Recent developments in AI chip efficiency suggest that future data centers may not need the same level of intense cooling infrastructure. This shift has immediate impacts on the market, especially for companies that make cooling equipment. While these businesses were once seen as major winners in the AI boom, fast technological changes might make their equipment less necessary in the future.

This highlights a fundamental change in infrastructure investing. In the past, this sector focused on stable assets like toll roads or airports. Those businesses do not change much over time. However, because modern infrastructure is now tied to technology, it carries a higher risk. A tech upgrade every few years can entirely change the investment equation for what was once considered a reliable and long term business.

Infrastructure investors historically have tended to like relatively stable returns. You invest in it because you expect this to be a pretty reliable business. But because you have infrastructure that is now tied to tech, it seems like there is a pretty big risk that every year or every two years there is going to be some huge tech upgrade that just changes the equation entirely.

There is also a growing focus on the physical world. In the previous decade, many investors made money through financial engineering. Today, success often requires a more hands on approach to physical assets. This trend is accelerated by massive amounts of public money and global events that require rebuilding energy and oil infrastructure.

Cohen and Steers is a long only asset manager. The firm focuses on real assets and alternative income strategies. They manage mutual funds and active ETFs for various investors. Tyler works specifically with global listed infrastructure and natural resource equity strategies. These strategies cover a wide range of sectors that keep the modern world running.

These are long only strategies investing across in infrastructure. It is what we call the cute subsectors. That is communications, things like cell towers, data centers and satellites, utilities, electric, gas, water, renewables, and transportation.

The scope of infrastructure includes communications like cell towers and data centers. It also covers utilities such as water and renewables. Transportation is another major category. This includes toll roads, airports, and freight rails. Energy midstream assets like pipelines are also part of the mix. The goal is to provide investors with exposure to these dynamic and essential parts of the economy.

Investment trends have shifted rapidly toward hard assets and infrastructure. What used to be seen as the old economy is now at the center of investor excitement. This transition is driven by massive energy needs for data centers and new power alternatives like nuclear and renewables. Tyler observes that the current cycle is unique because the capital requirements are so large they are creating significant challenges for companies and consumers alike.

In the world that I live in, which is the hard asset economy, having capital is very important. But we're now at a place in the cycle where the investment needs are so big that it's creating pretty big challenges for companies.

A major concern in the utility sector is affordability. When a utility doubles its investment in infrastructure without increasing its customer base, the costs for individual bill payers can skyrocket. This has turned utility bills into a central issue in local elections. However, the impact of data centers is not always negative for the average person.

The relationship between large-scale data centers and local utility rates depends on the existing infrastructure. If a utility has excess power generation that ratepayers are already subsidizing, bringing in a data center to consume that power can actually lower individual bills. The data center takes on a share of the costs that were previously resting on the shoulders of residents. This creates a wide dispersion in outcomes for different utility companies and their customers.

There are examples of data centers being really good for both the utility and the customer. But that's again very nuanced and it really depends on where you are and what your asset base looks like.

Data centers are not universally beneficial for local infrastructure. Their success depends heavily on local power generation, water access, and specific utility regulations. In some regions, like Wisconsin, negotiated tariffs ensure that hyperscalers cover the costs of new infrastructure. This protects local ratepayers from price hikes.

In Wisconsin, effectively the data center has zero impact on the local ratepayer. The hyperscaler has agreed to guarantee a return on a rate base for the capex that the utility is spending. It is not going to impact the rate payers there at all.

However, significant risks remain, particularly regarding stranded assets. If a utility spends billions to accommodate a data center that closes after a few years, local customers may be left to cover the costs. This creates a tension between the global need for more energy and the local challenge of rising electricity bills. The world needs more power for everything from AI to electric vehicles, but this growth must be balanced against the cost to residents.

Tyler describes his investment approach as being thematically informed relative value investors. The process involves identifying underappreciated themes and conducting deep fundamental research. Analysts visit sites and study local politics to find unique insights into utility contracts and regulations. By understanding these local nuances, the team identifies where regulation is improving or where a specific contract might benefit a small utility.

We want to find good themes that are underappreciated and underpriced and then find the best investment opportunities to take advantage of those. If you do that, we think you can generate really good investment results.

Tyler explains that as public markets investors, their focus is on identifying listed securities best positioned for growth over one, three, or five years. Unlike private equity where deal flow might arrive via offer letters, public market investing involves constant activity. Capital flows into their core strategy through open-end mutual funds, active ETFs, and separate accounts. The primary goal is to remain positioned in a way that outperforms established benchmarks.

For us, it is sort of your traditional equity research function where we are constantly trying to make sure that we are leading edge in terms of what is happening in markets. And identify, hey, we think that this thing is going to happen to the North Dakota utility as they invite a local data center customer there that is underappreciated by the market.

This approach relies on fundamental equity research to find value that the broader market has overlooked. It involves a boots on the ground methodology to identify specific catalysts. For example, understanding how a local utility might benefit from a new data center requires looking beyond top-level market data to find unique insights through specialized processes.

Classically cyclical companies like Caterpillar were once tied strictly to the rise and fall of the economy. However, many of these businesses are now being valued as secular winners. Tyler explains that while cycles still exist, they have become deeper and last much longer than in the past. This change is largely due to the natural resources sector being starved of capital for a long time. During that period, many industries consolidated, leaving the necessary expertise in the hands of only a few major players.

We've exited what we talked about as the era of abundance and we've entered the era of scarcity. We just don't have enough of all the stuff that we need for the economy to grow.

The transition from an era of abundance to an era of scarcity means that the companies capable of addressing these shortages are in a unique position. Because there is less competition following years of consolidation, these companies can earn above average returns and achieve more predictable growth. This reduction in volatility makes their current growth profile look very different than it did ten years ago.

The sort of secular growth and the reduction in volatility of that growth is very different now than it was 10 years ago.

Overcapacity in infrastructure is an unavoidable reality of the commodity cycle. Tyler explains that while infrastructure assets like freight rails and utilities often function as local monopolies, they are still subject to market cycles. For example, the shale pipeline boom in North America led to excess capacity when oil prices eventually dropped. Investors must try to sidestep these moments by identifying when market expectations become excessive.

You can't avoid it. That will happen. On the commodity cycle, the cure for low prices is low prices. The cure for high prices is high prices. Same thing on the infrastructure side. As investors, our job is to try to sidestep it. Understand when the market is getting too excessive in terms of its expectations.

Modeling global energy demand involves three main factors: population growth, the size of the global economy, and energy intensity. While population and economic growth continue to rise, the ability to convert energy into economic output is the most complex variable. Predictions for 2040 previously assumed significant gains in energy efficiency. However, recent trends show that economic growth is becoming more energy intensive, driving demand estimates even higher.

The scale of this transition is massive. To meet rising demand while also reducing coal consumption, the world needs to add roughly 60,000 terawatt hours of renewable supply by 2040. This is equivalent to recreating the entire global crude oil industry, which took a century to build, in only sixteen years. We are currently in a period of energy addition where the world needs more of every energy source.

Energy priorities have undergone a significant shift. Five years ago, the primary focus was on making energy clean. Stability and quantity were secondary concerns. Today, that order has reversed. The immediate need is for more energy that is stable and reliable. While cleanliness is still a goal, it cannot be prioritized over the basic requirement for a massive and steady supply.

We need more, it has to be stable and then we do want it to be clean, but we can't necessarily sacrifice the clean for the more in the stable part of it.

Tyler notes that large technology companies are responding to this reality by investing heavily in various energy sources. They are moving from coal to natural gas and trying to revitalize the nuclear sector. This includes funding for small modular reactors and reviving existing plants. The plan is to build enough capacity to meet demand first. Once the system is robust, the less desirable energy sources can be shut down. This change in approach is becoming more widely recognized within the market, even if it is not always the focus of public announcements.

Global energy models suggest that coal generation will likely be cut in half by 2040. While reaching zero coal usage is a common goal, Tyler argues that a pragmatic view shows the world will rely on coal for a significant period. Coal will represent a much smaller percentage of the energy market as it declines over time, but total elimination on a global scale by 2040 is improbable.

The sort of energy pragmatist would say, around the world coal will be relied upon for a really long time. It will be a much lower percentage of energy markets and we think it will sort of decline over time. But again, the idea of zero coal around the world by 2040 I think is highly unlikely.

Energy choices are often a reflection of what a country naturally possesses. In North America, the abundance of natural gas and crude oil creates a different economic calculation for renewables than in Europe. Because Europe must import much of its energy, it has aggressively pursued wind and solar to harness its own natural resources and move toward self-sufficiency. China is following a similar path for different reasons. As a major importer, China is working to reduce its dependence on the global supply by investing heavily in nuclear power, coal, and other energy sources.

China depends on the world for energy supply and they are trying to reduce that and they want to be more independent. They are going all in on nuclear. They are investing massive amounts in their nuclear economy, coal generation as well, kind of everything. And I think just an effort to be more self-sufficient.

This drive for energy independence is not unique to China. In a world shaped by the pandemic and rising geopolitical tensions, nations everywhere are prioritizing self-sufficiency. While some are skeptical about the future of nuclear energy in the United States, pointing to the long timelines of projects like the Vogtle plant in Georgia, Tyler believes the outlook is more optimistic. He suggests that the global shift toward securing domestic energy supplies may lead to a broader nuclear expansion than many expect.

Traditional energy sources like natural gas and coal offer reliability. They provide power 24 hours a day, 365 days a year. However, they have an emissions profile that is no longer desirable. Alternative energy sources like wind and solar are clean, but they are intermittent and variable. This creates a challenge for industries that require constant power.

Nuclear is the one resource that can serve both masters. It is 24/7, 365, low variable cost, very high capacity factor. And it's also pretty clean.

For a data center CEO, the primary concern is the power going out. These facilities are expensive shells that require constant cooling and electricity. They cannot risk the intermittency of solar or wind. They need baseload power. Tyler explains that fossil fuels like coal and natural gas effectively function as energy batteries. We currently lack viable batteries for wind and solar that can compete on that level. Until storage technology evolves, nuclear remains the most viable option for clean, reliable energy.

The nuclear renaissance is currently unfolding in stages. For two decades, nuclear capacity was being shut down globally, including in places like Germany. The first step in changing this trajectory is simply stopping the shutdowns. This effort is well underway. The second phase involves restarting facilities that were recently turned off. Over the next couple of years, the focus will shift to bringing places like Three Mile Island back online.

Phase three would be this sort of brownfield inside the fence nuclear facility build out. You worry about NIMBY issues, site supply, security, safety. I think that's been talked about here as well. We do think that that's going to start to pick up. But that's like a 2032 to 2035 in service.

Looking further ahead, newer technologies like Small Modular Reactors and thorium might become reality between 2035 and 2040. While there is optimism about seeing new nuclear facilities in the United States by 2040, Tyler believes no utility will take on that risk alone. There is almost no chance a utility will start a greenfield project without a guarantee against cost overruns. Instead, the government may need to step in, much like it has with critical minerals, to backstop these projects and shoulder the excess cost burden.

These supply chains are not going to come about because of market forces. You're not seeing a reaction in copper production, you're not seeing a reaction in uranium mining, you're not seeing a reaction in nuclear generation without direct government catalyst.

Ultimately, the transition requires a shift from allowing the nuclear sector to melt away to actively refreezing and rebuilding it. This process is not driven solely by market forces but by direct government intervention and public-private partnerships.

Investors and companies have forgotten natural resource markets. Copper and gold prices are at all-time highs. Yet major mining companies are cutting their spending plans. Usually, high prices lead to more supply. This time, investors are revolting against spending. They remember how much money was lost in shale oil between 2010 and 2020. They want discipline instead of growth.

The investors just revolt, they say no mas, they want the discipline. You have destroyed so much value. Shale destroyed a lot of value 2010 to 2020. And so it is not going to happen naturally. Maybe that is okay because these management teams should continue to be held to the discipline, but we do need the supply.

Market signals are not working. Now the government is stepping in to help. We see this with new contracts for nuclear power, lithium, and rare earths. Investors remain cautious. They worry about political changes and poor past returns. Tyler thinks people must see these old economy sectors as the new economy. This shift is needed for capital to flow again.

The spreadsheet math might say hey, start drilling again. But the history would say no, the returns need to be way better to justify that. It is a rational response by the investor base. People need to start looking at the old economy stuff as the new economy stuff.

Infrastructure investors focus heavily on regulatory risk, a factor that often receives less attention than it deserves. While a CEO might worry about operational failures like power outages, an investor's main concern is regulatory surprise. Tyler explains that assets such as airports and utilities are governed by regulators, and unexpected changes in their decisions can be devastating for an investment.

What keeps you up at night? For us, it is regulatory risk. You are investing in airports and utilities that are governed by a regulator. We worry about surprise.

Regulatory surprises can occur even in established markets. For example, utilities in Illinois might spend capital on improvements and request a higher return in their rate base, only for the regulator to deny it. To navigate this, investors must have a deep understanding of local regulatory environments to identify potential shifts before the rest of the market does.

Investment in regions experiencing political instability requires a deep understanding of local legal frameworks. Tyler explains that for large foreign direct investments in infrastructure, the risk of assets being nationalized or stranded is a major concern. Even if there is hope for quick resolutions during a regime change, it often takes a long time before the environment becomes safe enough for investors to commit capital.

As an infrastructure investor, the risk of expropriation, nationalization is very high. And so you don't want to go and suddenly see your assets stranded there.

The regulatory environment for building infrastructure in North America has also evolved. Tyler mentions a transition from a NIMBY mindset, meaning not in my backyard, to what he calls a BANANAS world. This acronym stands for building absolutely nothing anywhere near anything. Between 2018 and 2024, this sentiment made it nearly impossible to develop new projects. However, a more pragmatic approach is emerging as political leaders show a greater desire to build.

While oil prices might make some projects look questionable, certain types of infrastructure are still economical. The current growth is not in oil pipelines but in natural gas. Data centers are the primary customers driving this demand. Because these centers have an urgent need for power, they are willing to pay the prices required to make new natural gas pipelines viable.

While it is widely known that energy is a major bottleneck for the current technological boom, the scope of the problem extends beyond just data centers. Electricity demand in the United States was stagnant from 2007 to 2020, but it is now accelerating. Projections suggest a shift to 2.5% annual growth. Only about half of this new demand comes from data centers. The rest is driven by electric vehicles and the broader industrial system. This change from zero growth to significant expansion represents a massive shift for a large industrial system.

I'd start with we agree electricity is constrained. The demand is going to keep rising. I would note it's not just data centers. Only about half of it is data centers. There's lots of electricity demand coming from EVs and from the industrial system.

Tyler identifies a specific opportunity in the utility sector. Currently, the fastest-growing utilities are trading at a much smaller premium compared to average utilities than they have historically. Eight years ago, investors paid an 11% premium for a 1% growth advantage. Today, they only pay a 6% premium for a 2% growth advantage. This valuation gap exists because of concerns over regulation and affordability, but Tyler believes the best-performing companies will eventually pull away from the pack.

Another area of interest involves the companies providing the infrastructure for this growth. Engineering and construction firms, as well as consolidated industries like aluminum smelting, represent a picks and shovels approach to the energy gold rush. These businesses often offer better risk-adjusted returns because their growth is becoming more structural and predictable, even if they lack the high-octane volatility of direct energy plays.

Who made all the money in the gold rush? It was the Levi's and the picks and shovels companies and I kind of think that's where we are. You might get the best absolute returns in some of the direct ways to play this, but they might come with a lot more volatility.

Investing in infrastructure requires a deep focus on operational risk. For example, an oil spill in a pipeline represents a massive impairment to the asset. To assess these risks properly, Tyler and his team look beyond the CEO. They speak with plant managers and even ask leaders at competing companies about how well their peers maintain their equipment. It is important to understand the incentives and local relationships that keep these systems running safely.

With infrastructure, operational risks have major asset impairment risks with them. So we try to talk to the plant managers and the next rung down. We ask one CEO what he thinks about the assets of another company and if they are maintaining them well.

There is also growing local backlash against data centers. While these projects require a lot of power, they do not create many permanent jobs once the construction phase is over. This can lead to a situation where local residents see their utility bills rise by fifteen percent without receiving any clear economic benefit. This friction often plays out in utility commissions. Elected commissioners are typically more sensitive to local voter concerns than those who are appointed by a governor.

The scale of new power demand is staggering. One utility in the Midwest spent a hundred years building a system with eleven gigawatts of capacity. Now, that same utility is facing requests for fifteen gigawatts of new power just for data centers.

A hundred years ago to eleven, and by tomorrow they want another fifteen. That is really expensive. And so you are seeing this tension and I think that is reasonable.

The energy sector is currently positioned in front of a massive influx of capital. Between government spending, domestic security needs, and secular trends, it feels like standing in front of a fire hose of money. However, regulatory risk remains the primary concern. Shifting political administrations can change the viability of projects that rely on specific commodity levels or clean energy incentives over long periods.

The government is the key risk here, the regulatory risk. Even if the numbers pencil out at current commodity levels, it might not work because your assumption is that in three years we will have a new administration or the pendulum will swing back.

Investment in places like Venezuela remains a difficult proposition. Revitalizing their infrastructure would require upwards of $100 billion, yet the political regime and market uncertainty make it a hard sell. There is a desperate need for more visual data and reporting regarding the actual state of these facilities. Understanding the physical state of the equipment and the efficiency of loading tankers is essential before committing such massive capital.

The sheer scale of modern energy demand is difficult to comprehend. One utility company in the Midwest spent 100 years building 11 gigawatts of capacity, but now they project needing another 15 gigawatts in a much shorter timeframe. To visualize the scale, one gigawatt is roughly enough to power a city like Denver. While energy is currently the main constraint for technology companies, a breakthrough in chip efficiency could eventually reduce these staggering electricity requirements.

It took them 100 years to get to 11 gigawatts and now they are projecting 15 gigawatts more. These are staggering numbers. It makes sense why literally every company in this space is saying it is not chips, it is the energy.