Equity Research Industry Update — March 30, 2026 Independent Power Producers IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance    Our Call We are still bullish on the IPP space despite weak YTD performance and investor angst. And, despite key bear arguments, which we address in this piece, we are not swayed. YTD ’26 weakness presents an entry point.   Early ‘26 weakness in IPP’s have left investors searching for answers, and unlike the straightforward asset scarcity thesis of 2025, the thesis in 2026 has become much more complex. We address 4 key bear arguments against the IPPs. Overall – even with new alt. generation announcements and optical policy risks, the fundamentals remain unchanged, and pullbacks are an attractive entry point, in our view.   Bear Thesis 1: DCs choosing vertically integrated markets, BTM providers, driving down the scarcity value of existing assets. Recent datapoints show that there is plenty of DC demand to go around for competitive markets. Off grid BTM has only a handful of projects, none at COD, and both IPPs/T&Ds agree that DCs bring BTM as a bridge with a goal of on-grid pathway (opening for more virtual deals with IPPs). S/D remains tight and resolution is 3-5+ year path (i.e. scarcity has longevity).   Bear Thesis 2: Re-regulation or government intervention risks at the state level in the name of resource adequacy, instead of addressing missing market signals. We could see a route to a middle ground solution between IPPs and wires co.’s via Long Term Resource Adequacy Agreements (LTRAAs), open competitive procurements, more likely than full re-regulation, in states that want it.   Bear Thesis 3: Speculative Large Gen Projects such as US-JAP deals, nuclear consortium create market share risks. The commentary from vertically integrated companies on new AP1000 developments have raised eyebrows (ETR expecting a potential hyperscaler/ regulated nuke deal as early as '26, and DUK, D being open to the idea). In our view this doesn't impact deregulated markets and the development timelines for a new large-scale build is ~10+ years, which doesn't move the needle on scarcity today.   Bear Thesis 4: Power demand/AI backdrop risks – what if we hit a plateau in future AI- driven compute demand? Bears argue there is growing risk that this demand fails to materialize at the pace, scale, or durability currently embedded in forward curves and valuation assumptions. We do not find merit in this argument as it directly contradicts too many proof points of demand, including utility guidance, hyperscaler commentary, government and our internal data work. We provide our case within.     Shahriar (Shar) Pourreza, CFA Equity Analyst | Wells Fargo Securities, LLC Shahriar.Pourreza@wellsfargo.com | 212-214-5422 Constantine Lednev Associate Equity Analyst | Wells Fargo Securities, LLC Constantine.Lednev@wellsfargo.com | 212-214-5438 Alexander Calvert Associate Equity Analyst | Wells Fargo Securities, LLC Alexander.J.Calvert@wellsfargo.com | 212-214-5434 Whitney Mutalemwa Associate Equity Analyst | Wells Fargo Securities, LLC Whitney.Mutalemwa@wellsfargo.com | 212-214-6428 Anders Myhre Associate Equity Analyst | Wells Fargo Securities, LLC Anders.S.Myhre@wellsfargo.com | 212-214-6409 Andrew Kadavy Associate Equity Analyst | Wells Fargo Securities, LLC Andrew.Kadavy@wellsfargo.com | 212-214-5485 All estimates/forecasts are as of 3/27/2026 unless otherwise stated. 3/30/2026 5:00:00EDT. Please see page 16 for rating definitions, important disclosures and required analyst certifications. Wells Fargo Securities, LLC does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of the report and investors should consider this report as only a single factor in making their investment decision. This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research Table of Contents Table of Exhibits  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  3 PM Summary  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  4 Addressing Bear Arguments  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  5 Bear Argument #1: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Bear Argument #2: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Bear Argument #3: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Bear Argument #4: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Appendix: North American Power, Utilities & Infrastructure IPP research  . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .  14 2 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research Table of Exhibits Exhibit 1: IPP Performance....................................................................................................................................................... 4 Exhibit 2: Incremental DC Demand for Regulated and Deregulated Markets through 2030................6 Exhibit 3: Capacity Cost by Technology............................................................................................................................10 Exhibit 4: Total of Data Center Capacity Deployed and Planned Through 2030.........................................11 Exhibit 5: Operational and Future Capacity by Interconnecting Utility ..........................................................12 Exhibit 6: Selected Management Disclosures of Capacity Growth Contribution (in GW).....................12 Exhibit 7: Annual Demand Outlooks - PJM....................................................................................................................13 Exhibit 8: Annual Demand Outlooks - ERCOT ..............................................................................................................13 Exhibit 9: US Data Center Spending; Construction Put in Place vs Starts......................................................13 Equity Research | 3 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research PM Summary This year, it has gotten more complicated pitching IPPs to incremental buyers, growth investors etc. vs. 2025. Early ’26 weakness in IPP stocks has left investors searching for answers and posing theories about why the stocks have been down, but in our view the IPP landscape has not fundamentally shifted despite external noise. We still believe the IPPs are well-positioned and undervalued, and we disagree with bear theses around the rise of other sources or providers of generation, or a potential drop-off of demand. Adj FCF yields—the expression of under/over valuation for IPPS—have increased +110bps YTD, moving away from the mid-to-high single digits we believe they should trade at (i.e. contracted infrastructure). With the shares trading at adj FCF yields of 10%, 12%, 10% and 18% (TLN, VST, CEG, NRG respectively), we remain bullish on the IPPs as a group, and the early ’26 weakness presents an attractive entry point on what is, in our view, still one of the best ways to play power demand. Exhibit 1 - IPP Performance IPP Coverage Adj FCF Yield (2028E) 12/31/2025 3/27/2026 YTD (%) 12/31/2025 3/27/2026 TLN $375 $325 (13.4)% 8.3% 9.5% VST $161 $155 (3.6)% 11.5% 12.0% CEG $353 $301 (14.7)% 8.8% 10.3% NRG $159 $148 (7.2)% 16.5% 17.8% IPP Group Avg (9.7)% 11.3% 12.4% S&P500 $6,846 $6,369 (7.0)% UTY $1,068 $1,148 7.5% YTD Price Performance Source: Bloomberg, Wells Fargo Securities, LLC We have also been hearing with increasing frequency the IPPs have ‘been harder to pitch this year than in the past’ . From our perspective, though some aspects of the IPP argument have gotten more technical and harder to make resonate with generalist/growth investors vs. last year (policy intervention, capacity markets, backstop auction, BTM dynamics), we still think the IPPs are one of the best ways to play power demand – near term cash flows are significantly hedged, new entry into the markets have significant barriers including challenging new build economics, and incremental LT deals present opportunity to create counter-cyclical strategy. The backstop in the NT at current yields is increased allocation to buybacks. We will address 4 key bear arguments we have been fielding recently about the IPPs. Bear Case 1: Data centers choosing vertically integrated names, BTM providers, is driving down the scarcity value of assets. • Counterargument: Scarcity value is not going away. Data center demand continues to show a preference for deregulated over regulated markets (84GW vs 73GW future planned capacity). In deregulated markets, IPPs remain the most direct and levered way to gain exposure to rising power demand. We do not see vertically integrated utilities or speculative behind-the-meter projects meaningfully diverting demand away from IPPs. Plus, the scale of incremental load growth is large enough that multiple solutions can coexist, and in our view there is plenty of demand to support the IPPs alongside regulated names – this doesn’t need to be a zero-sum game. Bear Case 2: Re-regulation policy risks in states like PA, NJ, NY , MD where legislation could enable wires companies to build or own generation again, sidestepping, rather than fixing, the lack of effective market signals for IPPs to invest, instead shifting incremental supply into rate base and potentially diluting IPPs newbuild opportunities and scarcity value. • Counterargument: We see proposed state interventions as reflective of concern over resource adequacy, not spurred as a rejection of competitive markets. To this point, we think there could be a clear avenue for middle ground solutions in the form of Long-Term Resource Adequacy Agreements (LTRAAs), open competitive procurements/backstop auctions (where IPPs will be well-positioned to compete) that de-risk new entry through longer-dated contracts/capacity commitments (potentially backstopped by hyperscalers), rather than socialize generation through full re-regulation. 4 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research Bear Case 3: The vertically integrated Nuclear Consortium, US-Japan CCGT deal (GE Vernova Hitachi SMRs – TN/AL, Project South Mon – PA, and Project Anderson - TX), and PORTS Technology Campus - OH as well as any future deal announcements present headline risks and are reflective of a shift in new generation going forward away from the IPPS and backstopped by third parties. • Counterargument: We feel these large-scale announced projects remain too thin on details and long-dated to shift the fundamental IPP thesis. The vertically integrated nuclear consortium could potentially add 11GW (assuming 10 units), but nothing has been officially announced and any impact on supply could take over a decade to be realized and in traditional vertically integrated states. Plus, in our view, the implied scale of the recent CCGT gas announcements is still modest compared to underlying demand. PJM announced projects (excluding nuclear consortium) includes ~4GW in SW PA + ~10GW in OH (could include incremental datacenter load; i.e. net S/D neutral), do not offset ~46.2GW of new ELCC capacity we estimate is required to maintain current reserve margin levels through 2030. • Hyperscaler development timelines are far more aligned with IPP uprates, incremental capacity additions, and increased utilization of existing assets than with long-dated greenfield generation. Hyperscalers are in an arms race to bring compute online as quickly as possible, which makes waiting 3–5 years for new-build gas and a decade or more for nuclear untenable with NT needs. As a result, long-cycle generation solutions do little to alleviate near-term power scarcity, particularly in the markets where load is accelerating the fastest. This dynamic is already forcing hyperscalers to pursue short-term, flexible solutions, including BTM generation arrangements, to bridge the gap between immediate compute needs and longer-term system build-out. In this environment, virtual PPAs (VPPAs) and similar offtake structures provide a faster, more scalable solution that will directly benefit IPPs. • Bear Case 4: We address concerns around power demand and AI backdrop risks. • Counterargument: We do not view this bear argument as reflective of current planning realities. Our load outlook is not dependent solely on data centers - even under a more conservative scenario where AI-driven demand underwhelms expectations, we see a broadly constructive load growth environment supported by electrification, industrial expansion, infrastructure investment, and regional economic development. That said, we continue to believe AI- and data center–driven demand will continue to expand in the coming years, and this view is grounded in our proprietary data (see DC report HERE). We identify ~120 GW of planned and in-construction data center capacity through 2030/31, representing a meaningful source of incremental, high–load-factor demand that will need sustained generation investment rather than representing a diffuse or speculative load tail. Overall, we do not think the scarcity value of the IPP assets has been degraded, and we do not see meaningful movement in closing the supply-demand gap, especially in PJM, in the near term, unless it is through a mechanism where the IPPs are well positioned to compete and benefit. We do not expect alternative generation such as BTM, the Nuclear Consortium, or Japan-funded CCGTs to make a meaningful dent in demand, even if they can get off the ground, and insofar as market mechanisms such as open competitive procurements or the reliability backstop auction process may help address the issues, we feel IPPs are the best positioned in the market to benefit and potentially lock in guaranteed 15-year contracts to build. We also do not have concerns about AI power demand dropping off, given the projected demand for compute, and alignment with national security priorities, but we also highlight that non-AI industrial load is also significant and represents in and of itself a high floor for power demand should the AI market contract for any reason. We rank the IPPs by adj free cash flow yield to assess relative valuation from cheapest to most expensive. On this basis, the ordering is NRG (17.8%), VST (12.0%), CEG (10.3%) and TLN (9.5%). While this framework suggests CEG screens as the most expensive on Adj. FCF, our stated preference remains CEG given the NT 3/31 business update call, followed by NRG, TLN, and VST, reflecting our view that factors beyond absolute FCF yield—balancing valuation with NT catalysts, asset quality, durability of cash flows, and risk profile. Addressing Bear Arguments Early ’26 weakness in IPP stocks has left investors searching for answers and posing theories about why the stocks have been down despite very strong fundamentals. While we have heard a variety of Equity Research | 5 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research theories, we address the four bear arguments that have come up the most with investors, and why we disagree that they change the fundamental backdrop on IPPs. Bear Argument #1: Bear Thesis 1: Investors will argue that vertically integrated utilities and alternative non-grid solutions are eroding scarcity value of IPP assets. They argue that a growing share of incremental generation and infrastructure investment is being captured by regulated utilities, vertically integrated operators, and potentially non-IPP behind-the-meter (BTM) generators, diluting the scarcity value historically ascribed to IPP assets. Against this backdrop, bears argue there is little incentive to allocate capital to IPPs when vertically integrated states continue to offer scale, visibility, and more predictable returns, potentially capping valuation upside for the IPP space. Wells Fargo Take: Scarcity value is not going away. Data center demand continues to show a clear preference for deregulated markets, where IPPs remain the most direct and levered way to gain exposure to rising power demand. We do not see vertically integrated utilities or speculative behind- the-meter projects meaningfully diverting demand away from IPPs. As we discussed in our data centers piece (see HERE), the scale of incremental load growth is large enough that multiple solutions can coexist, and in our view there is plenty of demand to support the IPPs alongside regulated names. Despite growing attention on vertically integrated examples such as Entergy, Duke and Southern Co, data center growth through 2030 and beyond remains disproportionately focused in deregulated power markets despite optics. While we acknowledge nationwide data center demand, based on current data, ~84GW of incremental data center demand will land in deregulated markets vs. ~73GW of incremental demand in vertically integrated markets. Currently, deregulated markets lead regulated markets ~65GW to ~49GW in installed data center capacity. The majority of demand is landing in already tightening markets where scarcity accrues to merchant assets, not utilities, and we do not see this trend shifting meaningfully any time soon. This is reflective of a continued bias in data center siting toward deregulated markets due to the proximity to population centers, historic concentration of data center infrastructure and expertise in PJM, and speed to market and regulatory favorability in ERCOT . We do not see affordability or capacity market-related political concerns in PJM diminishing its position as the foremost data center market in the nation. As a result, we continue to see IPPs having the most earnings torque from incremental data center-driven load growth, with unregulated value upside from both higher margins and the anti-cyclical, LT nature of contracts (i.e. the scarcity value of merchant assets remains intact and/or improving). Exhibit 2 - Incremental DC Demand for Regulated and Deregulated Markets through 2030 Source: S&P 451, Wells Fargo Securities, LLC While we also acknowledge there have been several announcements of BTM off-grid or islanded projects. We see these as either 1) transitional, as a speed to market stop-gap while data centers 6 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research wait for grid interconnection - and both IPPs and T&Ds agree this accounts for many cases, or2) experimental, and we do not see them as representative of a larger industry trend, given that they cannot meet the reliability requirements needed by the vast majority of data centers.# This is consistent with commentary from IPPs, T&D utilities, and hyperscalers, all of which indicate that grid-connected solutions remain the first and preferred option. Data center developers do not want to run power plants, they want to build data centers. Partial BTM solutions which maintain grid connection we do feel are viable and could represent some degree of demand offset, but our analysis shows that BTM with grid connection actually seems to increase utility and IPP demanded power via the grid (~57GW across 47 projects after our estimated offsets, ~69GW w/ no offset) as compared to projects with no onsite power (~45GW across 47 projects). Consistent with this, the average project size is materially larger for sites with some onsite generation (1,454 MW) versus those without (960 MW). Our analysis of planned hyperscaler projects reveals that 52% of planned projects by hyperscalers will incorporate some form of onsite behind-the-meter power. We note that of these, only 11 of 98 projects have CCGTs, mobile gensets, or nuclear that could technically enable them to be “off grid” or “islanded,” although even these options will come with potential reliability issues. The rest of the projects with onsite generation have renewables, storage, diesel backup generators, etc. that would be insufficient to totally power a data center, and instead provide RECs or serve as emergency backup. Of those 11 planned hyperscale projects with sufficient dispatchable generation, we believe only 3 plan to attempt to be entirely off-grid – Crusoe Energy’s Project Jade, Prometheus Hyperscale, and Fermi Matador. These projects represent ~3% of the total number of projects and ~11.5% of total data center MW (**driven by planned 11GW Fermi Matador, excluding Fermi its 3.0%). Crusoe will use natural gas turbines, fuel cells, and BESS, Prometheus plans to use SMR nuclear, and Fermi plans to use nuclear. None are functional yet, and we wait to see if these proofs of concept are viable at scale. So what? Even where BTM adoption occurs, it does not undermine the merchant power thesis, in our view. Most BTM configurations are not a threat to grid demand and may even expand total load requirements rather than displace them, while truly off-grid solutions remain rare, speculative, and operationally complex. As a result, data center growth will continue to flow through competitive power markets, reinforcing the role of IPPs as the marginal suppliers of capacity. Equity Research | 7 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research Bear Argument #2: Bear Thesis 2:State-level policy increasingly favoring utility-owned generation as alternative IPPs. In states such as PA, NJ, MD, and NY, recent policy initiatives are considering structures to support generation investment by regulated utilities given the supply constrained environment and lack of economic basis for IPPs to build. Bears argue this dynamic disadvantages IPPs, as utilities benefit from guaranteed cost recovery, and lower capital risk, while IPPs face higher competitive hurdles. As a result, they argue that incremental generation growth in these markets may accrue disproportionately to regulated platforms, compressing the long-term opportunity set for IPPs. Wells Fargo Take: While we acknowledge rising political pressure around affordability and reliability, particularly in PJM states, we do not believe current policy initiatives amount to a meaningful re-regulation of generation or a structural shift away from IPPs. We believe IPPs are still the best positioned to build in PJM – they have the development capabilities, the turbine queues, the sites etc. Concerns about affordability and reliability, as well as pushes from wires companies, have state governments exploring reregulation opportunities which some view as potentially eating the lunch of the IPPs. But recent state-level actions simply reflect concern over supply adequacy, not a rejection of competitive markets. We think there are routes to a middle ground solution that eases supply side issues without yielding IPP market share directly to regulated segments. Specifically, we see Long Term Resource Adequacy Agreements (LTRAAs), competitive procurements, and the reliability backstop auction process as benefitting the IPPs as builders of choice in deregulated markets. These frameworks are largely designed to de-risk new entry through longer-dated contracts/capacity commitments, rather than to socialize generation through full re-regulation. Several states have ongoing legislative or regulatory processes exploring routes to generation acceleration in various forms, but proposals typically require competitive RFPs or third-party participation first, limiting the scope for utilities to displace IPPs outright – even in vertically integrated states like GA and MI, 3rd parties win a portion of the RFPs. Such is the case with Pennsylvania House Bill 1272 (see HERE) which would allow utilities that demonstrate "resource inadequacy" to petition the commission for approval to invest in new generation or enter long-term resource adequacy agreements to invest in new generation. Pennsylvania Senate Bill 897 (see HERE) is a similar bill. It would also enable utilities to own generation as part of long-term resource adequacy agreements where utilities invest in new generation resources as partial owners for a share of the revenues, or as full owners in some cases. In this version, before petitioning for this right, the utility must conduct at least one RFP to solicit third- party projects. Both bills remain in committee. In Maryland, utility ownership has also been discussed, but remains contentious, and the state ultimately took steps in another direction, though the Utility RELIEF Act in the Senate right now has language about a competitive procurement funded by alternative compliance payments to incentivize new construction. New Jersey has similarly emphasized IRP-driven identification of capacity needs, followed by open auctions, a structure explicitly endorsed by PEG management. These approaches preserve competition and position IPPs as the builders of choice, even if utilities play a coordinating or contracting role or are granted the ability to bid in themselves. Given their institutional knowledge, supply chain relationships, and incentive to find competitive efficiencies vs. regulated names coming out of hibernation to participate for the first time, we see the IPPs as poised to be key players in an emerging new resource development cycle, rather than bystanders. There is still market reform work to be done, and we will caveat that these solutions do sidestep, rather than fix, the lack of effective market signals for IPPs to invest. The first round of demand and supply datapoints for the next PJM capacity auction (28/29) were released last week and given preliminary numbers, we see peak load increasing +1.4GW, and both existing resource ICAP and implied UCAP moving lower (-3.1GW and -0.3GW respectively). In simple headline math – installed reserve margin remains below 20% target, pointing to pricing at cap signaling a tight market for years to come. Given the political constraints imposed by PJM governors and the Trump Administration, short of a true economic capacity procurement without the cap (which we see as politically unviable) the proposed RBA process will be a necessity and will favor the IPPs as preeminent owners and builders in PJM. The proposed PJM Reliability Backstop Auction (RBA) further underscores the IPPs favorable positioning in the tightening merchant market similar to 8 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research open procurements, plus in this instance the IPPs have the uprate route to bid, an additional high return, low capital play. The RBA process is being designed as a one-time, transitional procurement to address near-term reliability risks following capacity shortfalls driven by accelerated data center load growth, and would procure new-build capacity under long-term commitments of ~15 years, explicitly to enable financing and development of incremental supply. From an IPP perspective, the RBA highlights a tangible opportunity: long-dated contracted revenue streams to support new entry, accelerated pathways for capacity buildout, and further validation that scarcity is being addressed through competitive procurement. We do not see rate-based generation forming at scale, nor do we see vertically integrated utilities or speculative BTM projects crowding out IPPs. Instead, emerging frameworks can potentially lower risk through contracts while preserving competition, leaving IPPs best positioned to participate and win. Bear Argument #3: Bear Thesis 3:Long-dated, policy-backed new builds could pressure incumbent generation and reset market norms. Bears argue that speculative, large-scale new generation projects could ultimately pressure the value of existing generation assets and alter long-term market economics. Two developments are cited as potential overhangs: 1) renewed nuclear new-build risk from the AP1000 consortium; 2) NEE’s announced approval from the White House to build generation in support of the U.S.–Japan agreement (GE Vernova Hitachi SMRs – TN/AL, Project South Mon – PA, and Project Anderson - TX): and 3) PORTS Technology Campus – OH) which introduces uncertainty around how future capacity demand will ultimately be served. If these projects move forward—particularly under utility-led or consortium structures with strong policy and regulatory support—they could add meaningful baseload supply over time, compressing energy prices and capacity values for incumbent fleets. This risk is most acute for IPPs with exposure to legacy nuclear or thermal assets, as new builds would likely benefit from more favorable cost recovery mechanisms, longer asset lives, and better policy alignment, potentially placing existing assets at a structural disadvantage. Wells Fargo Take:Following the group’s ~11% pullback in response to the NEE Japan headlines, we acknowledge the headline risk for IPPs and the broader concern that government could “put a thumb on the scale” in a way that distorts returns and puts capital at risk of low/regulated economics. That said, we believe the magnitude and speed of recent headline-driven share price moves largely reflect increased participation from generalist investors exhibiting elements of “FOMO.” This dynamic often leads to crowded long positioning in IPPs. When non-IPP related, large-scale generation projects are announced, these incremental investors tend to exit quickly; because they were the marginal buyers, selling pressure emerges without a natural buyer stepping in to absorb shares, in our view. We would highlight that meaningful upside exists in optimizing and uprating existing gas assets— many of which have been running below potential (i.e., built to peak) and can operate at higher utilization levels as incremental hyperscaler demand materializes. Across the fleet, there are both asset-owned and acquisition opportunities for IPPs if additional contracts to serve hyperscalers come online. Importantly, this pathway offers hyperscalers a faster, lower-risk, speed-to-market solution versus long-dated new builds—by contracting existing generation, funding uprates/ efficiency improvements, and paying for targeted transmission upgrades to firm deliverability. Separately, we would caution that new-build gas economics remain challenging despite being viewed as the “lowest-cost” firm resource. Market chatter around recent large-load/gas arrangements underscores that these solutions can be very expensive on an installed basis (~$3– 4/kW) once you incorporate full EPC scope, interconnection, gas infrastructure, and execution contingencies. Put differently: while gas may serve as the interim backstop in the load buildout, the set of “doable” gas projects is not large enough (or fast enough) to be constraint-relieving at system scale, and the deals being discussed do not, on their own, solve the broader supply/deliverability bottleneck. Even the lowest-cost firm resource CCGT requires ~$500–$600/MW-day forward capacity pricing to pencil, consistent with our LCOE analysis (see here) and IPP commentary. Equity Research | 9 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research Exhibit 3 - Capacity Cost by Technology Source: PJM, S&P Capital IQ, BNEF, EIA, SEIA, Wells Fargo Securities, LLC Estimates Even in scenarios where a handful of gas projects advance, we view them as episodic / one-time in nature rather than a repeatable template that rapidly adds enough capacity to reset the market. Until meaningful incremental supply actually clears permitting, interconnection, and construction, scarcity still matters—which supports the case for contracting and improving utilization of existing assets as the “lowest-friction” near-term solution. Gas may bridge the gap, but new builds are expensive and limited in near-term scale, reinforcing the relative attractiveness of existing-asset contracting and uprates as the practical path to meet hyperscaler timelines. We would remind investors that neither the AP1000 consortium concept nor the referenced nuclear discussions have finalized commercial details—including structure, cost recovery, offtake, siting, permitting, interconnection, or construction risk allocation. As a result, a portion of the market reaction appears speculative, and may be more reflective of a political narrative around affordability and reliability than an executable development plan today. More importantly, timelines are measured in years, not quarters. Even under a constructive outcome for AP1000 development or the referenced projects, reactor builds are typically framed as ~10+ years from concept to commercial operation. That timeline reality means nuclear does not alleviate near-term constraints—and, until it is built, scarcity and deliverability constraints remain the binding factors. In that interim period, procurement is far more likely to continue favoring virtual PPAs, contracting existing generators, funding uprates/efficiency, and targeted transmission upgrades, with BYOG structures emerging over time. We also see a timing mismatch in the bear narrative: hyperscalers are unlikely to wait a decade for nuclear if near-term solutions exist. Accordingly, we view nuclear announcements and consortium discussions as long-duration options rather than near-term threats to incumbent assets. Said differently, even if nuclear ultimately arrives, the market has a long runway in which existing generators can re-contract and extend tenor (often 10–15 years across energy/capacity constructs), while counterparties absorb permitting, execution, and cost overrun risk. Finally, on scale, the nuclear items being discussed—even if they progress—only represents 11GW (10 reactors at 1.1GW each), a small subset relative to total system needs and the magnitude of incremental load being contemplated. They are not an immediate, broad-based reset of the supply stack; they are incremental, long-dated additions with significant development risk and long lead times. Bear Argument #4: Bear Thesis 4:Power demand growth may underdeliver relative to expectations. A central pillar of the bull case for IPPs is a sustained acceleration in power demand driven by data centers, electrification, reshoring, and AI-related load. Bears argue there is growing risk that this demand fails to materialize at the pace, scale, or durability currently embedded in forward curves and valuation assumptions. Project delays, permitting and interconnection bottlenecks, efficiency gains, or macro-driven demand softness could push incremental load further out. In such a scenario, supply-demand balances may not tighten as expected, limiting power price upside and reducing the operating leverage IPPs are positioned to capture. 10 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research Wells Fargo Take: We do not view this bear argument as reflective of current planning realities. Based on our prior work, we estimate approximately 120 GW of incremental load coming onto the grid by 2030 from datacenters alone, and importantly, our load outlook is not dependent solely on data centers. Even under a more conservative scenario where AI-driven demand underwhelms expectations, we see a broadly constructive load growth environment supported by electrification, industrial expansion, infrastructure investment, and regional economic development. Across our coverage universe, utilities and system planners continue to signal sustained demand growth, reinforcing our view that incremental load is diversified, durable, and supportive of tightening supply- demand balances and a constructive backdrop for IPPs. We continue to believe AI- and data center–driven demand will push utility load growth forecasts and capital budgets higher over the coming years, and this view is grounded in our proprietary data. In our most recent data center study, we apply a differentiated classification framework to projects that are either planned or under construction, allowing us to move beyond headline announcements and focus on projects with tangible development progress. Using this approach, we identify ~120 GW of planned and in-construction data center capacity through 2030/31, representing a meaningful source of incremental, high–load-factor demand that utilities must plan for. Importantly, this capacity is geographically concentrated within key power markets, reinforcing the need for sustained generation, transmission, and distribution investment rather than representing a diffuse or speculative load tail. Exhibit 4 - Total of Data Center Capacity Deployed and Planned Through 2030 Source: Company Data, S&P 451, Wells Fargo Securities, LLC Company disclosures corroborate our findings, with utilities reporting growing large-load pipelines supported by ESAs, engineering authorizations, and transmission planning activity, while hyperscalers continue to guide to elevated, multi-year AI-related capital spending. The alignment between our bottom-up database and company-reported pipelines reinforces our confidence that data center demand is real, visible, and appropriately embedded in forward supply-demand models. Equity Research | 11 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research Exhibit 5 - Operational and Future Capacity by Interconnecting Utility Source: Company Data, S&P 451, Wells Fargo Securities, LLC Importantly, IPP fundamentals are not singularly dependent on any one source of demand. Even if data center load ramps more slowly than expected, investors should recognize the increasingly diverse drivers of system-wide load growth. Company-level disclosures across our coverage point to sustained capacity growth tied to electrification, industrial expansion, reshoring, infrastructure investment, and regional economic development—not data centers alone. Utilities such as AEP , CMS, CNP , ETR, and SRE provide detailed multi-year capacity outlooks by source, reinforcing that incremental generation needs extend well beyond datacenters. Exhibit 6 - Selected Management Disclosures of Capacity Growth Contribution (in GW) Note: We have provided the exhibit above as a proof point behind mgmt incorporating other large load sources into planning. This list is not exhaustive. Source: Company Disclosures and Wells Fargo Securities, LLC Beyond bottom-up large-load work, commission and system-planner energy forecasts are moving higher, providing an independent confirmation that total energy demand is rising. Successive PJM forecasts step up materially up until 2025, with the 2026 forecast being down over previous in early years; however, it still, over time, adds and surpasses 2025, while ERCOT’s Annual Energy Forecast shows a similar upward revision. For IPPs, higher system energy—not just peak demand—implies a longer-duration call on generation, supporting sustained tightening risk, more frequent scarcity conditions, and a multi-year runway for merchant optimization, contracting, and incremental capacity additions even if any single load category ultimately underdelivers. 12 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research Exhibit 7 - Annual Demand Outlooks - PJM Source: PJM, and Wells Fargo Securities, LLC Exhibit 8 - Annual Demand Outlooks - ERCOT Source: ERCOT and Wells Fargo Securities, LLC We would also highlight additional ways we actively track to validate data center growth, with credit to the Technology & Services Team. First, we monitor U.S. Data Center construction activity, which serves as a leading indicator given the long lead times and upfront site work required before data centers become visible in utility load data. The most recently published US Consensus Bureau released Construction Data (Jan’26), in which we would see US Data Center Construction spending at $47.0 billion (seasonally adjusted), which is +31% y/y. Rolling 3-month Data Center Construction spend totals $138.2B, +30% y/y, accelerating from +27% in December—see note here. Exhibit 9 - US Data Center Spending; Construction Put in Place vs Starts (in Millions of Dollars) Source: US Census Bureau; Dodge Construction Network Wells Fargo Securities, LLC Technology and Services Team Outside of utility disclosures, recent semiconductor company commentary continues to validate the scale and visibility of the data center pipeline coming to life over the next several years. • Broadcom has framed AI demand explicitly in power terms, noting that across its six large LLM customers, expected deployments approach ~10 GW by 2027, reinforcing the importance of evaluating AI infrastructure build-out through a gigawatt lens rather than purely dollars. • NVIDIA’s commentary is similarly constructive, with management indicating that nearly 9 GW of Blackwell-based infrastructure has already been deployed and consumed over the prior five quarters. More importantly, NVIDIA now sees at least $1 trillion of cumulative, high- Equity Research | 13 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research confidence order visibility through 2027, implying roughly 30–35 GW of incremental infrastructure deployment over CY2025–2027 (with an estimated ~$800–$820 billion of orders remaining beyond what has already been recognized). • AMD commentary further supports this trajectory, with disclosed wins across its MI4xx platform exceeding 13 GW, which we estimate represents more than $225 billion of implied revenue opportunity. Appendix: North American Power, Utilities & Infrastructure IPP research Across management commentary, industry conversations, and recent NDRs, messaging has been consistently datacenter positive, with no indication of slowing demand and little evidence of a meaningful shift toward behindthemeter or offgrid solutions. Importantly, commentary has been pro- grid, not pro-BTM: hyperscalers continue to emphasize that their preference remains gridconnected power. As we have heard repeatedly, data center developers do not want to run power plants, they want to build data centers, leaving merchant generators and IPPs well positioned to serve incremental load growth. For our recent commentary from management meetings, please see: • SRE NDR - SRE NDR: The Comeback Kid Does It Again? Upsides to Upside and Catalysts for Rerating; Remains Top Idea, Raise TP $115 • ETR NDR – From NDR: Fastest Grower w/No End in Sight; ESAs and Reg Nukes In Mind; Upsides Evident but Not Tied to Analyst Day • DUK NDR - The Last of the Few Remaining 5-7% Growers? Too Big to Grow? Maybe Not Anymore as Mosaic Pieces Fall into Place For recent commentary on industry trends and other news, please see: • Meta Chips on the Table, Going All In on Lucky 7 (CCGTs) in LA; NOLA Poker Face Stuns to the Upside • TMI from CERAWeek? Context Is Key—Crane Comments Not an Issue for 2H27 Restart Date • Who Know’s More About Surprises Than Japan? US/JAP Deal Benefits NEE; Mild Optics of Shot Across the Bow for IPPs • So Now You Know… CEG in agreement to sell PJM assets to LS Power; More Ammo for Business Update • Hey Google, When Can We Raise the CAGR: Google + DTE new 2.7 GW Renewable Deal for 1GW Data Center Is One Step Closer to 8+% 14 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research Companies Mentioned in Report Company Name Ticker Last Price (03/27/26) American Electric Power Company, Inc. AEP $130.10 Centerpoint Energy, Inc. CNP $42.38 CMS Energy Corporation CMS $76.21 Constellation Energy Corporation CEG $301.49 Dominion Energy, Inc. D $60.88 Duke Energy Corp. DUK $129.99 Entergy Corp. ETR $109.88 NextEra Energy, Inc. NEE $91.40 NRG Energy, Inc. NRG $147.74 Public Service Enterprise Group Incorporated PEG $80.71 Sempra SRE $95.88 Talen Energy Corp. TLN $324.54 The Southern Company SO $95.55 Vistra Corp. VST $155.48   Source: Wells Fargo Securities LLC Estimates, FactSet Equity Research | 15 This document is for aweinberg@btig.com and should not be distributed further. Industry Update Equity Research Required Disclosures I, Shahriar (Shar) Pourreza, certify that: 1) All views expressed in this research report accurately reflect my personal views about any and all of the subject securities or issuers discussed; and 2) No part of my compensation was, is, or will be, directly or indirectly, related to the specific recommendations or views expressed by me in this research report. Wells Fargo Securities, LLC does not compensate its research analysts based on specific investment banking transactions. Wells Fargo Securities, LLC’s research analysts receive compensation that is based upon and impacted by the overall profitability and revenue of the firm, which includes, but is not limited to investment banking revenue. Additional Information Available Upon Request This is a compendium report, to view current important disclosures and other certain content related to the securities recommended in this publication, please go to Wells Fargo Disclosures or send an email to wellsfargoresearch@wellsfargo.com or a written request to Wells Fargo Securities Research Publications, 301 South College, 7th Floor, Charlotte, NC 28288. STOCK RATING OW=Overweight: Total return on stock expected to be 10%+ over the next 12 months. (BUY) EW=Equal Weight: Total return on stock expected to be -10% to +10% over the next 12 months. (HOLD) UW=Underweight: Total return on stock expected to lag the Overweight- and Equal Weight-rated stocks within the analyst's coverage universe over the next 12 months. (SELL) NR=Not Rated: The rating and price target has been removed due to lack of fundamental basis to support the recommendation or due to legal, regulatory or company policy considerations. FINRA regulation requires member firms to assign ratings to one of three rating categories: Buy, Hold and Sell. In accordance with FINRA regulation and solely to satisfy those disclosure requirements in the ratings distribution table and ratings history chart contained in these Required Disclosures, our rating of Overweight corresponds to a Buy rating; Equal Weight corresponds to a Hold rating; and Underweight corresponds to a Sell rating. As of March 29, 2026 53.5% of companies covered by Wells Fargo Securities, LLC Equity Research are rated Overweight. (BUY) 39.4% of companies covered by Wells Fargo Securities, LLC Equity Research are rated Equal Weight. (HOLD) 7.1% of companies covered by Wells Fargo Securities, LLC Equity Research are rated Underweight. (SELL) Wells Fargo Securities, LLC has provided investment banking services for 42.6% of its Equity Research Overweight-rated companies. (BUY) Wells Fargo Securities, LLC has provided investment banking services for 39.3% of its Equity Research Equal Weight-rated companies. (HOLD) Wells Fargo Securities, LLC has provided investment banking services for 39.7% of its Equity Research Underweight-rated companies. (SELL) Important Disclosure for U.S. Clients This report was prepared by Wells Fargo Securities Global Research Department (“WFS Research”) personnel associated with Wells Fargo Securities, LLC ("Wells Fargo Securities"). WFS Research may, from time to time, provide clients with short-term trading views in its research reports regarding subject companies on which Wells Fargo Securities currently has equity research coverage. A short-term trading view offers a view on how the market price of a subject company’s common equity may trend in absolute terms during the 30 days following the date of the short-term trading view. A short-term trading view on a subject company’s common equity does not impact our fundamental investment rating or price target for that company, which reflect our view of how the subject company’s common equity may perform over a one-year period. A short-term trading view may reach a different conclusion than the firm’s fundamental investment rating and price target for a subject company and, therefore, short-term trading views could result in short-term price movements that are contrary to our fundamental investment rating and price target. Short-term trading views are not ratings and the firm does not intend, nor undertakes any obligation, to maintain, update or close out short-term trading views. Short-term trading views may not be suitable for all investors and have not been tailored to individual investor circumstances and objectives, and investors should make their own independent decisions regarding any short-term trading views discussed in WFS Research reports. Important Disclosure for International Clients United Kingdom – The securities and related financial instruments described herein may not be eligible for sale in all jurisdictions or to certain categories of investors. For recipients in the United Kingdom, this report is distributed by Wells Fargo Securities International Limited (“WFSIL”). WFSIL is a UK incorporated investment firm authorised and regulated by the Financial Conduct Authority. For the purposes of Section 21 of the UK Financial Services and Markets Act 2000 (the “Act”), the content of this report has been approved by WFSIL, an authorised person under the Act. WFSIL does not deal with retail clients as defined in the Directive 2014/65/EU (“MiFID2”). The FCA rules made under the Financial Services and Markets Act 2000 for the protection of retail clients will therefore not apply, nor will the Financial Services Compensation Scheme be available. This report is not intended for, and should not be relied upon by, retail clients. For the purposes of the U.K. Financial Conduct Authority's rules, this report constitutes impartial investment research. EEA – The securities and related financial instruments described herein may not be eligible for sale in all jurisdictions or to certain categories of investors. For recipients in the EEA, this report is distributed by WFSIL or Wells Fargo Securities Europe S.A. (“WFSE”). WFSE is a French incorporated investment firm authorized and regulated by the Autorité de contrôle prudentiel et de résolution and the Autorité des marchés financiers. WFSE does not deal with retail clients as defined in the Directive 2014/65/EU (“MiFID2”). This report is not intended for, and should not be relied upon by, retail clients. 16 | Equity Research This document is for aweinberg@btig.com and should not be distributed further. IPP Bear Traps Ahead? Why We’re Not Taking the Bait: Answering Bear Arguments on the IPPs, and Reiterating Our Bullish Stance Equity Research Australia – Wells Fargo Securities, LLC and Wells Fargo Securities International Limited are exempt from the requirements to hold an Australian financial services license in respect of the financial services they provide to wholesale clients in Australia. Wells Fargo Securities, LLC is regulated under the laws of the United States and Wells Fargo Securities International Limited is regulated under laws of the United Kingdom. All such laws differ from Australian laws. Any offer or documentation provided to Australian recipients by Wells Fargo Securities, LLC or Wells Fargo Securities International Limited in the course of providing the financial services will be prepared in accordance with the laws of the United States or United Kingdom and not Australian laws. Canada – This report is distributed in Canada by Wells Fargo Securities Canada, Ltd., a registered investment dealer in Canada and member of the Canadian Investment Regulatory Organization (CIRO) and, Member – Canadian Investor Protection Fund (CIPF). Wells Fargo Securities, LLC’s research analysts may participate in company events such as site visits but are generally prohibited from accepting payment or reimbursement by the subject companies for associated expenses unless pre-authorized by members of Research Management. China – Strictly Private and Confidential. For the sole use of the recipient only. Not to be copied or redistributed within the People’s Republic of China. Brazil - This report was not created for distribution to investors resident in Brazil or to the Brazilian public in general. Wells Fargo Securities, LLC is a broker-dealer registered in United States of America with and regulated by the U.S. Securities and Exchange Commission. Wells Fargo Securities, LLC is not registered in Brazil and its products, including this report and the securities mentioned in this report, have not been and will not be publicly issued, placed, distributed, offered or negotiated in the Brazilian capital markets, and, as a result, have not been and will not be registered with the Brazilian Securities Commission (Comissão de Valores Mobiliários, the CVM). The offer of Wells Fargo Securities, LLC's products, including this report and any securities mentioned in this report, is intended only for residents in the countries in which Wells Fargo Securities, LLC is authorized to operate. About Wells Fargo Securities Wells Fargo Securities is the trade name for the capital markets and investment banking services of Wells Fargo & Company and its subsidiaries, including but not limited to Wells Fargo Securities, LLC, a U.S. broker-dealer registered with the U.S. Securities and Exchange Commission and a member of NYSE, FINRA, NFA and SIPC, Wells Fargo Prime Services, LLC, a member of FINRA, NFA and SIPC, Wells Fargo Securities Canada, Ltd., a member of IIROC and CIPF, Wells Fargo Bank, N.A. and Wells Fargo Securities International Limited, authorized and regulated by the Financial Conduct Authority. This report is for your information only and is not an offer to sell, or a solicitation of an offer to buy, the securities or instruments named or described in the report. This report, including any ratings it contains, does not constitute a personal recommendation or take into account the particular investment objectives, financial situations, or needs of individual clients. Clients should seek professional advice, including tax advice, to determine whether any advice or recommendation in the attached research report is suitable for their particular circumstances. The information in this report is provided as of the date of the report and has been obtained or derived from sources believed by WFS Research to be reliable, but WFS Research does not represent that this information is accurate or complete. Any opinions or estimates contained in this report represent the judgment of WFS Research, at the time that the report was published, and are subject to change without notice. All Wells Fargo Securities research reports published by WFS Research are disseminated and available to all clients simultaneously through electronic publication to our internal client websites. Clients may also receive our research via third party vendors. Not all research content is redistributed to our clients or available to third- party aggregators, nor is WFS Research responsible for the redistribution of our research by third-party aggregators. Equity Strategists focus on investment themes across the equity markets and sectors. Any discussion within an Equity Strategy report of specific securities is not intended to provide a fundamental analysis of any individual company described therein. The information provided in Equity Strategy reports is subject to change without notice, and investors should not expect continuing information or additional reports relating to any security described therein. Clients of WFS Research are permitted to store, display, analyze, modify, reformat, copy, duplicate and reproduce this report and the information contained within it for their own internal use and for no other purpose. Without the prior written consent of Wells Fargo Securities, no part of this report may be copied, duplicated or reproduced in any form by any other means. In addition, this report and its contents may not be redistributed or transmitted to any other party in whole or in part, directly or indirectly, including by means of any AI Technologies (defined below) through which this report or any portion thereof may be accessible by any third-party. “AI Technologies” means any deep learning, machine learning, and other artificial intelligence technologies, including without limitation any and all (a) proprietary algorithms, software, or systems that make use of or employ neural networks, statistical learning algorithms (such as linear and logistic regression, support vector machines, random forests or k-means clustering) or reinforcement learning, or curated data sets accessible by any of the foregoing or (b) proprietary embodied artificial intelligence and related hardware or equipment. In addition, certain text, images, graphics, screenshots and audio or video clips included in this report are protected by copyright law and owned by Wells Fargo Securities, its affiliates or one or more third parties (collectively, “Protected Content”). Protected Content is made available to clients by Wells Fargo under license or otherwise in accordance with applicable law. Any use or publication of Protected Content included in this report for purposes other than fair use requires permission from Wells Fargo Securities or, in the case of content attributed to any third party, the third-party copyright owner. You may not alter, obscure, or remove any copyright, trademark or any other notices attached to or contained within this report. These Conditions of Use are not intended to, and will not, transfer or grant any rights in or to the report or the information contained within it other than those rights that are specifically described herein. All rights not expressly granted herein are reserved by Wells Fargo Securities or the third-party providers from whom Wells Fargo Securities has obtained the applicable information. Any external website links included in this publication are not maintained, controlled or operated by Wells Fargo Securities. Wells Fargo Securities does not provide the products and services on these websites and the views expressed on these websites do not necessarily represent those of Wells Fargo Securities. Please review the applicable privacy and security policies and terms and conditions for the website you are visiting. Copyright © 2026 Wells Fargo Securities, LLC SECURITIES: NOT FDIC-INSURED - MAY LOSE VALUE - NO BANK GUARANTEE Equity Research | 17 This document is for aweinberg@btig.com and should not be distributed further.