Fall 2025 Edition

It could become the largest IPO in history. It could raise $30 billion in new equity. And according to the Trump administration, it could "restore" Fannie Mae and Freddie Mac to the private market after 17 years in federal conservatorship. Behind this bold language lies a concerning reality: the push to take the government-backed mortgage giants public is being driven more by political optics than any semblance of financial logic.

In the years coming out of the Great Depression, 25% of the nation's outstanding mortgage debt was in default. In an effort to support affordable housing, Fannie Mae was formed as a part of FDR's New Deal. Its initial goal was to support local banks by providing federal money to finance home loans. Banks could only lend as much as they held in deposits; they were incredibly restricted, and rates were high. Fannie Mae would buy loans from these restricted lenders and provide immediate cash. Fannie pooled mortgages to guarantee payment of principal and interest to large investors who were not traditional mortgage investors. Lenders could then sell their loans and use that money to issue more mortgages, which drastically lowered rates and increased mortgage availability.

Fannie Mae's ownership changed in a multitude of ways in the decades after its founding. Most notably, Fannie was already privatized once before in 1968 to remove its debt from the federal budget, however they were put under conservatorship after the 2008 crisis.

Freddie Mac was formed in 1970 as a government-chartered private company in response to Fannie Mae's monopoly status. Freddie was formed to make the market more competitive, so they share a nearly identical function, and was similarly put under conservatorship. Like Fannie, Freddie buys and securitizes loans, which frees lenders up to issue more loans. The major difference is that Freddie works with substantially smaller lenders, often credit unions or community banks.

Currently, Fannie and Freddie are government sponsored enterprises, or GSEs, a hybrid of being both private and owned by the government. They are publicly traded, but the government acts similar to a holding company that exerts massive influence in the finances and operations of the two companies. This would be the most significant public offering of a government-controlled enterprise in modern U.S. history.

What the Proposed IPO Would Actually Be — And Why You Should Care

While a government-backed secondary market that originated in the 1930s may seem incredibly irrelevant to us graduating college nearly a hundred years later, this history explains why we are precisely the demographic that will be most impacted by changes to Fannie and Freddie. According to the Federal Reserve Bank of New York, "as of June 2025, loans securitized by government-sponsored enterprises (GSEs) like Fannie Mae and Freddie Mac comprise around 52 percent of all balances of outstanding mortgage debt at roughly $6.5 trillion." And of that, 58% of all agency loan purchases come from first-time home buyers.

As many graduates plan to move to major cities where there are already rental crises, the support Fannie and Freddie give to multifamily affordability and rental conditions become increasingly relevant. Changes in the government-sponsored enterprises will impact the next generation of renters and homeowners immensely, as instability in the mortgage market could raise borrowing costs for first-time homebuyers, make it harder to qualify for loans, and impact rental affordability. Far from restoring stability, a rushed or politically engineered IPO could raise mortgage rates, weaken investor demand, and undermine confidence in the institutions that guarantee America's $12-trillion mortgage market. This proposal is not meaningful reform. This potential IPO is a political stunt dressed as capital markets policy.

IPO Optics Against Market Reality

The administration's pitch is that the IPO could be the biggest of all time, relieve the government of $30 billion in debt, and transform one of the world's most important secondary markets. How can all three be true at once? Who actually benefits? If executed poorly, as the current process strongly suggests, the lending environment could easily become more fragile, not less.

In a typical IPO, the government would work in the background while experts model risk, determine capital needs, and craft a structure that supports long-term stability. This process has been unusually public, personality-driven, and chaotic. CEOs of major banks have reportedly made repeated visits to the White House, where they are accused of tailoring pitches directly to the president and incorporating celebrity cameos. Meanwhile, Fannie Mae has experienced senior leadership turnover, internal controversies, and high-profile firings.

The message to markets is clear. It is a political process first, financial second. That matters because Fannie and Freddie underpin roughly 60% of all U.S. mortgages. Their role is essential as they provide liquidity by purchasing loans from lenders and converting them into securities. For this system to function, investors need predictability. They need clarity around capital requirements, guarantee fee setting, and the independence of risk pricing from politics.

The current spectacle offers the opposite. It signals that short-term political goals may drive decisions on structure, valuation, and governance. When markets detect political influence creeping into financial decision-making, they demand higher returns to compensate. In housing finance, that translates directly into higher mortgage rates.

Why an IPO is not Feasible

The biggest obstacle to a clean IPO is also the issue the White House discusses the least: capital. Under existing regulatory requirements, Fannie and Freddie require on the order of $150 to 300 billion in capital to exit conservatorship under current regulatory standards. According to the Wall Street Journal, "Trump officials have been envisioning IPOs that value the combined firms at roughly $500 billion and raise roughly $30 billion." A $30 billion equity sale, while a historical amount, would barely dent that gap.

Without adequate capital, the firms cannot responsibly return to private ownership. Instead, the administration risks creating a worst-case scenario. The current government-sponsored entities would become public enough to face shareholder scrutiny but still government-controlled enough to remain vulnerable to political pressure. This hybrid structure would be unstable and unattractive to long-term private investors.

The Overlooked Risk: The TBA Market

Compounding the risk, the conversation around a potential IPO has almost completely ignored the anchor of American housing finance: the TBA (To-Be-Announced) market for GSE mortgage-backed securities.

The TBA market is the second most liquid bond market in the United States, behind only Treasuries. In 2024 alone, an average of $345 billion in GSE mortgage-backed securities traded every day. It attracts trillions of dollars in global capital to the U.S. housing system and ensures that families across the nation have access to 30-year, fixed-rate mortgages at stable, relatively affordable rates.

There is no international equivalent; the TBA market is a uniquely American innovation that delivers unique benefits. The fact that the U.S. can even support a 30 year fixed rate mortgage is a result of the secondary market these GSEs provide — outside the U.S. longer term fixed rate mortgages are virtually unheard of. The TBA market's stability rests on a single condition: investors must believe that GSE mortgage-backed securities will remain liquid, predictable, and insulated from political turmoil. If a chaotic, politicized IPO leads investors to demand a premium for governance uncertainty, the shock would reverberate through the entire housing system. Mortgage rates would rise, liquidity would thin, and regional disparities in credit access would widen — which would undo decades of progress toward a unified national mortgage market. The administration's IPO rhetoric ignores all of this.

The Case for Privatization and Its Limits

An interesting argument in favour of privatization can be formed by making a parallel to the reasons why Fannie Mae went private in the 70s. If the GSEs are taken off the government's balance sheet, not only would the taxpayer be off the hook for insuring any financial troubles, but the government would raise a significant amount of money. There is also a sentiment that appeals to creative destruction — thrusting them into the public market will force them to innovate and provide new mortgage structures without the restrictions of the government. However, this could lead to a 2008-like scenario where, to keep pace with private competitors, they began taking on increasingly risky mortgages. Each time these GSEs have strayed from their core mission of making mortgages more accessible, there have been horrific consequences for the American homebuyer.

Alternative Reform Pathways

The IPO is often framed as a benefit for taxpayers, but the actual distribution of benefits tells a different story. While the administration could claim a policy victory and investment banks can earn enormous underwriting fees, there still remains the rest of the US. Borrowers could face higher mortgage costs if political risk premiums rise. Investors would buy shares in undercapitalized and politically influenced firms. Taxpayers could once again be responsible if instability triggers future losses — the same cycle that led to the 2008 bailout. Housing finance is too central to the U.S. economy to be reshaped by a rushed transaction.

Reforming Fannie and Freddie is not only possible, it is necessary. But the sequencing matters. A responsible path would establish clear, enforceable capital requirements and allow time to meet them. They would create a predictable, apolitical framework for setting guarantee fees and explicitly protect TBA market liquidity. Introduce private capital ahead of taxpayers, but in realistic, sustainable amounts.

Only once these foundations are in place should policymakers consider an IPO. Anything earlier risks systemic stability for political theatrics.

Housing finance rarely captures national attention, but its stability affects everything from mortgage rates to the broader economy. While the administration may be drawn to the optics of a massive equity sale, the risks to market stability and household borrowing costs are real. In the political economy of housing finance, careful design of such an IPO must come before spectacle. Spectacle is winning and the country's housing system may pay the price.

In less than a decade, sports betting in the United States has transformed from an underground pastime into a multibillion-dollar financial industry embedded in everyday consumer life. This shift began in 2018, when the Supreme Court overturned the Professional and Amateur Sports Protection Act, allowing states to legalize and regulate sports wagering.

Since then, sportsbooks have adopted market-based pricing, real-time data modeling, and aggressive consumer incentives that closely resemble those used in traditional financial markets. However, this resemblance masks a critical imbalance. Unlike those markets, legalized sports betting has expanded with far fewer regulatory safeguards, exposing consumers to financial risk without the protections typically afforded to investors. A legal system that operates like a financial market should be regulated like one.

One of the strongest arguments in favor of legalization is its impact on the illegal sports betting market. According to the American Gaming Association, "As legal sports betting has grown, the illegal market has shrunk dramatically — from $150 billion pre-legalization to around $64 billion today." Clearly, sports betting has already been successfully integrated into the United State's formal economy. However, legalization alone does not ensure consumer protection.

A huge part of the growth of sportsbetting is founded in the convenience of online sportsbooks. In fact StatNews reports more than 90 percent of wagers are now placed online rather than in physical sportsbooks and yet, this participation is still growing. According to Business of Apps, "sports betting apps in the United States brought in $13.7 billion in revenue in 2024, a 25.6% increase from the previous year." With a single click, bets are placed instantaneously, allowing continuous participation with minimal friction.

These books such as DraftKings, FanDuel, BetMGM and others function like market makers in equity and foreign exchange markets, providing continuous liquidity in the market. With just a quick google search, these sports books are always ready to trade, executing bets almost instantaneously, even during periods of "low trading activity." In fact, Sports books' primary purpose is to manage exposure.

To describe this system in another perspective: sports-bettors speculate on fluctuating prices, respond to information shocks, exhibit identifiable behavioral biases, and compete in a marketplace structured by institutional players who manage liquidity and risk. Yet this market uniquely operates with minimal regulatory oversight and without the investor protections built into traditional financial infrastructure.

Examining the Counterarguments

This characterization of sports betting as a financial market can be criticized, specifically when arguing that key economic features distinguish wagering from investment. A sports bet has no intrinsic value, does not generate future cash flows, and cannot be held as a productive asset. From this perspective, a sports bet is a zero-sum transaction, not a market. Others suggest that sportsbooks do not fully mirror market makers because they do not always neutralize risks through balanced books — instead sports books can occasionally assume directional exposure, especially in low liquidity environments.

These arguments do not negate that these sports books act in an increasingly similar way to traditional markets. The lack of intrinsic value in these sports bets should not halt the comparison. Options and futures are both derivative products that trade on regulated exchanges. Additionally, the occasional directional exposure assumed by sportsbooks does not undermine the fact that they continuously set prices and adjust spreads in ways that are very similar to market makers. The fact that sports bets lack intrinsic value does not mean they should escape regulation. Financial markets already regulate speculative products that carry no intrinsic value because they still expose participants to real financial harm.

Additionally the idea that sports betting is written off because it is legally classified as "entertainment" is an oversight. This industry is not merely entertainment. This argument undermines the speculative environment and true economic consequences of sports betting.

Behavioral Risk and Structural Exploitation

Behavioral economists have long shown that individuals systematically overestimate their ability to outperform markets and display strong loss aversion, leading to increasingly risky behavior following losses. These same dynamics are evident in sports betting. Despite the evidence that the "house always wins," sports bettors still wager their money, attempting to redeem their losses by making riskier, larger bets.

In fact, the market actually exposes participants to higher levels of behavioral risk. These sports books will strategically implement promotions and parlay "multipliers" in order to generate more revenue. According to an experimental study on effects of inducements on sports gambling and decision-errors, "self-reported survey data also indicate[s] that inducements increase gambling participation and spend as well as risky behaviours, such as impulsive in-play betting." With these tactics, sports bettors increasingly act as inexperienced traders during unpredictable periods.

When sportsbooks profit from predictable human biases, losses can no longer be explained as purely the result of individual choice. In markets where firms actively encourage risky behavior, regulation exists to limit harm, not to remove personal responsibility.

Market Structures and Liquidity Parallels

Sports betting also shows differences between high-liquidity and low liquidity assets. Games like NFL playoffs, Baseball World Series, March Madness basketball, and more attract more significant betting volume and in turn exhibit highly efficient pricing. This is comparable to large cap equities.

In contrast, smaller conferences, lower-division sports, and lesser known players do not attract large volumes of bets and thus exhibit significant inefficiencies. In these bets, opportunities still exist, but there is more price volatility and less information surrounding these bets — and therefore comparable to emerging markets and small-cap stocks.

Speculation and gambling are synonymous depending who is defining the pair. Speculation involves taking calculated risks, and many may argue that is the intent of gambling on sports. Looking at statistics and probability, sports betting can be incredibly calculated — not just based on chance and luck.

Bets on long term outcomes such as MVP winners or world champions function similarly to derivatives. These bets are low probability and prices fluctuate depending on conditions like performance and injuries. Meanwhile, with in-game bets or live betting, odds are recalculated after every play in real time response, similar to high-frequency trading in equity markets.

Sports betting operates as a dynamic financial marketplace that features speculation, market making, arbitrage, liquidity management, and robust behavioral components. Yet it lacks the regulatory guardrails that protect participants in traditional financial systems. Treating sports betting more like a financial market would not eliminate risk, but it could introduce clearer disclosures, limits on promotional practices, and stronger consumer protections.

While financial markets are subject to suitability requirements, disclosure standards, and consumer protections, sports betting operates largely outside these frameworks. As millions of Americans enter this market daily, policymakers and analysts face an urgent imperative: should a system that mirrors Wall Street so closely continue to operate without comparable oversight?

Progressive lawmakers like Zohran Mamdani argue that raising taxes on high earners will close budget gaps and strengthen social programs. But recent IRS and Census migration data suggest that the mobility of wealthy taxpayers is far more relevant than policymakers assume. In a world of remote work, mobile businesses, and widening interstate tax gaps, a small number of departing high earners can cause disproportionate revenue losses. This article examines the elasticity of high-income taxpayers and argues that beyond a certain threshold, aggressive progressive taxation backfires — reducing state revenues instead of expanding them.

Why This Matters for College Students

Graduates deciding where to live and work are entering a landscape shaped by interstate tax competition. States with unstable revenue bases often cut funding for public universities, financial aid programs, and early-career professional services — while fast-growing low-tax states expand them. Remote work has widened your menu of choices, but it's also amplified states' need to retain high earners. Understanding how tax mobility affects state budgets helps explain why opportunities flourish in some regions and contract in others.

Are High Earners "Sticky?"

Supporters of higher top-bracket taxes argue that wealthy residents are deeply rooted in their communities. They point to entrenched industries, cultural amenities, family obligations, and strong professional networks as anchors that reduce geographic mobility. From this lens, incremental tax hikes merely tap into an abundant and reliable revenue source. Because high earners historically did not migrate at high rates, the assumption is that they lack both incentive and ability to relocate.

Yet the conditions that once kept high earners geographically anchored have changed. Remote work, industry decentralization, and expanding tax arbitrage opportunities — the restructuring of income in a way that legally reduces one's tax burden — have weakened the traditional constraints on mobility. What was once a safe assumption — that the wealthy cannot or will not leave — no longer holds in the same way.

Migration Data Tells a Different Story

IRS Statistics of Income (SOI) migration data reveals persistent outflows of high-income households from high-tax states like New York and California. Between 2013 and 2022, both states experienced steady losses of residents earning over $200,000, while Florida and Texas saw strong net inflows among the same group. In 2022, New York lost roughly $14.1 billion in adjusted gross income to other states, while California lost approximately $23 billion.

Critics often argue that weather, housing costs, or lifestyle changes drive migration more than taxes. While these factors matter, they do not fully explain the selective exit of top earners. The data show that higher-income households account for a disproportionate share of out-migration from high-tax jurisdictions. This pattern indicates that tax differentials play an important role, particularly when alternative states offer comparable urban amenities or business environments.

Further, Florida and Texas' reliance on sales and property taxes, rather than income, creates more stable revenues that are less vulnerable to shifts among top earners. The success of these states reflects structural tax design that reduces sensitivity to migration. High-tax states ignore these structural differences at their own peril.

The fiscal consequences are significant because the top 1% contribute an outsized share of state tax revenue. In New York, roughly 46% of income tax revenue comes from the top 1%. In California, the figure is close to 49%. When even a small portion of this group relocates, the resulting loss materially affects state budgets.

What About States With High Taxes and Strong Economies?

Some argue that high-tax states like New York or California remain economic powerhouses precisely because they invest heavily in public services, infrastructure, and innovation ecosystems. From this view, raising taxes on the rich strengthens long-term competitiveness by funding education, transit, and research. These states, the argument goes, offer unique value that low-tax states cannot replicate — reducing the incentive for high earners to leave.

This perspective has merit, but it assumes that the value proposition remains strong enough to offset rising tax differentials. The current migration data suggests that this balance is shifting, and that the economic "premium" of living in high-tax states is no longer compensating for the widening fiscal gap.

Elasticity at the Top

The economic concept of tax elasticity measures how sensitive taxpayers are to changes in tax rates. Recent studies challenge the assumption that high earners have low mobility. Research from the Stanford Institute for Economic Policy Research (2022) finds that millionaires do migrate in response to higher taxes, particularly when rate differentials widen significantly. Federal Reserve research (Kitao and Sahin, 2020) similarly reports moderate but meaningful elasticity among top-income earners.

Because high earners represent such a large share of tax receipts, even minimal elasticity carries major fiscal implications. Policymakers who ignore elasticity risk basing budgets on projections that collapse when mobility increases.

Progressive taxation works only when taxpayers have limited alternatives. In an economy where high earners and businesses can relocate with minimal friction, ignoring tax-driven mobility carries significant fiscal risk. The evidence shows that once tax rates exceed competitive thresholds, the elasticity of high-income taxpayers becomes large enough to shrink the revenue base. For states that rely heavily on their wealthiest residents, sustainable fiscal policy requires respecting these constraints rather than dismissing them. If policymakers ignore mobility, they risk undermining the very social programs they seek to expand.

There seems to be a contradiction in the U.S. economy: restaurant operators complain about falling foot traffic and cautious customers, yet aggregate consumer spending and the stock market remain strong. The answer to that puzzle sits in two places: the drive-thru and the trading screen. Lower-income Americans are quietly pulling back on fast food, while higher income, stock owning households keep spending thanks to a prolonged, big tech-driven bull market. The question for the QSR sector isn't just if traffic will rebound — it's what happens if the market tailwind disappears.

The "Recession" That Doesn't Show Up in GDP

Recent earnings calls from big QSR names paint a consistent picture of weak traffic, especially among budget-conscious customers. In the first quarter of 2025, McDonald's reported a nearly double digit drop in traffic from low and middle income consumers. It also experienced its largest U.S. same-store sales drop since 2020 and singled out low and middle-income customers as the main source of the decline.

Industry data back this up. The National Restaurant Association has recorded net declines in customer traffic for eight consecutive months, even as many operators still report positive same-store sales because of higher prices. This consistent traffic decline is one of the most concrete signs that a full blown recession should be occurring within the sector, but it isn't.

This weakness is easy to miss — consumer spending remains quite strong, and GDP isn't signaling recession. But at the level of a $10 combo meal, the downturn is obvious. For many low-income households, fast food has crossed from "cheap treat" into "luxury." This kind of thinking gives an idea as to why consumer sentiment is at an all time low, but the market is seemingly not reacting accordingly.

A Bull Market Props Up the Top Half

On the other side of the split, higher-income households are looking resilient. Studies from the Boston Fed and McKinsey show that spending growth has been much more resilient among consumers earning over six figures, who are less likely to cut back and more likely to maintain or upgrade their purchases.

The stock market is playing a big role here. Big tech and related giants now represent roughly a third of the S&P 500 on their own, and nearly half if you include adjacent mega-caps. That concentration means that strong performance in a narrow slice of tech can keep the entire index hitting new highs, boosting the wealth of the households most likely to own equities.

The result: a gigantic yet risky wealth effect. Wealthier consumers, buoyed by portfolio gains and relatively strong wage growth, continue to eat out, travel, and spend. Recent reporting even suggests that for some chains, traffic from higher-income guests is growing nearly as fast as it's falling among low-income diners. This produces a misleadingly stable average.

What If the Music Stops?

If QSR is in a recession at the bottom while being propped up by a tech-driven wealth effect at the top, a sharp correction in tech stocks is not just a stock problem — it's also a fast-food problem.

If the energy crisis is not resolved and big tech momentum stalls, the households currently offsetting low-income weakness could start cutting back as well. The same tech wobble that drags the S&P down could also drag QSR traffic down from the "comfortable" tier of consumers that have been dining out freely.

That doesn't guarantee a collapse in fast-food demand as people still need convenient meals, but it does remove a hidden stabilizer. QSR chains might be forced to lean harder into value menus, promotional pricing, and cost cutting, pressuring margins and profitability.

Two Futures for Fast Food

If the current environment holds, QSR chains can wait out the uncertainty driving down consumer sentiment and causing low income consumers to stay at home. Same-store sales may inch higher as companies lean on pricing, loyalty apps, and menu tweaks, quietly shifting toward higher-margin items and more targeted promotions. Growth will slow, but the model more or less works as management teams refine strategies and ride the market tailwind.

If that tailwind fades and the tech rally stops, the adjustment becomes more painful. With less support from wealth and confidence effects, QSR brands would have to buy back their customers with real value: sharper discounts, simpler menus, and tighter cost controls. Franchisees might delay remodels, scale back expansion plans, or close weaker stores altogether. In both cases, the sector adapts — either gradually through strategic tweaks, or abruptly through margin compression. But these outcomes remain closely tied to how long the current political and economic uncertainty, and the tech trade beneath it, can last.

Traditional U.S. banks are retreating from high-risk lending, constrained by post-2008 regulations and volatile markets. Into the void have stepped private credit funds — nonbank lenders like Blackstone Credit, Apollo Global Management, and Ares Management — fueling a $2 trillion private credit boom. These shadow lenders now finance everything from leveraged buyouts to middle-market acquisitions, raising urgent questions about whether credit risk is quietly migrating into a largely unregulated, opaque financial system. This shift does not simply represent innovation; it may signal a dangerous repeat of the shadow banking buildup that helped trigger the 2008 financial crisis.

Banks Pull Back

In the wake of the financial crisis, banks faced tougher capital rules and oversight. This made financing speculative or highly leveraged deals more expensive and less attractive. Rising rates and volatile markets since 2022 have added to the caution. When banks couldn't offload $80 billion in buyout loans in 2022, many reduced exposure further.

As a result, syndicated loan issuance declined 15% year-on-year in early 2025. Former Goldman Sachs executive Mike Koester noted that volatility makes it harder for banks to price and place syndicated deals. This risk aversion opened the door for private lenders.

Private Credit's Rapid Rise

Private credit refers to loans made outside the traditional banking system. These direct loans bypass bond markets and often involve just a few funds. Private credit assets under management have grown from a few hundred billion in 2010 to over $1 trillion today, rivaling the high-yield and leveraged loan markets. Growth surged after 2019 as low rates pushed investors toward higher-yield alternatives.

Major players like Apollo and Ares now manage credit portfolios worth hundreds of billions. Apollo's credit business alone jumped from $75 billion in 2018 to $268 billion in 2023. BlackRock expects global assets in the sector to reach $3.5 trillion by 2028.

Private funds can execute deals quickly, especially during market stress. In 2025, U.S.-China tariff tensions disrupted syndicated markets. Banks paused, but private lenders stepped in. Their flexibility and speed are giving them an edge.

Mega-Deals Go Private

Direct lenders now finance billion-dollar transactions once handled by banks:

• Zendesk (2022): $10 billion LBO financed by a consortium of direct lenders.
• Jeppesen (2023): Blackstone and Apollo provided $4 billion in debt.
• Vista Equity (2023): Issued $4.8 billion in private credit for a software acquisition.
• Noosa Yogurt (2025): A $200 million unitranche loan replaced a Citigroup-led bank loan.

Private credit financed 60 percent of U.S. leveraged buyouts in 2023, up from just 7 percent five years earlier. While banks have regained footing in calmer conditions, direct lenders have permanently shifted the credit landscape.

Risks in the Shadows

Despite its benefits, the growth of private credit brings a set of complex and potentially destabilizing risks. Chief among them is opacity. Unlike public markets, private credit operates behind closed doors. These loans are not rated by credit agencies or regularly traded, and their valuations are based on internal models rather than market prices. This can delay the recognition of losses and mask the true volatility of credit portfolios.

Another concern is the steady weakening of underwriting standards. Borrowers in the private credit space often carry higher debt loads and benefit from looser or even non-existent financial covenants. These light protections reduce lender control and bear unsettling similarities to the covenant-lite structures that proliferated ahead of the 2008 crisis.

The creditworthiness of borrowers is also a point of vulnerability. Many firms turning to private credit are smaller, highly leveraged, and lack public ratings. According to the IMF, more than a third of these borrowers now have interest expenses that exceed their earnings — an unsustainable position in a high-rate environment.

Liquidity mismatch further amplifies the danger. While most private credit funds lock in investor capital for years, a growing number are offering semi-liquid structures, particularly to retail investors. If redemptions spike during a downturn, these funds may be forced to sell assets under pressure, magnifying market stress.

Additionally, the notion that banks are insulated from this shift is misleading. Many provide credit lines to private funds, invest in them, or structure their deals. Federal Reserve data suggests banks have roughly $79 billion in such exposures. Should private credit funds falter, the effects would likely reverberate across the regulated banking sector.

Finally, regulatory oversight has not kept pace. Private credit funds face no capital requirements, stress testing, or consistent disclosure mandates. While the SEC has proposed modest reforms, oversight remains piecemeal. Without stronger frameworks, risks could accumulate undetected until they spill over into broader markets.

As one investor recently remarked, shadow lenders now write the biggest checks in corporate America. Their growing role has turned them into kingmakers, but with little public accountability.

A Crisis Repeating in Slow Motion?

This expansion of private credit mirrors the structure and logic of the pre-2008 shadow banking system: financial risk pushed into the unregulated corners of the market, where return-seeking behavior overwhelms prudence. Though modern private credit funds differ in funding structure and asset type, their explosive growth, lack of transparency, and weak oversight are all red flags.

Jamie Dimon and IMF officials have voiced alarm. Investors believe they are protected by structure and sophistication, but the last crisis taught that correlation, leverage, and poor underwriting can unravel with speed. The assumption that these funds will quietly absorb losses may prove overly optimistic. Unlike banks, they are not required to carry capital buffers or submit to stress tests. That makes the system more fragile, not less.

As one advisor put it, this isn't innovation. It's déjà vu from 2008, only bigger.

Private credit is not merely replacing banks in certain lending markets. It is recreating the same systemic blind spots that led to the last financial crisis, but on a much larger scale. Regulators must not mistake complexity for resilience. Without robust oversight and greater transparency, private credit is poised to become the next fault line in global finance.

The proliferation of AI training infrastructure has proven to place significant demand pressure across North American electricity markets. Goldman Sachs projects AI-specific compute will add 160 TWh of incremental demand by 2030, roughly equal to Florida's total annual consumption. NVIDIA's largest customers have executed multi Giga Watt power supply agreements, positioning technology firms as active energy market participants rather than passive consumers.

Geographical Constraint

Unlike previous data center buildouts that sought cheap hydropower in the Pacific Northwest, AI infrastructure is clustered around tech hubs — regions where power grids are already strained. GPU clusters need reliable and low cost electricity in specific locations, not just anywhere power is available. This is creating severe pressure on regional grids. In Texas (ERCOT nodal system), electricity price spreads in areas with heavy GPU concentration have widened more than 50% year-over-year. In the Mid Atlantic (PJM), the queue of projects waiting to connect to the grid has reached 270 GW with average wait times exceeding five years.

The core problem is timing, as major grid upgrades in the US take 7 to 10 years from planning to completion, while AI companies are expanding on quarterly cycles. Demand is growing faster than the infrastructure can respond, creating a structural supply deficit.

Difference with AI Demand

Traditional industrial users reduce electricity consumption when prices spike. Factories slow production, as per basic economics. On the other hand, AI operators cannot. For them, shutting down GPU clusters to save on electricity costs means delayed model training and competitive disadvantage. The opportunity cost of interruption is greater than the power bill savings for these firms. This changes how electricity gets priced; AI firms will pay prices that would not be economical for regular firms since margins on computers are worth the costs. Federal regulators are even questioning whether AI infrastructure should get priority in grid connection queues, which would further dynamize the shift in North American power markets.

This results in a feedback loop: high AI profitability (as well as a lot of hype from stock speculators and investors) justifies extreme electricity prices which then disrupts pricing models.

Trading Opportunities Emerging from Grid Stress

Three instruments are becoming particularly relevant:

Congestion Rights: as price differences widen between power generation areas and consumption zones, traders can profit from betting on transmission bottlenecks.

Virtual Power Agreements: AI firms are using virtuals not solely for renewable energy credits, but as hedges against price spikes in regions where they operate data centers. Meta has recently created a new power trading subsidiary that likely will aim to save money from hedging.

Real-time volatility: increase in bottlenecks means greater uncertainty and volatility, thus creating opportunities for traders who can forecast these deviations. Hedge funds are already positioning for this shift.

Unlike traditional electricity demand forecasting based on weather and industries, AI buildout follows VC investing cycles and semiconductor supply. Traders who can anticipate where AI companies will expand next, relative to grid constraints, can extract returns from location-based price differences — underlining a new change in how power markets are traded due to the proliferation of AI.

The Long-Term Structural Shift

The winners will be those who can anticipate where AI infrastructure will expand before it becomes public knowledge and understand which parts of the grid will face the worst bottlenecks. Traders should more and more treat power as a commodity tied to AI deployment rather than a regulated utility service.

Unlike semiconductors where supply shortages can eventually be solved with enough capital and time, power grid constraints face regulatory, environmental, and physical limits that money alone cannot quickly overcome — such as battery storage and renewable sources only generating power in certain conditions.

The power grid, and not the chip supply chain, may be the ultimate constraint determining which companies dominate the AI race. For power traders and infrastructure investors, the opportunity is structural rather than cyclical. AI demand will continue to increase price volatility, expand geographic price differences, and create a market where understanding grid infrastructure and location becomes a source of persistent advantage.

Price: $257.87 · Market Cap: $420B · Sector: Semiconductors / AI Infrastructure

Investment View

Advanced Micro Devices (AMD) continues to expand its role in data-center AI infrastructure through an open-ecosystem strategy that integrates silicon, systems, and software. With MI350 ramping, ROCm 7 maturing, and multi-OEM adoption broadening, the key question shifts from participation to durability — and whether AMD can sustain share and margin momentum amid rising competitive intensity.

Product and Platform Momentum

Advanced Micro Devices (AMD) designs high-performance CPUs and data-center GPUs for cloud and enterprise AI Workloads. The Instinct lineup has progressed from MI300 to MI350, which is now supported by the ROCm 7 software stack, an open platform for both training and inference that has demonstrated significant performance gains compared to ROCm 6, addressing prior concerns about software maturity. Azure's (MSFT) ND MI300X v5 virtual machines deploy eight MI300X GPUs per node, which support fine-tuning and high-throughput inference at production scale. AMD and Azure jointly provide detailed provisioning driver guidance, which reduces the friction for infrastructure teams and signals readiness for large-scale deployment.

Supermicro introduced H14 systems that feature Instinct MI350 GPUs, which broaden its OEM base beyond a single hyperscaler. The availability of both air-cooled and liquid-cooled options for MI350 systems demonstrates flexibility across various data center configurations, thereby shortening deployment timelines.

Building Toward Full-Stack Integration

AMD's strategy now extends beyond chip design to full-stack systems. AMD acquired Enosemi, which brings in-house photonics and co-packaged optics expertise, critical for managing power, latency, and interconnect costs as clusters scale. AMD has stated that Enosemi's technology will be quickly incorporated into next-generation AI systems, reinforcing its rack-scale ambitions.

Financial Performance and Capital Allocation

Financials show momentum alongside policy noise. In Q2 FY25, AMD reported record revenue of $7.7 billion, led by strength in the data center and client segments. The earnings deck detailed export-control-related inventory and charge effects, along with non-GAAP adjustments, which helped investors model the margin trajectory through the product transition.

Street consensus for 2H 2025 assumes continued data center growth as MI350 volume ramps and ROCm 7 adoption deepens. The focus now lies on sustained shipments and software progress, rather than one-off announcements. As OEMs ship H14 platforms and Azure expands regional availability, the question shifts from participation to the durability of share and margin.

Capital returns provide a cushion: In May 2025, AMD approved a $6 billion share-repurchase program, increasing total capacity to approximately $10 billion and offering flexibility as AI investments accelerate.

Investment Thesis

AMD's open-ecosystem approach can sustain demand beyond a single cloud anchor. Azure validation reduces perceived risk for hyperscalers seeking a second GPU supplier, while ROCm 7 advancements narrow the software gap with NVIDIA, improving developer retention and workload portability. The Enosemi acquisition strengthens AMD's long-term path toward rack-scale optical systems, aligning with the broader industry shift toward photonics.

Key Near-Term Catalysts

• Incremental MI350 design wins and third-party benchmark results
• Additional OEM platform launches and Azure regional availability
• Earnings commentary on AI mix and data center margin trends

Risks

• Execution on ROCm 7 maturity and ecosystem adoption
• Competitive pressure from rival GPU roadmaps
• Policy or export-control constraints affecting shipment timing and gross-margin mix

Bottom Line

AMD's open stack platform now has tangible production footprints. The investment debate transitions from proof of concept to proof of consistency. If MI350 deployments expand across clouds and OEMs, while ROCm and optical integration progress as planned, Street expectations could rise. Conversely, delays in execution or policy headwinds could leave limited room for multiple expansions.

Oracle appeared unstoppable, or so the research thesis held when work on this article began in September — an excellent month for Oracle. The stock of the revered database giant, founded 48 years ago to sell "relational" databases, had surged to an all-time high on September 10, buoyed by blockbuster optimism. For a brief period, co-founder Larry Ellison, a legendary and enduring figure in Silicon Valley, overtook Elon Musk as the richest person in the world. The company's $300 billion, five-year commitment from OpenAI and its astounding remaining performance obligations — now up 438% year-over-year to $523 billion — were often cited as exemplary of a once-in-generation AI infrastructure land grab.

That optimistic narrative has collapsed. Oracle's stock fell 10.83% on December 11 and slipped another 4.66% on December 12 after its December 10 Q2 Earnings Release, extending a drawdown that has left it down more than 30% from its September 10 peak. This sudden reversal offers a sobering case study in the speed at which market euphoria can transform into skepticism when financial fundamentals fail to keep pace with ambitious visions. Now, Oracle is a bellwether for investor sentiment towards AI.

The AI Investment Thesis and Its Cracks

The AI investment thesis appeared to be supported by Oracle's recent growth metrics. It still has many positive figures to report. In fiscal Q2 2026, Oracle reported total revenue of $16.1 billion, representing 14% year-over-year growth, only slightly below consensus expectations and a much higher than expected $455 billion contract backlog. On top of that, the company's cloud business remained a bright spot: Cloud Revenue was $8 billion (up 34%) and Cloud Infrastructure Revenue was $4.1 billion (up 68%).

However, warning indicators are appearing beneath this upbeat exterior. Credit default swaps — which are essentially derivatives that offer insurance against the risk of a bond issuer not paying creditors — started to increase gradually. Oracle is now amongst the top 20 most traded corporate credit default swaps, which is unusual for a non-financial company, and it now costs three times more to insure against the risk of Oracle defaulting than it does Microsoft.

The Debt Burden

A fundamental conflict is reflected in the discrepancy between Oracle's credit risk and stock performance: the company is leaning heavily on unprecedented borrowing to fund its accelerating AI aspirations. Oracle's debt has grown by $11bn to $88bn and its total debt to operating profit (before depreciation and amortization) is 4.2, substantially higher than the traditionally safe range of zero to three. On top of this, Oracle reported negative $10 billion in free cash flow for fiscal Q2 2026 (almost twice StreetAccount's consensus for Q2), and management also increased the full-year capital expenditure forecast to $50 billion.

In credit markets, this played out sharply: Bloomberg reporting described Oracle's investment-grade notes as trading "more like junk," with investors in the $18 billion high-grade note September bond sale sitting on paper losses of roughly $1.35 billion. All major ratings agencies cite Oracle's high debt-to-EBITDA ratio (higher than 400% per Moody's) as reasons for its credit downgrading — all three major rating agencies have shared their concerns about Oracle's elevated risk, and Morgan Stanley has recommended dumping Oracle's bonds.

OpenAI Dependency and Execution Risk

Furthermore, Oracle may be overexposed to a customer currently in the glaring spotlight over profitability concerns. OpenAI has agreed to purchase $300 billion of computing capacity from Oracle over five years, accounting for over half of Oracle's $500 billion in pledged revenue. To service these payments, OpenAI must significantly increase its own revenue. This is a future under pressure as Google, Anthropic, and open-source alternatives gain ground at the same time as OpenAI is in the spotlight for mounting worries of an unpredictable AI bubble — its software products have yet to prove they can generate returns that justify the extraordinary investment levels we are seeing. This event has sent shockwaves through the market with AI-related stocks like Nvidia, AMD, Coreweave, and Micron seeing sympathy drops.

Execution timing risk has also joined the narrative. Oracle already has limited experience building physical infrastructure and was intending to get data center complexes online years ahead of typical industry timelines. Already, Bloomberg has reported that, according to people close to the work, some US data centers Oracle is developing for OpenAI have had completion dates pushed from 2027 to 2028 due to labor and material shortages (Oracle disputed that characterization in follow-on coverage). It is becoming clearer that Oracle's enormous infrastructure investments will materially precede actual revenue conversion.

Structural Disadvantages vs. Hyperscalers

The differences between Oracle and its hyperscaler competitors highlight the company's particular weaknesses. In the words of The Economist, "its larger rivals look nothing like a utility." Even though they also make significant investments in AI infrastructure, Amazon, Microsoft, and Google Cloud have more diverse revenue streams and better credit ratings. Oracle's lower credit rating poses an acute problem at a time when capital requirements are skyrocketing, since it must contend with financing costs that are higher than those of its rivals.

The 126 basis points annually to insure Oracle's debt — much higher than competitors — are the most striking example of how the market's initial enthusiasm for this strategy is giving way to skepticism.

From this it is clear that under outward signals of extreme strength can be concealed underlying weaknesses that beget acknowledgement. In the case of Oracle, that acknowledgement has arrived swiftly. The ongoing episode has become a live barometer of whether AI infrastructure will translate into durable, profitable growth, and whether the narrative surrounding AI infrastructure can be extended.

Oracle serves as a warning to investors about the dangers of continuously extending current trends, with its downfall from market darling occurring rapidly. Database management systems and enterprise software, the company's core business, continue to generate significant operating cash flow. However, Oracle lacks the capital needed for the bold move to become an AI infrastructure hyperscaler, which has led to a debt-fueled gamble on ongoing demand from OpenAI. The credit markets — where investors demand 126 basis points annually to insure Oracle's debt — are the most striking example of how the market's initial enthusiasm for this strategy is giving way to skepticism.