Asset allocation is integral to the success of any investment portfolio, and it is among the most important decisions an investor will make. In a year that has been full of surprises so far, we offer you a behind-the-scenes glimpse of the issues and themes at the forefront of the meetings of the NEPC Asset Allocation Team.
Participants:
- Jennifer Appel, CFA, Principal, Senior Investment Director
- Robert Goldthorpe, ASA, Senior Investment Director
- Phillip Nelson, CFA, Partner, Director of Asset Allocation
- James Reichert, CFA, Partner, Senior Director of Portfolio Strategy
- Aparajita Bubna, Managing Editor, facilitated this roundtable
What are the biggest surprises of 2026 so far, and rank them in terms of their impact?
Jennifer: Before ranking surprises, I think it’s useful to start with what the market consensus was coming into the year and assess whether those views are still intact or have broken down.
Two consensus views stand out. The first was that U.S. interest rates would move lower over the course of the year. That view has clearly been challenged. Expectations for rate cuts have been pushed out as inflation concerns have resurfaced and the labor market has remained resilient.
The second was a positive outlook for U.S. growth. I think that view largely still holds, although the rationale has shifted somewhat. Earlier in the year, optimism was driven in part by expectations that fiscal policy—the One Big Beautiful Bill—would provide a meaningful tailwind. More recently, however, higher energy prices and the reemergence of some inflation pressures have offset some of its stimulative effects.
With that framework in mind, the biggest surprise to me is interest rates. The market entered the year expecting a significantly different path for monetary policy than what has ultimately materialized. That said, markets have historically not been especially accurate at forecasting the trajectory of Fed funds rates, so perhaps this shouldn’t be as surprising.
Robert: Starting with the least surprising, I would go with AI-related spending. It’s hard to characterize that as a major surprise given how clearly it was telegraphed by the hyperscalers’ capital expenditure plans last year. The surprise within that theme is the performance of some of the secondary, less obvious beneficiaries. A number of hardware suppliers have seen outsized gains amid high demand for the underlying infrastructure supporting AI.
As for the biggest surprise, it has to be the geopolitical conflict in the Middle East. By definition, these events are difficult to predict, and the timing, scope, and market implications are invariably largely unanticipated.
Jennifer: A geopolitical conflict or event in any year is probably going to rank at the top of the surprises list.
James: I had geopolitics as my number one surprise too, but the other big surprise for me is consumer resilience. Coming into the year, it wasn’t at all clear that consumer spending would remain as strong as it has. Consumer sentiment was relatively weak as a result of the government shutdown, and there was uncertainty around spending.
While the One Big Beautiful Bill provided some support, it was far from certain that consumers would continue to keep spending. Yet, that’s what happened. Despite ongoing inflation pressures and, more recently, higher gas prices, consumer spending has remained remarkably resilient. The composition of that spending may be evolving—there appears to be stronger spending among more affluent consumers—but the fact remains that consumers are still opening their wallets.
As long as spending remains healthy, companies generally find ways to preserve profit margins and support earnings growth. That’s why consumer resilience stands out as one of the biggest surprises to me. It simply wasn’t the outcome that many expected at the start of the year.
On the other end of the spectrum, AI spending and index concentration are among the least surprising. Both trends were already firmly in place coming into the year.
Jennifer: I’m not entirely surprised by consumers’ resilience because it has been supported by a labor market that has remained stronger than many expected. One of the potential surprises I had in mind coming into the year was the possibility that the labor market would pick up steam again.
In the second half of last year, there were concerns around a softening labor market and the potential for jobs displaced by AI. Instead, the labor market has remained remarkably resilient, and if people are employed, they will still spend. In that regard, consumer resilience may be less of a surprise than the labor market holding up far better than expected.
Phillip: It’s interesting because if you had told me 10 years ago that the 10-year Treasury yield would be above 4.5%, oil prices would hit over $100 per barrel, and equities would be generally higher, I would view that combination as highly unlikely.
That said, coming into this year, many recognized that some sort of geopolitical surprise was likely. The specifics are always impossible to predict…would we have put Iran or Venezuela on the Bingo card? Probably not. But the idea that a geopolitical shock to markets could emerge wasn’t surprising. To me, the details were unexpected, but not the overall event.
What’s perhaps more notable is how markets have responded. We’ve seen repeated examples of geopolitical events initially spooking investors, only for markets to recover relatively quickly. Markets certainly reacted to developments earlier in the year, but the initial response was short-lived, and markets reversed course, and continued to solidify gains. It feels as though markets have become increasingly conditioned to shrug off geopolitical disruptions.
If I had to rank the biggest surprise, I might actually put AI at the top of the list—not because elevated AI spending or capital expenditures were unexpected, but because of how central AI has become to the investment landscape and the U.S. economy. The extent to which economic growth, corporate spending, and investor sentiment are tied to AI is most surprising to me. AI appears to be driving everything.
What is the biggest danger to the U.S. economy?
James: Interestingly, I identify what Phill just said as the biggest risk looking ahead. My concern is the extent of spending, investment and market sentiment that has become tied to AI. To be fair, AI is clearly creating value and has significant real-world applications, but the challenge is determining its most valuable uses, and whether capital is being directed towards them.
Jennifer: Exactly the same!
James: There will be losers. That’s a challenge for investors and a danger to the economy. The AI boom is fueling infrastructure growth, creates construction jobs, and bolstering local economies. A slowdown poses a big risk to markets and our economy.
Jennifer: Speaking of AI spending, if that tailwind starts to fade, we are going to start talking about recessions again. The volume of AI spending is probably one of the reasons there hasn’t been a recession in the last few years.
James: I think timing matters here. Much of this spending is already committed and there are long-term contracts in place and large-scale projects are already in development; after all, data centers aren’t built overnight. It may not be imminent, but the real risk may be if there is an AI bubble and the investment cycle starts to fade in one, two, or three years from now—and what that means for the U.S. economy.
Robert: I don’t know what will stop the AI investment cycle, but that has to be the biggest surprise if it occurs in the near term.
Phillip: But would we do anything different? Would we choose not to put trillions of dollars to work? It feels like the U.S. economy is collectively making a bet on AI—we are still not fully all in, but the economy is setting itself up for that.
I’m not sure that’s necessarily a bad thing. I find myself trying to compare the AI investment cycle to historical parallels, and the closest analogy might be the railroad expansion in the mid-1800s, where every policy, every dollar, every build-out was focused on expansion.
Robert: Would you say we’re all in on AI? It seems like all of this investment is being driven by the belief that AI is the future—and if companies don’t invest now, they risk being left behind.
As an economy and society, are we all in on AI? And is that a risk? What happens if AI ultimately isn’t what we expect it to be, or doesn’t deliver, or we need multiples of the investments we are seeing today to make it viable? These are the questions I keep coming back to. This feels dangerous to me because no one is going against the grain.
Phillip: At the same time, this AI investment cycle still feels nascent and early. There are pockets of tightness in areas like memory chips, but I think we are still early in the process of building out the infrastructure that’s been announced.
Think about all of the data centers that are expected to be built out over the next two to three years. Right now, there is a coordinated push from governments, regulators, corporations, and communities to ensure these projects move forward. The bigger risk to me is that the capital that is being pledged doesn’t get put to work.
If we’re talking about $3 trillion to $4 trillion of AI-related investment over the next three to four years relative to the size of the GDP of the U.S. which is roughly $30 trillion, I’m not sure it’s all that different from a massive infrastructure investment where we rebuild bridges and roads.
James: There are potentially significant productivity benefits on the other side as we build out the AI infrastructure, and technology continues to advance. We’re already seeing newer models become more energy-efficient and capable. It’s positive if people are able to advance scientific discoveries over the longer term; the faster those gains can be realized, the better.
Phillip: I think there have been many valid questions over the past decade about the United States’ ability to deploy capital domestically. Whether it’s infrastructure, housing, or other large-scale investments, most people would agree that it hasn’t been the easiest environment. But in this case, capital is flowing and investment is happening. To some degree, this is how the economy should work. This is what a dynamic economy looks like. To me, the biggest risk still comes back to inflation and not proactively managing it
Jennifer: One item on this list that we haven’t discussed yet is the possibility of a less independent Federal Reserve. That’s not our base case, but it is a risk with significant market implications, and not just from an economic standpoint; if investors lose confidence in the Fed’s independence and its commitment to checking inflation, you could see Treasury yields move higher. This would represent a massive structural shift that markets are not pricing in today.
James: That would pose significant danger, but I think the probability of it is low at this juncture. There remains widespread support for the continued independence of the Fed from both the new chair and all the governors that sit there today.
Jennifer: I don’t know if it’s because we are exhausted of it, but no one has mentioned tariffs so far.
James: Tariffs have been coming down too, and that’s important. Tariff rates hit a peak in the fall and since then, it’s a tailwind.
Robert: Like the spike in oil prices, markets seem to look past tariffs. With the tariff announcement, initially, there was a sudden sell-off, and then it became yesterday’s news.
Robert: Events like tariffs, oil prices, and other geopolitical conflicts could shock markets a few decades back, and unravel the global economy in certain ways. But now, is the global economy so diversified and resilient that events like these are seemingly isolated, and don’t have the broader impact they used to?
If you had a crystal ball, what’s the one thing you would like to know?
James: I want to know when fusion energy will become commercially viable. I think that it has the potential for data centers to lower their costs, consume less energy, and it could change the way we think about power in general.
Robert: I’d go with future of AI. I want to know the winners and losers. Some companies today could be overvalued, but potentially you could see trillions of dollars of added market value in the future.
Jennifer: To take the really short-term time horizon of the future of AI, I want to know if the earnings growth guidance from the Street and companies are real for the next couple of years. If they are, you’d want to buy the stocks, right?
Robert: You mention the point of there being winners and losers; all these companies can’t be successful in this space.
James: The losers will be those companies that are shedding customers. And then over time, those companies will tweak their system to make it more efficient and effective to improve their own margins. So, as long as each of these companies have paying customers that drive some cash flow, they’re all going to win because the service that they’re providing is going to be a little different and unique.
The idea that all the models need to do everything is how we think about AI now, but in the future, it could move to more specialized models. There will be losers, but it’ll be the companies who poorly executed their business strategy.
Phillip: Winners and losers in any space are normal. Personally, I think AI is broader than the dotcom buildout from more than 25 years ago. I would love to know about the change in productivity of the labor force over the next five years due to AI. How much more efficient do we become? To me, that is the telling point of how big AI can be in terms of changing capital markets and the economy.
What is your most important recommendation for investors?
James: This may sound boring but I’m going to say diversify. Diversification matters. With high yields around the globe there’s an opportunity that when this warmer inflationary environment reverses, the market will start discounting a new rate path.
Jennifer: Not being afraid of equity risk is important; be mindful of strategic targets, but don’t shy away just because returns have been so strong on paper. That’s the kind of bias that people have.
Phillip: This is a somewhat conservative answer, but I would say really making sure portfolios aren’t underweight the largest corporate names around the world because this is where the majority of earnings growth is coming from right now.
What should clients not be doing right now?
James: Two things: don’t reach and don’t run.
It’s easy to be fearful of valuations, but earnings appear to be solid for now. There might even be tailwinds if we see rates back off if inflation starts to slow down. At the same time, being diversified is important. To Phill’s point in the earlier question, about making sure portfolios aren’t underweight the largest corporate names around the world, you have to ensure you have that exposure to the big drivers of global growth. However, don’t put all your eggs into that basket.
Robert: What about private markets? I read headlines around credit and equity risks and write downs. Should an investor not be re-upping commitments?
James: I think investors should be re-upping commitments. Private markets are all about manager selection and finding those gems who do it better than everybody else, and when you find them you keep doing it till you find something better. You should trust in the process and look at investments that have good teams behind them, a sound strategy, and the right vehicle with the right fees.
Jennifer: Diversification by vintage year is important.
Phillip: One of the things we really warn against is a naive view of diversification, that is, just spreading your bets around for the sake of spreading your bets around.
We see this, to some degree, in how people allocate equities around the world because they need to be in countries and regions to balance their exposures. But there are differences in quality, and we should be tilting towards the best quality, best visibility for earnings growth, the most dynamic regions, and the highest potential for productivity.
James: Sounds like a growth tilt.
Phillip: If it quacks like a duck…
Jennifer: Sounds like a U.S. tilt.
James: I would push back on that. I think it’s a growth tilt because there are plenty of non-U.S. investments that are contributing to this productivity engine.
Jennifer: One other thing investors should not do is try to add public real assets exposure tactically if they don’t have it already.
James: To add to that, we believe liquid real assets can be part of your strategic asset allocation. However, now is not the time to start that blindly.
Phillip: In line with that, be purposeful with how you allocate capital and make sure there’s a reason behind it.
We believe diversifying portfolios with different asset classes are important, but what is your top performing asset class for the rest of 2026? Has it changed from your prediction from a year ago?
Jennifer: I will stick to my answer from 2025 of diversified public real assets. I think that even with a resolution around the Strait of Hormuz that commodity markets, specifically, will take time to normalize. You could also see more expenditure on infrastructure as a result of this event. So, I feel comfortable with my answer.
James: I said long bezzle and I’ll stick to that.
Robert: I would pivot towards commodities from the S&P 490 (that’s the S&P 500 excluding the top 10 names).
Phillip: I will stick with U.S. mid-cap stocks and add growth-oriented stocks.
Related Insights
