US Chamber of Commerce Hosts CEO Summit of the Americas in Los Angeles Anna Moneymaker
Alphabet Inc. 🇧🇷NASDAQ: GOOG) stock has been hit hard this year. Down 39% for the year, it underperformed the NASDAQ-100 benchmark.
Google vs spy vs NASDAQ (Searching Alpha Quant)
Google’s underperformance is a recent development. Earlier in the year, it actually avoided the worst of tech stocks’ slump, faring much better than the beaten names that dragged the index lower. If you look at the chart above, you will see that Google has been ahead of the NASDAQ for most of the year, most recently on November 25th. It has dropped to underperforming in the last month.
So what happened here?
Basically, there were two things:
Google’s first-quarter earnings release showed a large increase in headcount despite a slowdown in revenue growth. Overall results missed analysts’ estimates. Predictably, stock plummeted after the launch.
ChatGPT was recently launched and has become a huge phenomenon on Twitter, with many people commenting that it could replace Google Search for some use cases.
After these two events took place, Google stock dropped below the NASDAQ-100 on an annual basis.
ChatGPT’s risk to Google was covered extensively in Seeking Alpha. Some contributors feel that total disruption is imminent, others believe the risk is overstated. I myself wrote an article arguing that the risk was exaggerated, due to the fact that ChatGPT does not overlap with all areas of Google’s business.
I still believe what I wrote in my recent article on ChatGPT. Google Search makes up 57% of Google’s business, and ChatGPT has nothing to do with search use cases such as news because its training data ends in 2021. There is little overlap between ChatGPT and other Alphabet businesses , such as YouTube, Google Cloud and Gmail .
However, there is a real risk to Google shareholders, not coming from ChatGPT, but from Google’s own management:
Arrogance
Arrogance, described as “excessive pride or self-confidence,” is the cardinal sin of success. This makes the person think that his past success will continue into the future and that nothing can stop him from getting what he wants.
At the end of 2022, it’s hard not to believe that Google isn’t being affected by some degree of arrogance. The company’s management persisted with expensive hiring campaigns long after it became obvious they were impeding performance, lagged behind OpenAI in consumer-facing AI applications, and allowed its culture to become slow and bureaucratic (according to some insiders). None of these signals alone “prove” that Google’s management considers itself untouchable, but collectively, they balk.
Another factor that suggests Google’s leadership is getting too confident is the claim that search results are getting worse. Many programmers now say that Reddit is better than Google at retrieving information, and others say the same thing about ChatGPT. It would be strange for Google to lose its status as the world’s best way to find information, as it has the resources to stay at the top. Sadly, management doesn’t seem to know there’s a problem: when people ask Google why the search results are getting worse, they’re usually brushed off. For example, former Googler Marissa Mayer once said that it was the internet, not Google, that was getting worse.
There’s a strong argument to be made here that the people running Google feel their position is unassailable. This is a real risk that shareholders should be aware of. ChatGPT itself won’t kill Google, but enough ChatGPT-type outages could eventually erode the Google moat. To this day, the moat is quite wide, but nothing is destined to last forever. In this article, I’ll explore the possibility that Google’s moat could narrow due to management complacency, concluding that it’s a real long-term risk (though not a big deal in the short term). However, I remain bullish on stock in general because it has a large collection of products and services that are generally very profitable.
Hubris in business history
The idea that arrogance leads to failure is one of the oldest tropes in literature. In Greek tragedy, the arrogance of the protagonist usually led to self-destruction, or at least a major setback. The same pattern has been seen many times in business. in your essay “The Psychology of Human Misjudgment”, Charlie Munger describes a case where HP Inc. (NASDAQ:HPQ) decided to hire Carly Fiorina as CEO because management thought she appeared “articulate and dynamic.”
As management was used to success and believed in its own abilities, it felt confident in his positive opinion of Fiorina and hired her, although (according to Munger) a methodological recruitment process had ruled her out as a candidate. Ultimately, HP underperformed the S&P 500 during Fiorina’s tenure, leading to her being forced out.
There are many other examples of companies struggling due to arrogance. A classic example is Blackberry (BB) missing the transition to touchscreen smartphones after Litter (AAPL) launched the iPhone because its sales were already very good. Ultimately, Apple’s vision triumphed and BlackBerry’s stock plummeted.
Could Google have a “BlackBerry moment” on the horizon?
At the moment, it doesn’t look like any of Google’s products will be discontinued overnight. As I mentioned in my last article on ChatGPT, the most obvious competitive threat is only a partial threat. ChatGPT cannot compete with Google Search for news or product searches because its training data ends in 2021. Product searches are the most profitable part of Google’s search business as they allow Google to show people ads relevant to the products they are looking for. Therefore, the impact on ChatGPT revenue should be minimal in the short term.
However, if Google’s management is becoming complacent, a series of ChatGPT-type threats could erode the company’s moat. Imagine for a second that ChatGPT added new data to the ChatGPT training database so it could summarize the latest news and pull product reviews from Reddit. If that happened, ChatGPT would be competing with the entirety of Google search. There are technical challenges to training AI in the latest news – the subject is changing all the time, making validation a challenge. But we could see this kind of thing emerge in the coming years.
In the face of such possibilities, we have stories of excessive bureaucracy at Google, programmers saying that Reddit is better than Google at retrieving information, and Google executives dismissing claims that Search is deteriorating. It certainly seems like arrogance is at play here; if it is, it can erode value over time.
Why am I still optimistic?
Having explored the signs that Google’s management is getting overconfident, it’s time to explain why I’m still bullish on the stock. So far, my review makes Google seem like a minefield of risks, but overall, I think the stock is a really good buy. In the next few paragraphs, I’ll explain why I feel this way.
The first reason is evaluation. Thanks to this year’s downgrade, Google is now cheaper than any time in recent memory. At today’s prices, Google is trading at:
17.8 times earnings.
Four times more sales.
4.5 times book value.
12.4 times operating cash flow.
The first and last of these multiples are genuinely low, the middle two are still high, but lower than they were in the past. You can see some more valuation rates below, courtesy of Seeking Alpha Quant.
Google rating (Searching Alpha Quant)
Google is also very cheap on a discounted cash flow (“DCF”) model. I’ve done complete Google DCF models, with modeled financial statements and estimated growth, in previous articles. You can check out one of these models here. For now, I’ll just point out that, in a terminal value model, GOOG has an upside with no presumed growth. If you assume 0% growth and discount the company’s $4.76 in free cash flow per share by 5%, you get a price target of $95. Treasury’s current yield is 3.75% , so the chosen discount rate includes a risk premium. Typically, equity risk premiums tend to be greater than 1.25%, but since my growth assumption is incredibly conservative (0%), the 5% discount is justified. If Google never grows again, it should be worth at least $95.
In addition to your assessment, another reason I’m bullish on Google is that it’s still showing appreciable revenue growth. Google’s earnings have been declining in recent quarters, but it still has revenue strength. In the most recent quarter, Google’s revenue grew 6%, when other big tech companies like Metaplatforms (GOAL) and nvidia (NVDA) saw their growth turn negative. Obviously, the earnings growth achieved at this revenue level has not been strong. But increased revenue provides a path to future profit growth, assuming management eventually sees cost discipline. The arrogance I mentioned earlier may delay the realization that such discipline is needed, but we have indications that cost cutting could start as early as the first half of 2023. It would be great news if it materializes.
Recent Google Title Earnings Metrics (Alphabet Inc)
Conclusion
The bottom line with Google is that it’s a big company whose management needs to make some big decisions right now. Cost discipline is urgently needed in the short term, while longer-term investments in AI and the quality of search results will have to be made. It won’t be easy, but a little volatility in the stock market could be just the wake-up call management needs to start acting.