Every month, The Prudent Speculator produces a newsletter that includes a market summary, helpful charts and graphs, recent equity market news, economic outlook and specific stock investment strategies focused on value stock investing. This month, we celebrate the 700th edition of the newsletter, we look at historical market barometers and we offer our take on DeepSeek. Note that the entire list is available to our community of subscribers only.
Editor’s Note: DeepSeek, January Barometer and the Super Bowl Indicator
We are pleased to present the 700th edition of The Prudent Speculator and your Editor is thrilled to have been on board for more than 480 of those publications. It was 38 years ago that our founder Al Frank took a wet-behind-the-ears 21-year-old under his wing, teaching him very early on that the secret to success in stocks is not to get scared out of them.
With the equity markets enduring significant volatility to start February, given President Trump’s announcement of tariffs on Canada, Mexico and China, and the drama as those countries considered retaliations, we share wisdom Al penned on the occasion of our 70th edition dated November 1979.
I live in a quiet apartment on a quiet street in quiet Santa Monica. Life goes well and I have much to be thankful for. But when the stock market averages drop 10% or so, I find myself writing about “massacres” and stocks being “murdered” by panicking investors. Watching the market too much each day, rooting too hard for my portfolio (my “team” of stocks), reading and listening to too much jargon, I sometimes lose perspective and feel that I am in a battle with almost life and death outcomes. Will the Bull gore the Bear, or will the Bear maul the Bull? Perhaps we should see the market more as populated by chickens, what with all the clucking and squawking at every ruffled feather.
Interest rates and inflation will abate, and the USA will survive Iran’s traumas. Along the way there could be some terrific dislocations in the economy, for instance, when crude oil averages twice its current price. As in times past, adjustments will be made, like the horseless carriage driving buggy whip makers into manufacturing other profitable products. They are having a sale in Wall Street, with recently increased discounts. Now is a good time to buy selected common stocks for long-term capital gains in a widely diversified portfolio.
Believe it or not, the Dow Jones Industrial Average was then quoted at 818.94, with the popular gauge enjoying a total return including dividends of more than 20,000% over the ensuing 45+ years, or 12.46% per annum.
True, the major market averages have not fallen by 10% today, though corrections of that magnitude have occurred every 11 months on average, and they are not far off all-time highs after a superb January. Still, we think Al’s admonitions now are as valuable as ever, given that there is plenty of hand-wringing about possible trade wars, not to mention the unpredictable nature of the new Administration and its influence of actions on the geopolitical stage. We could also add rich valuations for many stocks, DeepSeek wreaking havoc on A.I.-related companies (see our Graphic Detail) and questions about future Federal Reserve interest rate decisions to the mix.
Indeed, the uncertainty that always permeates the equity markets seems a little more extreme these days, so we turn back to Al for comments written in the first edition of the newsletter back in March, 1977: After several years of helter-skelter buying and selling, I learned that successful speculating is more a matter of character than mathematics, analysis, or luck. Obviously the latter are required, but the great gains and losses seem to occur in consequence to individual psychology.
Anything can happen as we go forward and we know that trips to the downside are part of the game, but we remain very comfortable with the inexpensive valuations and generous capital returns for the majority of our stocks. In fact, TPS Portfolio sports a forward P/E ratio of 14.5 and a 2.5% dividend yield vs. respective figures of 25.7 and 1.2% for the S&P 500. Yes, tariffs could impair corporate profits and lead to an economic slowdown, but stocks are dramatically higher today than when the Trump 45 Trade War began in March 2018, reinforcing Al Frank’s conclusion, “If you are willing to get rich slowly, the market is the place to be to make or maintain a fortune.”
“Nature does not hurry, yet everything is accomplished.” — Lao Tzu
Graphic Detail: As January Goes
With stocks experiencing positive returns in the first month of 2025, the January Barometer would argue for stronger-than-usual performance over the balance of the year. Of course, the past is merely a guide and not the gospel as stocks rallied sharply last year, in line with Presidential 4th Years and the Year of the Dragon in the Chinese Zodiac, even as the Kansas City Chiefs Super Bowl LVIII victory might have suggested otherwise. Of course, no matter the calendar study, almost every review of the historical data favors Value-oriented strategies.
Graphic Detail: Artificial Intelligence
In a major “aha moment” for artificial intelligence (A.I.) technology, DeepSeek, a Chinese A.I. laboratory, presented a new A.I. reasoning model called R1 that is fast, reasonably accurate and uses novel logic shortcuts that purportedly result in a 90% to 95% reduction in computing power consumption. R1 uses reinforcement learning to generate chains of complex thought autonomously and self-verifies the information, resulting in a scalable, ultra-efficient model like none before it.
The DeepSeek team actually released a research paper on DeepSeek-V3 December 27, explaining that it is “a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token…We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours for its full training.”
It did not get much attention outside the A.I. community until DeepSeek-R1 was posted to GitHub on January 20, a repository for open-source code. Instead of walling off the team’s contributions to A.I. and privatizing the model, R1 was made available to everyone (provided they had the hardware to run the code). Software engineer and early-stage investor Marc Andreessen called the release “a profound gift to the world,” but A.I.-related stocks tumbled on fears that A.I. investments were worth less if models could operate with a 90%+ reduction in cost.
We argue the knee-jerk dumping of many A.I.-related stocks ignored the fact that DeepSeek’s claims excluded the potentially hefty costs tied to “prior research and ablation experiments on architectures, algorithms or data.” That does not mean that DeepSeek is not a major step forward, but the billions of dollars headed toward A.I. investments, especially those related to hardware, are not likely to be undone. Projections for A.I. capital expenditures were never accomplishable given current constraints. There are plenty of instances where chip demand far exceeds available supply, power generation capacity is insufficient to fuel data centers (a single data center can consume as much energy as 50,000+ homes) and human interaction with models is inefficient.
On Threads, Chief A.I. Scientist at Meta Platforms (META) Yann LeCun wrote about infrastructure investments, “Much of those billions are going into infrastructure for *inference*, not training…Once you put video understanding, reasoning, large-scale memory, and other capabilities in AI systems, inference costs are going to increase. The only real question is whether users will be willing to pay enough (directly or not) to justify the [expenses].” He was quick to add, “The market reactions to DeepSeek are woefully unjustified.”
A.I., while groundbreaking, is still a costly and resource-intensive endeavor, particularly in its early stages of development and deployment. Models require significant computational power, leading to high infrastructure costs that can represent material expenses for even the largest tech firms. Companies face mounting costs for training and maintaining models. Indeed, a single AI query can demand nearly 10 times the energy than that of a normal Google search. However, as history has shown, technological advancements that reduce costs often trigger Jevons Paradox: as efficiencies increase, so does demand.
Not unlike what we have seen in industries like electricity and air travel—this dynamic underscores the need for A.I. to overcome its current cost challenges, as lower expenses will drive broader adoption and deeper integration across industries. In its current form, A.I. can be deployed profitably by just a few companies. It’s expensive, labor-intensive to train and requires massive budgets to get useful answers. By making R1 open source, anyone can start tinkering with A.I. provided they have a relatively reasonable budget and plenty of persistence. What a change! And we don’t think for a minute that the engineers at the “big firms” are ignoring R1. Surely, they’re reading every word in the documentation, likely applying it to their proprietary models in the same way that Alphabet (GOOG) used open-source Chromium as a foundation for Chrome and Microsoft (MSFT) for Edge. So, open-source initiatives, like the DeepSeek project, exemplify how collaborative efforts can push the boundaries of cost reduction while democratizing access to powerful AI tools.
Developments in the past few weeks have reaffirmed our position that A.I. is here to stay. Perhaps we need to update our latest Market Outlook, in which we wrote “monetization of the technology with consistency will come further down the road sooner than we think.”
Recommended Stock List
In this space, we list all of the stocks we own across our multi-cap-value managed account strategies and in our four newsletter portfolios. See the last page for pertinent information on our flagship TPS strategy, which has been in existence since the launch of The Prudent Speculator in March 1977.
Readers are likely aware that TPS has long been monitored by The Hulbert Financial Digest (“Hulbert”). As industry watchdog Mark Hulbert states, “Hulbert was founded in 1980 with the goal of tracking investment advisory newsletters. Ever since it has been the premiere source of objective and independent performance ratings for the industry.” For info on the newsletters tracked by Hulbert, visit: http://hulbertratings.com/since-inception/.
Keeping in mind that all stocks are rated as “Buys” until such time as we issue an official Sales Alert, we believe that all of the companies in the tables on these pages trade for significant discounts to our determination of long-term fair value and/or offer favorable risk/reward profiles. Note that, while we always seek substantial capital gains, we require lower appreciation potential for stocks that we deem to have more stable earnings streams, more diversified businesses and stronger balance sheets. The natural corollary is that riskier companies must offer far greater upside to warrant a recommendation. Further, as total return is how performance is ultimately judged, we explicitly factor dividend payments into our analytical work.
While we always like to state that we like all of our children equally, meaning that we would be fine in purchasing any of the 100+ stocks, we remind subscribers that we very much advocate broad portfolio diversification with TPS Portfolio holding more than eighty of these companies. Of course, we respect that some folks may prefer a more concentrated portfolio, however our minimum comfort level in terms of number of overall holdings in a broadly diversified portfolio is at least thirty!
TPS rankings and performance are derived from hypothetical transactions “entered” by Hulbert based on recommendations provided within TPS, and according to Hulbert’s own procedures, irrespective of specific prices shown within TPS, where applicable. Such performance does not reflect the actual experience of any TPS subscriber. Hulbert applies a hypothetical commission to all “transactions” based on an average rate that is charged by the largest discount brokers in the U.S., and which rate is solely determined by Hulbert. Hulbert’s performance calculations do not incorporate the effects of taxes, fees, or other expenses. TPS pays an annual fee to be monitored and ranked by Hulbert. With respect to “since inception” performance, Hulbert has compared TPS to 19 other newsletters across 62 strategies (as of the date of this publication). Past performance is not an indication of future results. For additional information about Hulbert’s methodology, visit: http://hulbertratings.com/methodology/.
Portfolio Builder
Each month in this column, we highlight 10 stocks with which readers might populate their portfolios: Corning (CLW), General Motors (GM), United Parcel Service (UPS)
Kovitz Investment Group Partners, LLC (“Kovitz”) is an investment adviser registered with the Securities and Exchange Commission. This report should only be considered as a tool in any investment decision and should not be used by itself to make investment decisions. Opinions expressed are only our current opinions or our opinions on the posting date. Any graphs, data, or information in this publication are considered reliably sourced, but no representation is made that it is accurate or complete and should not be relied upon as such. This information is subject to change without notice at any time, based on market and other conditions. Past performance is not indicative of future results, which may vary.