Predictions: 2025
Predictions are tough ... especially about the future!
It’s that time of year again … New Year’s Eve and predictions for the future. It's time to reflect on the year gone by and look forward to the new year. As Yogi Berra said, “It’s tough to make predictions … especially about the future.” With that spirit in mind, here are my 2025 predictions.
Note: these are in order of decreasing likelihood of proving out to be correct. I will try to give a quantifiable “success” criterion for each prediction. That way, these can be scored at the end of next year, and I can fall flat on my face :)
Prediction: D.O.G.E. has their homework eaten by the Beltway
After President Trump was elected by a comfortable margin, we heard that Elon Musk and Vivek Ramaswamy were going to lead some sort of new governmental efficiency entity. It is going to be called the Department of Governmental Efficiency (cleverly nicknamed “D.O.G.E”.) The purpose of this new entity is to reduce government waste and spending. I certainly think this is a worthy idea, but I believe it will soon become evident that this whole initiative is a mirage. I hope I am wrong.
First of all, I use the word “entity” deliberately. These two billionaires are basically unpaid advisers to President-elect Trump. Of course, they are two very powerful men, especially Elon Musk. So, we do have to take them seriously. However, they will initially have no authority to reduce one dime of Federal spending. That will fall completely on the Congress. To give them that authority, Congress would have to pass a law authorizing them to cut spending. Or simply cut spending themselves, through regular order. Without that, they’re merely unpaid consultants, and we all know how effective consultants are at reducing spending.
The second reason I doubt that D.O.G.E. hunts is that the GOP hold on Congress, especially the House, is very tenuous. The current margin in the House is 219-215. That means only one House member on the elephant side can defect, if the House Dems vote as a block, to get a bill passed.
There are already problems. Thomas Massie (KY) has said that he won’t vote for Mike Johnson as speaker, so we don’t even know if there will be a speaker of the House for the first days of 2025. And without a speaker, no bills can even move to the floor. Another GOP House member, Rep. Spartz has announced that she won’t caucus with the GOP. She did say that she supports the D.O.G.E. so perhaps she would vote in favor of any legislation to cut spending. But two other GOP reps are stepping down to join the Trump administration: Elise Stefanik (NY) and Mike Waltz (FL.) It will take until the spring for special elections to fill these two vacancies. Once they leave, the effective House majority may be a mere single vote.
It is hard to see Republicans ramming through any legislation to cut government spending or fire Federal workers with this razor-thin majority. One wild card in the deck is the Democrats. Nancy Pelosi had hip replacement surgery in December, and may not be back for floor votes for a while. It may be wise for Mike Johnson to try and ram through any D.O.G.E. enablement bill very quickly – perhaps early in January while she is still recuperating and Waltz and Stefanik are still around. However, wisdom and Mike Johnson don’t generally go together in a sentence. This is the same guy who had his 1600+ page “CRomnibus” shot down by Musk and Trump, producing an early victory for D.O.G.E as the resulting bill was slimmed down to a mere 116 pages or so. And Johnson is the same guy who put a $60B aid bill for Ukraine and Israel on the House floor back in April. At the time of this writing, he appears to be hanging by a thread and could be ousted as House Speaker.
Finally, anything D.O.G.E. tries to do will no doubt be challenged in the courts. That alone will delay and frustrate Musk and Vivek.
Sidebar: Federal remote work policy is one area where D.O.G.E. might make inroads without having to go through Congress. It appears that the party may be ending for Federal workers and telework. If D.O.G.E. can get policy changed to eliminate remote work through executive action and bypass Congress thereby forcing federal workers back to the office, many may quit or retire rather than face soul-crushing commutes again. This could be a back-door way for D.O.G.E. to get rid of some Federal workers. However, without the big appropriations bills actually cutting FTE’s I don’t think these positions will really go away. Agencies would be free to re-hire new workers to replace the quitters, albeit with job descriptions no longer as attractive, as they’d be tied to offices and dreadful commutes.
Success criterion: No Federal agency gets eliminated, even secondary ones like the National Endowment for Democracy (NED) or the National Endowment for the Arts (NEA.) Federal spending for the next year’s budget (2026) will rise over 2025. The total number of FTE’s for the Federal Government will be roughly the same as in 2024.
Conclusion: D.O.G.E won’t hunt.
Prediction: The US Stock market takes a breather
TL;DR:
Definition of correct: S&P closes lower on Dec. 31, 2025, than on Jan 2nd, 2025.
Rationale – it’s overpriced, with too much concentration in a few darlings (NVDA, AVGO)
Bonus prediction: Bonds stink, too. There will be nowhere to hide … even BTC and Gold will tank. US treasuries will be the “winner” as you don’t lose money in them.
My thinking here is that the US stock market has been the strongest of all the major economies for a while, at least since the pandemic year of 2020. The US market is led by tech stocks, especially the so-called “Magnificent Seven” stocks like Nvidia, Microsoft, Google, and Meta. Despite higher interest rates, these growth stocks have thrived based on huge expectations of future revenue, in part due to a narrative that “AI” is going to massively change the future the way the World Wide Web did in the 90’s.
I think 2025 is the year when the narrative faces reality, especially around AI (see next prediction, below.) Many of these companies have absurd price-to-earnings or price-to-sales ratios. I don’t follow the stock market as closely as I used to, but some of the best Elliott Wave theorists I follow are pointing to extreme danger signs, as the market appears to be topping in a massive 5-wave pattern that spans decades. One analyst I follow even predicts a 13-20-year bear market is about to begin.😯
I am not quite that pessimistic, but I think the jig is up for Mr. Market in 2025. While stocks won’t outright crash, the “Mag Seven” group that bubble-vision babes love to drool over on CNBC will correct significantly, and because they make up so much of the overall market weight of indices like the Nasdaq and S&P, market returns will be nominally negative for 2025 vs. today’s closing prices.
Scoring criterion: This will be easy to score: If the S&P closes higher for 2025 vs. December 31, 2024, I’m wrong. If it is lower, I’m right.
Conclusion: Hang out in T-bills or a money-market fund if you don’t like to see big drawdowns in your trading accounts.
Disclaimer: This prediction is for amusement purposes only. It is not financial advice. I could easily be wrong, and stonks could rocket to new all-time highs in 2025. If you’re looking for actual trading advise, seek out a professional service.
The Artificial Intelligence bubble starts to pop.
This one is the big Kahuna that I hope most comes true, but I may be too early. Note that I said “starts” to pop. I believe the analogy might be to the Housing bubble, circa 2007. That was a full year before we had the full destruction of the economy, with Lehman Bros. failing and Hank Paulsen and Ben Bernanke working long nights and weekends on the next bailouts. We’re in the late innings, but the ballgame isn’t over. The full AI collapse is likely still several years away.
There is no doubt that Wall Street and Silicon Valley are spending billions to create an AI bubble. Bubbles are characterized by:
Excessive greed and/or a herd mentality.
A common refrain that “this time is different” or “we’re in a new paradigm”
Ridiculous statements become accepted as gospel. As in, “Home prices will never drop again” or “we’re in a permanently high plateau!”
The AI debate features all three. Stocks like Nvidia and Broadcom have seen insane rallies, and every time I tune in to CNBC I hear stories about AI taking over and resulting in a new paradigm for tech companies. Virtually every commercial on bubble vision has a sales droid screaming “powered by AI” or “our new AI platform” as a selling point. Even for companies that have nothing to do with the technology.
As I wrote on LinkedIn, in response to a poster claiming that AI was going to decimate knowledge-based work:
There is nothing really new, here. Artificial intelligence has been around as a concept since the 1950s. The current craze is based on computing power reaching the point where large language models are capable of impressive shows of pseudo-knowledge. But there is a catch -they can only perform using massive data sets that rely on human input. If humanity stopped being curious and questioning the conventional, scientific progress would be over. Science always requires someone who questions the conventional wisdom. If it didn't, AI would tell us that the earth is flat.
There will be disruption to some job roles, as always. Just as the Internet and the cloud changed the way IT does its job. But there will always be a need for human thinking. In the legal area, it is very unlikely that gatekeepers like the ABA will allow "AI" to replace judges, lawyers, and prosecutors.
So, what we have here is a herd mentality. Check one.
Second, we hear that this time is different. AI is going to not only eliminate legal, financial, IT, and customer service jobs, it is going to make all our lives easier. We’ll never have to worry about thinking again – AI will do it for us. This is rather dangerous, in addition to being delusional.
Certainly, there is value in large language models in being able to synthesize information and produce reasonably good first drafts for things like briefing a court decision or replacing search engines. I have some experience with this, being in law school. Legal tools like Lexus/Westlaw have already integrated some AI features, and can generate quick legal briefs. And these tools will get better. However, it is quite a leap to think that they could ever replace an actual lawyer. The heart of lawyering is the legal analysis phase, not the ability to list facts or memorize rules. Remember too that court decisions don’t always follow stare decisis. If we ever replaced judges with AI, we’d never have overturned Roe v. Wade, or allowed for gay marriage as in Oberfell. Law would become a “dead” field, like Latin or ancient Sanskrit.
And that leads to my main reason for skepticism on the AI FOMO-fest. AI is only as good as the data sets you feed it. That means it can only tell us about the past. Without skepticism, science never advances. And AI is incapable of skepticism. It’s inherently probabilistic. Thus, it will never produce anything novel, only pale imitations of past human creativity.
Conclusion: Yes, Virginia, there is an AI bubble.
Scoring:
Definition of correct: No widespread layoffs due to AI replacing jobs, and at least one AI-related public company goes bust, or a major scandal erupts with AI being found to produce fraudulent results and lawsuits filed against the owners of the company hiding behind the AI.
Note: As I write this, I found out that a Canadian-based SaaS accounting firm that possibly used AI to replace accountants has already failed. Whether they relied on AI or not is unclear at the moment, but it would not surprise me if they did. I’ll be doing some research on this and leave an update if it turns out to be significant. I expect to hear more stories like this in 2025.
Bench shuts down, leaving thousands of businesses without access to accounting and tax docs
Other markers:
Wall Street starts to warn on companies overspending on AI with no ROI or significant revenue boosts;
Talk turns to using AI to make existing employees more productive, not replacing them;
AI for cost-cutting survives as a thing; we get more stories about people unable to reach a human for customer service; the crapification of everything continues.
So, there you go, three predictions for 2025 that are score-able. Best wishes to everyone on a Happy New Year!
Bonus Geopolitical musings:
You’ll note that I refrained from making any predictions on:
The wars in Gaza, Ukraine
Whether Trump gets any of his agenda through (other than D.O.G.E. which I predict will fail) including ending the war in Ukraine within 24 hours of assuming office.
I am tempted but I believe that current events are simply too unpredictable to say with any certainty where things are headed in Ukraine or the Middle East. And I have no specific insight or edge here, vs. other pundits out there in the blogosphere. I’ll note that not a single pundit I follow or know of predicted the collapse of the Assad regime in 2024.
If I had to make a prediction, it would be that the war in Ukraine continues throughout the entirety of 2025. That’s a very sad prediction that I hope doesn’t come true. But I really don’t see Trump able to overcome the neo-cons in any meaningful way. If he tries to cut the money flow to Zelensky, I suspect the Senate GOP ghouls like Mitch McConnell will stop him. Even if he somehow gets Congress to clawback money from the DoD budget, the war mongers will also use the court system to impede him, much as they did with immigration in 2017-2020.
Again, best wishes in 2025 to anyone reading and have a happy New Year!