Our Insights

Michelle Weaver: Welcome to Thoughts on the Market. I'm Michelle Weaver, Morgan Stanley's U.S. Thematic and Equity Strategist.

David Arcaro: And I'm Dave Arcaro, U.S. Power, Utilities, and Clean Tech Analyst.

Michelle Weaver: Today, a hot topic. Are data centers’ raising your electricity bills?

It's Tuesday, December 23rd at 10am in New York.

Most of us have probably noticed our electricity bills have been creeping up. And it's putting pressure on U.S. consumers, especially with higher prices and paychecks not keeping pace. More and more people are pointing to data centers as the reason behind these rising costs, but the story isn't that simple.

Regional differences, shifting policies and local utility responses are all at play here. Dave, there's no doubt that data centers are becoming a much bigger part of the story when it comes to U.S. electricity demand. For listeners who might not follow these numbers every day, could you break down how data centers' share of overall electricity use is expected to grow over the next 10 years? And what does that mean for the grid and for the average consumer?

David Arcaro: Definitely they're becoming much bigger, much more important and more impactful across the industry in a big way. Data centers were 6 percent of total electricity consumption in the U.S. last year. We're actually forecasting that to triple to 18 percent by 2030, and then hit 20 percent in the early 2030s. So very strong growth, and increasing proportion of the overall utility, electricity use.

In aggregate, this is reflecting about 150 gigawatts of new data centers by 2030. Just a very large amount. And this is going to cause a major strain on the electric grid and is going to require substantial build out and upgrading of the transmission system along with construction of new power generation – like gas plants and large-scale renewables, wind, solar, and battery storage across the entire U.S.

And generally, when we see utilities investing in additional infrastructure, they need to get that cost recovered. We would typically expect that to lead to higher electric rates for consumers. That's the overall pressure that we're facing right now on the system, from all these data centers coming in.

We've got these substantial infrastructure needs. That means utilities will need to charge higher prices to consumers to cover the cost of those investments.

Michelle Weaver: What are the main challenges utilities companies face in meeting this rising demand from data centers?

David Arcaro: There are a number of challenges. If I were to pick a few of the biggest ones that I see, I think managing affordability is one of the biggest challenges the industry faces right now, because this overall data center growth is absolutely a shock to their business, and it needs to be managed carefully given the political and regulatory challenges that can arise when customer bills are getting are escalating faster than expected. The utility industry faces scrutiny and constant attention from a political and regulatory standpoint, so it's a balance that has to be very carefully managed. There are also reliability challenges that are important.

Utilities have to keep the lights on, you know, that's priority number one. The demand for electricity is growing much faster than the supply of new generation that we're seeing; new power plants just aren't being built fast enough. New transmission assets are not being built, as quickly as the data centers are coming on. So, in many areas we're seeing that leads to essentially less of a buffer, and more risk of outages during periods of extreme weather.

Michelle Weaver: And you mentioned, companies are thinking about how can they insulate consumers. Can you take us through some of the specifics of what these utility companies are doing? And what regulators are doing to respond, to protect existing customers from rate increases driven by data centers?

David Arcaro: Definitely. The industry is getting creative and trying to be proactive in addressing this issue. Many utilities, we're seeing them isolate data centers and charge them higher electric rates, specifically for those data center customers to try to cover all of the grid costs that are attributable to the data center's needs.

A couple examples. In Indiana, we're seeing that there's a utility there who's building new power plants, specifically for a very large data center that's coming into the state and they're ring fencing it. They're only charging the data center itself for those costs of the power plants. In Georgia, a utility there is charging a higher rate for the data centers that are coming in to the Atlanta area – such that it actually more than covers the costs and compensates other consumers in the form of bill credits or even bill reductions as those data centers come on.

Similarly, then, in Pennsylvania, there's a utility that has excess transmission infrastructure than the state’s [infrastructure]. They're better able to absorb data center activity. They're able to lower customer bills as the data centers come on, as they spread their costs over a larger customer base in that case. So, this isn't universal though. There are some areas around the country where there are costs related to data center growth that get socialized across all consumers.

One approach I also wanted to mention that we're seeing data centers pursue more and more actively is to power themselves. Essentially bring their own power, and they're using gas turbines, engines, and fuel cells that they're deploying right on site. This is actually in many cases faster than connecting to the grid, but it also avoids any consumer impact. Companies like Solaris Energy and Bloom Energy are two providers of that type of solution. And we're also seeing at a broader industry level. Another approach is the idea of data centers being flexible or turning off and not consuming power from the grid at certain times when the grid is facing stress, in an extreme weather scenario in the winter or summer. And that idea is gaining traction as well. So, we think the industry is looking for approaches that could ease the pressure on the system and on reliability, manage the affordability issues while continuing to enable and build data centers.

Michelle Weaver: You mentioned what a few different states are doing on this front. But data centers are not evenly distributed through states or evenly distributed across regions. Are there regional differences in how data center growth is impacting electricity prices?

David Arcaro: There are a couple of key differences that we're seeing around the country. Some areas just aren't getting that many data centers, you know, so I'd point out the northeast – in New England, in New York, we're just not seeing that much data center growth. So, it's less of an issue, the impact of data center power demand impacting customer bills in those areas. And then in some regions around the country, the utility structure is important to be aware of. There are some regions where the price of electricity fluctuates based on the supply and demand of power, rather than being directly set and controlled by a regulator. In those markets, data centers can actually more directly impact the price of electricity and there just isn't an easy way in that case to ring fence them and protect consumers from the impact of price increases.

So that's where we think unique challenges can arise. And over time, we would expect to see the most meaningful rate impacts to consumers in those areas specifically. And examples would be New Jersey, Maryland, Illinois, Pennsylvania, Ohio. Those are a couple of the states where we're seeing those more volatile and directly impacted prices.

So, as we look at utilities, we think the state exposure is going to be more and more important. And so, a few companies like NextEra, Sempra and AEP are a few utilities that are in states that have less affordability concerns and less direct exposure to rate impacts from data centers. And then several power companies like Vistra and Talen have more of their power plants that are in states that have excess infrastructure; and as a result, potentially less affordability concerns.

So, clearly the energy sector is facing real challenges and changes. So, Michelle, how are rising electricity bills actually affecting U.S. households?

Michelle Weaver: It's putting even more pressure on a consumer that's already being stretched thin by multiple years of inflation and elevated price levels, and electricity is a really different type of good. It's very different from gasoline or other consumer goods or staples – in that it's an essential good. You need to have it. And it's a network service that households are structurally locked into. Unlike gas where you could adjust your trip frequency or take a different type of transport, there really aren't good substitutes for electricity.

And so this dynamic weighs on consumers. They have to continue paying these bills, and it weighs particularly heavily on lower income consumers where utility bills make up a much larger portion of their household budget.

So, it crowds out some of that other potential spending.

David Arcaro: That makes a lot of sense. It's an important expense to consider in terms of the impact on consumers. And, you know, as a result, are consumers blaming data center electricity demand for this rise that we're seeing in bills or are they pushing back?

Michelle Weaver: Yeah. Data center development is quickly becoming a NIMBY or “not in my backyard” issue with communities pushing back and even getting projects canceled. Companies really need to find ways to address local concerns about environmental and water related externalities. And message that they're able to insulate consumers, or do something to mitigate these potentially higher electricity bills.

A recent poll of around 2200 voters found that just over half of respondents attribute overall electricity price increases to AI data centers, at least somewhat. While around another third, consider them very responsible. And these responses are consistent across all regions and across political affiliations. And I think this consistency across regions is really interesting. As we're talking about before, data centers are not impacting bills in every region. But consumers are still blaming them and still attributing bill increases there.

It’s clear that both the energy sector and U.S. consumers are navigating a complex landscape with data center growth at the center of the conversation. As policy responses evolve and the U.S. midterm elections approach, this issue is only going to gain more attention. And we'll be sure to bring you the latest. Dave, thanks for taking the time to talk.

David Arcaro: Great speaking with you, Michelle.

Michelle Weaver: And thanks for listening. If you enjoy Thoughts on the Market, please leave us a review wherever you listen and share the podcast with a friend or colleague today.

Welcome to Thoughts on the Market. I’m Serena Tang, Morgan Stanley’s Chief Cross-Asset Strategist.

Today – does the 60/40 portfolio still make sense, and what can investors expect from long-term market returns?

It’s Monday, December 22nd at 10am in New York.

Global equities have rallied by more than 35 percent from lows made in April. And U.S. high grade fixed income has seen the last 12 months’ returns reach 5 percent, above the averages over the last 10 years. This raises important questions about future returns and how investors might want to adapt their portfolios.

Now, our work shows that long-run expected returns for equities are lower than in previous decades, while fixed income – think government bonds and corporate bonds – still offers relatively elevated returns, thanks to higher yields.

Let’s put some numbers to it. Over the next decade, we project global equities to deliver an annualized return of nearly 7 percent, with the S&P 500 just behind at 6.8 percent. European and Japanese equities stand out, potentially returning about 8 percent. Emerging markets, however, lag at just about 4 percent. On the bond side, we think U.S. Treasuries with a 10-year maturity will return nearly 5 percent per year, German Bunds nearly 4 [percent], and Japanese government bonds nearly 2 [percent]. They may sound low, but it’s all above their long-run averages.

But here’s where it gets interesting. The extra return you get for taking on risk – what we call the risk premium – has compressed across the board. In the U.S., the equity risk premium is just 2 percent. And for emerging markets, it’s actually negative at around -1 percent. In very plain terms, investors aren’t being paid as much for taking on risk as they used to be.

Now, why is this the case? It’s because valuations are rich, especially in the U.S. But we also need to put these valuations in context. Yes, the S&P 500’s cyclically adjusted price-to-earnings ratio is near the highest level since the dotcom bubble. But the quality of the S&P 500 has improved dramatically over the past few decades. Companies are more profitable, and free cash flow -- money left after expenses -- is almost three times higher than it was in 2000. So, while valuations are rich, there’s some justification for it.

The lower risk premiums for stocks and credits, regardless of whether we think they are justified or not, has very interesting read across for investors’ multi-asset portfolios. The efficient frontier – meaning the best possible return for any given level of portfolio risk – has shifted. It’s now flatter and lower than in previous years. So, it means taking on more risk in a portfolio right now won’t necessarily boost returns as much as before.

Now, let’s turn our attention to the classic 60/40 portfolio – the mix of 60 percent stocks and 40 percent bonds that’s been a staple strategy for generations. After a tough 2022, this strategy has bounced back, delivering above-average returns for three years in a row. Looking ahead, though, we expect only around 6 percent annual returns for a 60/40 portfolio over the next decade versus around 9 percent average return historically. Importantly though, advances in AI could keep stocks and bonds moving more in sync than they used to be. If that happens, investors might benefit from increasing their equity allocation beyond the traditional 60/40 split.

Either way, it’s important to realize that the optimal mix of stocks and bonds is not static and should be revisited as market dynamics evolve.

In a world where risk assets feel expensive and the old rules don’t quite fit, it’s essential to understand how risk, return, and correlation work together. This will help you navigate the next decade. The 60/40 portfolio isn’t dead – and optimal multi-asset allocation weights are evolving. And so should you.

Thanks for listening. If you enjoy the show, please leave us a review wherever you listen and share Thoughts on the Market with a friend or colleague today.