How should I be positioned? with Torsten Slok (Apollo) and Jason Draho (UBS CIO)
The desk's thesis emphasizes a cautious outlook on U.S. economic growth in the context of evolving monetary policy and ongoing global uncertainties. Per the full note source, the commentary from Torsten Slock and Jason Draho suggests that the anticipated acceleration in U.S. growth may be impacted by persistent data noise, fiscal adjustments, and the introduction of AI into various sectors. This nuanced view prompts a recommendation to manage risk as investors navigate an environment characterized by both opportunity and volatility. The current consensus target for USD pairs indicates a central expectation around 1.075, implying traders should remain vigilant for shifts in economic data that could affect market positioning.
What the desk is arguing
The desk posits that while market participants have been operating under optimistic assumptions regarding U.S. economic growth, recent data suggest a more complex reality. Torsten Slock's insights during the UBS Market Moves podcast highlight that the noise in incoming economic data due to governmental shutdowns last fall may skew perceptions of growth trajectories.
Supporting evidence indicates that growth forecasts have become increasingly dependent on underlying fiscal policies and potential disruptions from AI technologies. Investors should monitor how these factors interplay with the Federal Reserve's future actions, as the implications could shape market dynamics significantly.
Where it sits in our coverage
Our current consensus target for the USD stands at 1.075, reflecting a moderate bullish stance. Notably, firms such as jpmorgan project a target of 1.10 for March 2026, while bofa is more conservative with a 1.04 target for the same period. This positions our desk's call slightly above the center of the wider consensus range, suggesting a leaning toward caution amid evolving economic conditions.
How other firms see it
The commentary aligns with views from firms like jpmorgan, which remains bullish, while bofa adopts a contrary stance with a more bearish outlook on USD positioning. Market participants should closely watch shifts in major currency pairs like EUR/USD, which are likely to mirror the Federal Reserve's monetary decisions and the overall macroeconomic climate. Detailed insights into AI's impact on economic recovery could also translate into volatility across various asset classes.
What the calendar says
Currently, there are no major economic events on the horizon which could act as catalysts for movement in the FX markets. Traders should remain proactive in assessing ongoing economic data releases and the sentiment in risk assets as indicators for future positioning.
01Cautious outlook on U.S. economic growth amid evolving monetary policy.
02Noise in economic data complicates growth forecasts significantly.
03Current consensus target for USD pairs is 1.075, suggesting vigilance in positioning.
04AI disruptions and fiscal policies could lead to increased market volatility.
Market implications
Traders should monitor U.S. economic data closely, as any surprises can lead to significant position adjustments. Watch for movements around the 1.075 mark in USD trade, which may act as a pivotal level in light of forthcoming fiscal and monetary policy changes.
Risks to this view
The call could be invalidated by unexpected economic data that significantly diverges from forecasts, or if the Federal Reserve signals a more hawkish turn that pressures the USD higher than anticipated.
ubs
Hi everyone, Dan Cassidy here. Welcome back to How Should I Be Positioned on the UBS Market Moves podcast channel. On this podcast, we like to exchange macro and market views with our industry colleagues and thought leaders along with thinking when it comes to asset allocation.
With that, joining us as we're making our way through the blizzard of 2026. We're not in studio today. We are doing this remote though.
We're very happy to have back on the podcast with us, Torsten Slock, Partner and Chief Economist at Apollo Global Management. We're joined today as always from the UBS Chief Investment Office as well by Head of Asset Allocation for the Americas, Jason Draho. So with that, Torsten, Jason, thank you as always for dedicating some time to be with our listeners, our clients here on How Should I Be Positioned.
I know there is a lot of topics we want to get through today. So, Jason, let me turn it over to you to lead the conversation with Torsten. All right.
Well, thank you, Dan. And thank you, Torsten, for joining us under suboptimal weather conditions. It is still, the blizzard is still kind of going on as we record this in New York City.
Definitely not hot, but I got the kind of the flip side of the mic. The first question I wanted to discuss was, you know, the consensus narrative starting last fall for the markets was that we're in this sort of running hot environment and that the idea that growth is going to accelerate, it's going to benefit from stimulus measures, the policies that were enacted last year, and then we would see growth surge in the first quarter, first half of 2026. You know, the data coming in looks sort of consistent with that, I think, but it's also really noisy because we're still dealing with kind of a backlog of data that was delayed because the government shut down last fall.
When you look at the state of the economy right now, do you think that that narrative from a macro perspective is justified? Is the data coming in that way? Do you still have confidence in that?
Or maybe we need to kind of recalibrate the thesis? No, I do think still that there's a number of tailwinds to the US economic outlook that continue to play out in the background. First of all, last year, we had the AI spending boom continuing into 2026.
And that spending boom, of course, continues to provide a tailwind, especially to the data center buildout, also to the associated energy buildout. Secondly, we also still have an industrial renaissance. There's a political will and strong opinion that we should home shore or onshore several things that used to be produced abroad, most importantly, semiconductors, pharmaceuticals, also defense.
And this industrial renaissance continues to also be a very important tailwind to growth at the moment. And lastly, we also have the One Big Beautiful Bill that took effect here on January the 1st. And that will mean that households that normally get around $3,000 in tax refunds in March and April, they will now this year get $4,000 in refunds.
Because remember, the One Big Beautiful Bill was done retroactively. And that means it starts instead on January the 1st, 2025. So the consequence, of course, of that is that we have all been paying too much in taxes in 2025.
In other words, withholdings have been too high. And as a result of that, again, households will get on average about $1,000 more. And given that there's about 130 million households in the U.S., that brings you into many billions of dollars of consumer spending that is likely going to come out over the next several months.
So the short answer, Jason, to your question is, there's a number of structural tailwinds. Importantly, tailwinds are not a function of what the Fed Funds Rate is, but tailwinds that mainly are the AI and data center build-out and the Industrial Renaissance and the One Big Beautiful Bill. And those are reasons why we should continue to be optimistic on the growth outlook as we go through 2026.
So you mentioned a few different things, and I want to come back to them in terms of kind of a question of, you know, if we're going to be wrong, like which of those factors might be most wrong, but are they kind of a third or fourth kind of wild card that's been thrown in just in the past few days regarding tariffs? You know, last time I think we did a podcast, it was in the summer of 2025. Already then, the markets felt like they're kind of moving on from tariffs to the main story.
And certainly, other than, you know, wondering when the Supreme Court is going to rule on tariffs, it's not been a factor really in market discourse so far this year. Given this reaction of the Supreme Court striking down the tariffs very quickly by Saturday morning, President Trump has signed an executive order implementing effective, I think it's midnight essentially, or just past midnight on the 24th of February, a 15% tariff across the board on all goods imported from the U.S., which will likely expire by July 24th. The question I guess I have on that topic specifically is, how much do you think, if at all, do you think as the state paid for the growth pieces that you laid out, because ultimately, I can still tell you that the effective tariff rate is going to be lower all sequel than it would have been as of just a week ago, but that's probably like a lot of uncertainty in existing trade deals for businesses that maybe thought they had some clarity, now they have to kind of rethink about it, you know, how this is going to play out.
So, how much of a drag do you think, if at all, is this latest development tariff story in altering your thoughts on, you know, the near or medium term for the U.S. economy? Yeah, this is important, Jason. If we go back and look at what was the level of the average effective tariff rate, meaning what was the average tax that was imposed on imports, what was it before the Supreme Court decision, and that was, according to the Yale Budget Lab, around 16%.
When the Supreme Court then struck down the AIVA tariff, then that moved down to 9%. And then Trump, of course, here has added now 15% back on top of that. That brings us back to the average effective tariff rate is now moving from 9 to 14.
So, that means that effectively, where we are today at an average effective tariff rate of 14%, is basically not too far away from the 16% that we were at before the Supreme Court decision. So, from that perspective, it is not much different. And in that sense, to your question, the impact on the economy is probably going to be relatively modest.
If anything, tariffs are going to drop a little bit from 16 to 14. And that, of course, should be positive for growth. And that should also mean that downward pressure modestly, at least, on inflation.
What really is, exactly as you're highlighting, the key challenge now is that we have now, unfortunately, a new uncertainty appearing on the horizon, namely that the Section 122 tariffs that were imposed by Trump, the 16%, they will only be allowed to stand for 150 days. And after that, they will need congressional approval. So, that means that in about six months' time, which means just before the midterm election, now Congress needs them to decide whether they're going to continue the 15% tariff on the rest of the world, or whether they're going to say, no, we're going to take this away, and then lower the effective tariff rate to 9%.
The issue also is that the IEVA tariffs allowed, of course, Trump to put tariffs on individual countries. Section 122 is for all countries that are importing, that the U.S. is importing from. So, that makes also a very big difference that it's no longer possible to the same degree from the President to impose tariffs on individual countries, whereas the Section 122 is all about 15% on all countries at the same time.
But from a macro account perspective, you're right, this creates a new uncertainty about what will happen in 150 days. What does this mean now that it's suddenly all countries? And it is correct what you're saying, and I agree with that, namely that given that elevated level of uncertainty, that is a very, very modest headwind, at least here and now.
But I still think that headwind is relatively small compared to the strong tailwinds that I mentioned, namely AI spending, the industrial renaissance, and the one big critical bill. So, on those factors, I think we're on the same page in that if the direction of travel for the U.S. economy was to accelerate from here, maybe this tariff news alters the slope a little bit, maybe slightly lower, maybe slightly higher, but it doesn't really change the general trajectory. So, there's other risks that growth could not materialize, and you alluded to the one big bill, the stimulus, and I guess the question everyone is trying to track right now is the refunds, tracking ahead to suggest what happened, or in general, could the stimulus from that bill not be as much as people think?
The industrial renaissance argument, you may like to get into construction, that's true, but will it actually matter so much in the near term? Other factors that could be in play is labor markets, a lot of questions regarding will job growth actually accelerate, especially given companies may be anxious about the economy, anxious about trying to deploy AI instead, and so would hiring stay low, in which case, you start to wonder, you start to weigh on consumer sentiment if they're worried about keeping or not even getting a new job. So, when you think about the different drivers that you laid out for why growth would accelerate, including the add-on to labor markets and the story getting better, what would be obvious factors, what's one that you think is the biggest risk of not materializing, or if it does materialize or doesn't materialize, would be the biggest drag on the growth pieces?
Yeah, this is very important, because let's take the two things, namely the labor market and AI, because you're right that the labor market has indeed shown that job growth has been slower. A lot of attention is being paid to the slowdown in job growth last year relative to 2024, but a very, very important reason for that is that there was a very significant decline in immigration. For the years 2022, 23, and 24, net immigration into the U.S., both legal and illegal, was about 3 million people came into the country on average for those three years.
Then, when Trump entered, of course, as president starting, of course, early 2025 last year, and this year, on average, net immigration will drop from around 3 million for those years to now around 500,000. And the consequence, of course, of this is that job growth should also be slowing down. The Brookings Institute has been calculating that in the years 2022, 23, and 24, non-farm payrolls on average was around 200,000, and now the break-even rate because of immigration declining so much, according to the Dallas Fed, is only around 30,000.
So, that's a very dramatic decline in job growth, simply because we have fewer immigrants coming into the country. And that is important, because also when people begin to talk about, well, what has AI done to the labor market, or what has the trade war done to the labor market, well, that then needs to also take into account that those effects need to be calculated, while at the same time recognizing that immigration has dropped as dramatically as it has. So, the first conclusion, in my view, is I still think the key reason why the labor market has shown slower job growth is by far because of the decline in the labor supply, and not so much because of the decline in labor demand.
If you look at the private sector, labor demand has been relatively steady, around 1% growth in jobs year-over-year throughout the last 18 months. So, the first conclusion is I think the labor market is better than its reputation, simply because it is mainly labor supply that has thrown things down. And on AI, I think still very early, as Jay Powell has been saying at several press conferences when he's been asked about this, he still says, and I agree with that, that there is no sign of AI today in the labor market.
There's no sign today of AI in inflation. There's no sign today of AI in productivity. Perhaps most importantly, if you look at earnings expectations to the magnificent seven over the last year, the magnificent seven earnings expectations have gone up a lot.
But if you then over the same period compare that with earnings expectations for the S&P 493, they have basically moved sideways. So, that's a different way of saying if AI really was such a revolutionary technology, you should also expect that to show up in higher profitability in the S&P 493. But that's simply not what the market is expecting.
So, I still think the jury is out in terms of what AI will do. And at this point, it's just clear, at least in the data that we look at, that there's just not signs yet of AI showing up in the data. Well, we'll come back to the CIO question later on, because it is not so much in the interim, rather than perhaps the hyperscalers, the amount of CapEx they're going to spend to be a driver for just investment this year and therefore GDP growth.
But structurally, there's just some important questions that I want to come back to. But speaking a little bit with the more cyclical outlook, we are only 10 plus minutes into this podcast. We've talked about growth, we've talked about the drivers, fiscal policy, tariff policy.
We have not yet talked about inflation. And the biggest concern when tariffs are first introduced is that it would be inflationary. And that's not necessarily, but it's evident in the data, but we have not seen a massive inflation surge.
And when you start to look at the components of inflation, it still looks like these inflation trends are at work. The shelter piece looks like it's going to continue to go lower. Good volatility should roll over in terms of inflation once the tariff shock fully works its way through.
But the risk, of course, is that we're a little too complacent on this. The right-to-hot narrative implies that we almost have overheating in the economy, which would mean typically inflation stays elevated. So I guess how confident are you right now that inflation will actually roll over by year-end?
It's trending towards 2% comfortably. It's not going to get stuck closer to 3%. Yeah, this is very important.
We just got also recently the latest data for PCE inflation, which is the measure of inflation that the Fed is watching most carefully. And it did, unfortunately, move up to 3% year-over-year. So if you look at a chart for headline PCE inflation, for core PCE inflation, even for super core PCE inflation, it comes to the very clear conclusion that those pictures tell you exactly the worry, namely that inflation is at the moment sticky at around 3%.
In fact, for super core, it's begun to move up back towards 4%. So I think this is the key reason why Jay Powell went out and recorded the video on a Sunday afternoon, because he is basically trying to tell us that they are worried and the FOMC, at least the majority of FOMC members are worried that they cannot cut interest rates when inflation is sticky at around 3%. And the last point on this is also, there is now an important paper that was written at the Peterson Institute by Adam Poston and Peter Orszag, which is saying that inflation by the end of the year may be as high as 4%.
So I still think that the unfortunate thing for the Fed is that it's going to be very difficult for them to cut interest rates this year when inflation continues to be as sticky at around 3% as it is at the moment. We'll pick up on the last point regarding the Fed. Market pricing is still for two cuts this year.
The Fed was not fully priced until June, which is when Kevin Warsh, the nominee, presumably will be in office, assuming he gets through the regular confirmation process. Again, the question was if that's going to happen or not, but let's take that at face value that that's going to ultimately work its way through, even if the timing is a little bit lower or slower. And so, a cut by June in his first meeting, another cut by year-end, which, you know, maybe the economic conditions were warranted, although, you know, this is where inflation would perhaps suggest it isn't.
If the labor market is holding up okay, you could argue that the Fed really doesn't need to be cutting rates. I think part of the rationale is with Warsh, that he's kind of made the case for cuts. One might think that in order to get the job, he had to say, yes, I want to cut.
He made the argument that AI is going to be disinflationary, that's the reason to cut, and so on and so forth. So, we can speculate on what exactly the Fed will do this year, but I guess my question is, what would a Warsh-led Fed look like? Like, what is the policy reaction function?
You know, we'll talk about who wanted to shrink the balance sheet. What do you think is sort of relative from a policy family that's different than just, you know, what the Fed is already doing? How could that shift maybe this year and more so, you know, next year?
Like, what do you think that you would look for? And also, the ways that might be kind of out of consensus for what the market is currently thinking? Agree.
I agree. This is an important issue, and the key question is, of course, whether Warsh 1.0 will be saying the same things when he is now Warsh 2.0. The key issue for him, of course, is that there are 12 voting members on the FOMC, and at the last FOMC meeting, 10 of these members said we should not cut interest rates.
Two members said that we should cut interest rates. Well, if you now swap one of the 10 with, meaning Jay Powell with Kevin Warsh, maybe there will be three people saying that they should cut interest rates, but there will still be nine people saying that they should not cut interest rates. So, in some sense, I think at least the starting point is that it's business as usual for him.
He will be having a hard time to convince those members who just voted to not cut interest rates that interest rates should be going down, especially with inflation having moved up and especially with growth likely accelerating over the coming months. So, that's why the challenge for Kevin Warsh will be that it's the 12 voting members that decide what should happen, and the chairman can have some views, but the other members can have their own views, and it remains to be seen whether the other members are going to change their views or not. But at least where I stand right now, I do think that when he joins on May the 15th, and as you said, the first meeting is in June, I still think that the challenge for him is that he needs to really get the committee together, and if he goes out himself and says, I think interest rates should have gone down, and the majority of the committee says, no, we didn't think so, then, of course, the press conference is going to be awkward.
So, that's why it will be a challenging time for him and for the committee, given that these tensions are underneath their views at the moment. Separate from race is the balance sheet. Warsh has been critical about the Fed going beyond its remit or it's over its key in terms of how big its balance sheet should be.
He seems to advocate a small balance sheet. At the same time, there's also, especially last week's discussions about a kind of deal or framework that the Fed has with the Treasury, and this kind of goes back to 1951, kind of in the post-Second World War era. Right after that, the Fed actually engaged, and I don't know the details of it, actually in conjunction with the Treasury, but it's going to have a cap of 2% on 10-year Treasuries.
Inflation surged, then it came down. It helped, ultimately, kind of to get the economy back on track as it sort of recovered from the Second World War. But the deal that was sort of struck in 1951 seemed to create sufficient independence for the Fed to manage its balance sheet sort of separate from the Treasury.
Given Warsh's view on the balance sheet, given sort of questions about could the Treasury under Besant do things that look like, well, the Fed's not going to do operations with, for example, we'll just change the supply and issuance of the leading market bills, for example. It's not been the case for the first year or so with Besant as the Treasury Secretary. But now that you've got a maybe greater meeting of the minds between Besant and Warsh, what do you think is in scope for change and what realistically might happen and what sort of implications could that have for either the conduct of monetary policy or financial markets overall?
Yeah, this is potentially very important because any agreement or any accord between the Fed and the Treasury, of course, will need to be studied very, very carefully because what is the new idea and what is the goal with such an accord? What is it that's not working in the relationship that the Fed and the Treasury have today? At the moment, of course, the Fed is operating completely independently of the Treasury.
But if the accord has the purpose of coordinating more, of course, there will begin to be risk of a steeper curve in rates markets because the result might be that investors could begin to worry about whether fiscal policy is beginning to drive more of the decisions at the Fed. And that's, of course, an situation where Fed independence will be debated very heavily in market, namely, what does it mean that the Fed is independent? What would such an accord mean if the Treasury has some handle on what the Fed should be doing?
What Kevin Warsh has been saying is that he wants to, indeed, as you say, shrink the balance sheet and get them back to their core mission, namely of simply just doing monetary policy. At the same time, if you do shrink the balance sheet today, you run the risk, of course, that you will have scarce reserves. And that runs the risk of creating all kinds of challenges in money markets like we saw in September of 2019.
So there's a number of things that make it difficult for Kevin Warsh to shrink the balance sheet the way that he's been talking about earlier. So that's also why I think he's more pragmatic. I think he's going to go in and say, well, the goal here is really to get the inflation to 2% and have full employment.
And with that can likely be done with a balance sheet that is at least closer to where it is today as a share of GDP without having to shrink it dramatically. Because that raises a whole host of other problems and challenges in financial markets, if he were to make the balance sheet a lot smaller. I want to pivot back to the AI discussion we were having earlier.
Any of you know, it's kind of Mark talks in the context of is it impacting labor market or productivity right now? Or are we looking back last year as an explanation for what's going on with either? Ultimately, AI, we know it's kind of a multi-year implication.
So looking less cyclical and more circular three, five, 10-year horizon. We know that AI will have an impact. It's probably going to be, you know, it should enhance productivity.
Otherwise, you know, it has very little value if it doesn't do that. But if you were to like lean in certain directions of like the impact being large or small or the impact on the labor market, you know, the argument that like ultimately like an aging technology, there'll be a disruption in the labor market, but it tends to create more jobs and it loses like, how are you thinking about more like multi-year perspective, like, you know, the impact of AI and where you feel more confident like this is likely to happen. This is maybe less likely to happen given some of the discussions that are taking place right now.
Yeah, this is obviously a very, very important risk to any asset allocation decision at the moment, because if AI succeeds, it will have consequences. If AI disappoints, it will have consequences. Let's just agree up to this point that AI has at least not yet, as Jay Powell also repeatedly says, shown up in the data for employment, productivity, and inflation.
And it's also not yet shown up in the aggregate for data for profitability and for earnings outside of the tech sector. So in that sense, where we stand right now, a lot of the discussion around what AI will do remains very speculative. But the way I think about it is two ways.
First of all, if we really think about what AI is, it is simply doing tasks better in our jobs. So this then boils down to the question, okay, so if I do in my job and you do in your job, say 10, 20 different tasks every day, if I can make one, two, three of these tasks more efficient by using AI, then of course I will be more productive. But the key issue here is that there are basically no jobs in the U.S. economy that only have one task.
Most jobs that have one task have already been automated a long, long time ago. So in that sense, it's difficult to see this traumatic increase in the unemployment rate that some people talk about, because there are no jobs in the U.S. that only have one task. Even jobs as a paralegal lawyer, someone who's very careful, looks at x-rays, that's not just one task, that's a lot of other dimensions than simply the AI part of their jobs.
So this whole idea that they could just be replaced by a computer, it just seems very far-fetched in my view. So that's why my view is that so far it has not shown much up in the actual data, and I'm also skeptical that it will show up in the significant way that some people have been suggesting. And the second element also of the AI discussion is the following.
Think about, for example, when a company invents a new product. That product could, for example, be, say, Oceantic. Well, those who have Oceantic and have produced it, when they produced it, had some years where they have some pricing power, some monopoly profits, where they could squeeze some extra profits out and generate a lot of revenue.
But what is the product today in AI? Well, the product is a large-language model. But the large-language model, there are many different large-language models.
There's Brock, there's Claude, there's ChatGPT, FlexGPT, Gemini. And the users, it's really difficult to see the difference between these large-language models. So for that reason, the price of the product today is likely going to go down to zero.
So the value is not the product itself as it was with Oceantic, but the value is instead how the product is used. In other words, how AI is disseminated through the economy. And that remains the key issue, namely, will it be disseminated?
Will it actually generate those significant gains in productivity and significant gains in income that people are assuming? And again, at this point, we're just not seeing that yet. Yes, certainly, in some parts of the economy and coding, in particular computer science, we have seen some improvements.
But in the aggregate, it still remains questionable that these significant effects that some people talk about are going to materialize. Well, I think the area that most people agree with is that it's going to enhance some efficiencies, like operational efficiencies, but bring costs down, as you mentioned. Ultimately, these models are somewhat interchangeable.
And it's not a matter of debate, but let's say they're all pretty similar for a lot of tasks. In competitions, it drives the cost more. That means our ability to go out and apply the services and the tools, the costs go lower.
So that's the context where, in the past few weeks, we've seen in the market, equity, but it's covered in fixed income. AI could just go and disrupt all sorts of business models because it can do things at so much lower cost. It doesn't actually expand your revenue.
On top of that, it doesn't much lower costs. So software, the sector was being hit, and then we kept it out. We saw insurance companies reacting, brokers, logistics companies.
So there's been this almost shoot for us to ask questions later between who are the winners of all this, who has the most, and who is the bulk of losers. One of the non-profit effects in the debt market is a lot of debt was issued to pay for investments, especially software companies, technology companies, in the public market, but even more so in the private market. We've seen some high-profile examples of where there's been some stress there.
How worried are you about some financial system stability, contagion, or just a drag on economic activity if this continues and suddenly access to capital gets tied up or there's defaults that are picking up? Is that a significant risk in your mind or still maybe a little bit overstated given the current fears in the marketplace regarding what AI is going to disrupt? I do think that the software sector is certainly threatened by AI, but I also think that the software in a similar way as we just talked about before, it is really AI is able to change individual tasks and do tasks better.
But the question is, if you and I have a small business and we have some software that they run the business, either say the production or our customer relationship management, are we going to stop subscribing to the software we use today and then sit down and code and program new software for our business? It seems also a little bit difficult to see this happening, both in the short and in the long run, because if we were to do that, then we need to hire a lot of programmers and a lot of AI managers who could then instead install the software on our own computers that were written by AI rather than subscribing to the software that we were used to subscribe to. So the bottom line is, I still think that it will make a difference.
But I also think that there are some sectors and certainly some parts of software that will be threatened. But I also think that this also becomes an issue of what types of software is it then? And that, of course, is the hard homework to look under the hood of individual sectors, of course, in particular software, and see which parts of the software sector is it that's going to get hurt.
And that, of course, continues to be a major challenge for investors. At the moment, the entire software sector, of course, is selling off, and that may be, of course, the right trade if you are a fast money trader. But if you then begin to look two, three years down the road, then it seems very unlikely that all software companies will be going out of business.
So that's just to say that the stock selection or credit selection continues to be absolutely key when one thinks about what the consequences are of the new technology that is AI. So we've covered AI, we've talked about the Fed, tariffs, growth outlook, and risks to the growth outlook. Are there things that we haven't touched on, whether it's factors, risks, geopolitics, the midterms, something that you think is important to cover that maybe investors in general are focused not enough on?
Do you think about what would matter for the economic and investing outlook for the rest of the year? Yep. So the way we look at things is that we are driving the ball down the fairway and we're going towards the flag and the hole.
And for most of 2024 and 2025, our view was that there was a 90% chance that we would stay within the guardrails of staying on the fairway, and only a 10% chance that we would land out in the rough. But given the challenges with geopolitics, given the challenges with debt levels, not only in the US, but also in Germany and in Japan, given the challenges, again, with now a trade war potentially coming back with elections around the world, then I think that the likelihood of landing the ball out in the rough and getting some tail risk has increased from 10% to 30%. So therefore, investors should begin to construct their portfolios in a way where, in our view, there should be more focus on non-AI exposure, because AI is already a significant part of the S&P 500.
And with the hyperscalers issuing more debt, then AI is also becoming a more significant part of the public ID market. And finally, AI is also becoming a bigger part of venture capital. It used to be that venture capital was biotech and pharma and fintech and growth.
But now, venture capital, two-thirds, is also AI. So if I think about the pie chart for asset allocation, normally, we would say, oh, I have debt, I have equity, those things are uncorrelated. But today, that's no longer the case.
Debt and equity are highly correlated, because equity is all AI in the S&P 500. And in public credit, it's also all AI. And finally, venture capital is also all AI.
So I think investors, the best thing to do at the moment is to try to find some exposure to things that are not AI, meaning to things that are not exposed to the situation, either where AI is unsuccessful, or paradoxically, also to the scenario where AI is widely successful. Well, that actually kind of segues into a closing comment I'll make, that one of our methods with certain equities is an idea of a great broadening. The market, the U.S. equity market specifically, has been fairly concentrated, either driven heavily by Mank7, or some iteration of the Mank7 could go back to more than 5, 10 years, in terms of like the FANGs cost.
But this sense of like, there's now, we've entered a period where there's just about a reason for other parts of the market, whether it's typical stocks in the U.S., or other markets around the world, to have this sort of game decided, not just for a couple of weeks, but for an extended period of time. So kind of broadening the first line, because ultimately, as you mentioned, I would agree that, you know, what would actually get us in the rough, you know, that 30% chance, you know, there's a lot of different factors, not one necessarily kind of jumps out to me, but, you know, the best way to sort of manage that is through kind of being diversified. So, Torsten, thanks for your respect, really appreciate your insights.
And I'm sure we'll be talking again fairly soon. Thanks for having me. I really appreciate it.
Okay. With that, Jason Torsten, always a rich and insightful conversation. Thank you again for joining us today on How Should I Be Positioned?
And to you, our listeners, thank you as well for your engagement from UBS Studios. I'm Dan Cassidy. Thank you for joining us.
Thank you for tuning in. Be sure to visit ubs.com slash studios to view the entire UBS Studios suite of podcast channels, along with our video offerings, such as UBS Trending. You can also follow us on Instagram for content highlights at UBS Trending.
UBS Studios is part of the UBS Chief Investment Office within UBS Global Wealth Management. Visit ubs.com slash CIO to view the latest research. UBS Chief Investment Office's investment views are prepared and published by the Global Wealth Management business of UBS AG or its affiliates.
The views and opinions expressed in this material by external guest speakers are those of the author, speaker, and are not those of UBS, its subsidiaries, or affiliates. Accordingly, UBS does not accept any liability over the content of this material or any claims, losses, or damages arising from the use or reliance of all or any part thereof. This material has no regard to the specific investment objectives, financial situation, or particular needs of any specific recipient and is published for informational purposes only.
For a full legal disclaimer applicable to the independent investment views produced by UBS, please visit our website at UBS.com forward slash CIO dash disclaimer.