advantages within U.S. companies in semiconductor ecosystem, we have a chance to go back to the drawing board and step away from counter-productive competitive strategies. The Big 4 are the leaders in the industry and good leaders are not threatened by good ideas. By going back to the coopetition culture during the peak of innovation in EDA, they can influence a growth culture in the industry.
The Big 4 can invite and support participation from smaller companies in defining and developing new solutions, use internal R&D dollars more efficiently, refrain from anti-competitive pricing deals with customers that include bundling in sub-standard products unable to fully solve user problems, and allow smaller companies a meaningful economic path to success. This will put EDA back on a roadmap of growth opportunities being priced into stock values and the Big 4 can reduce reliance on shaving and/or controlling expenses to boost margins.
The users can also resist a short-term cost advantage from these bundled portfolios and focus on purchasing the best of breed solutions for each of their complex categories. They often underestimate the total cost of ownership in terms of engineering time lost in pursuing solutions that simply cannot scale. As such, they can indirectly force their vendors to compete on differentiation and feature advantage as opposed to the best pricing deals. If the users remain invariant in demanding excellence in products and service from their vendors, they naturally force the development of the best solutions, which in turn enables the users to produce their best in the shortest possible time.
EDA and the semiconductor industry in general are neither fully mature nor do the margins have to be razor thin. If all companies and stakeholders take an end-user centric, altruistic approach and collaboratively focus on longer-term industry gains, work to recreate the entrepreneurial spirit of innovation in chips and EDA, there is every possibility that the U.S. sustainably maintains significant leadership in the global industry.