Five months from initial “purchase” to “neutral”: Mizuho downgraded two house ratings, privatization expectations rapidly cooled

The Zhitong Finance App learned that Mizuho Securities recently downgraded the ratings of FNMA.US (FNMA.US) and Freddie Mac (FMCC.US) from “buy” to “neutral”, and that their target prices have both dropped: Fannie Mae was lowered from $10 to $5, and Freddie Mac was lowered from $9 to $4. At this point, only five months have passed since Mizuho first covered and gave a “buy” rating.

Analyst Dan Dolev admitted in the research report that the probability that the two houses will be privatized by 2028 is declining. He cut the probability of Fannie Mae's privatization from 30% to 18%, and Freddie Mac from 20% to 9%. This shift echoes the operation of the “big short” Michael Burry — the latter has sold two common shares and switched to preferred shares.

Mizuho: Probability of privatization lowered drastically

When Mizuho first covered two houses in May, the core logic was that the market seriously underestimated the probability of an IPO before 2028. The “quick exit” plan set by Dolev at the time included a 2.5% capital requirement, repayment terms for government preferred shares, and accelerated capital accumulation by increasing guarantee fees (g-fees).

However, after five months, all three assumptions were loosened.

Dolev listed three reasons in the latest research report: First, rising mortgage interest rates make privatization more complicated. If the two houses raise the g-fee in order to withdraw from government control, it will directly drive up mortgage costs. In the current political environment where housing affordability is sensitive, the viability is extremely low. Second, US Federal Housing Finance Administration (FHFA) Director Bill Pulte recently stated that the two houses will buy more mortgage-backed securities (MBS). Although this move will help lower interest rates in the short term, it will consume capital. Third, VantageScore joined the FICO Classic system and re-introduced the “double merger” credit scoring model, increasing the complexity of credit risk management.

Dolev pointed out that the first obstacle would bring political complexity to privatization, while the latter two would slow the pace of capital accumulation. He expects that the two houses will not make up for losses by increasing the g-fee, that capital accumulation will be slower, and that the schedule for full capitalization will move further back.

Burry sells common stock to buy preferred stock

Burry revealed in a Substack post last weekend that he had emptied two houses of common stock and instead bought multiple series of sub-preferred shares. The core judgment is that the US Treasury holds $388 billion in senior preferred shares. If converted to common shares, existing common shareholders face dilution of at least 90%.

But Burry didn't underestimate the two rooms themselves. He pointed out that the current market value of Fannie Mae's sub-preferred shares is about 6 billion US dollars, with a face value of 19 billion US dollars, and the transaction price is about one-quarter of face value. If dividends are restored or converted at face value, the expected return is about 3 to 5 times. In the capital structure, the dividend distribution and settlement order of preferred shares take precedence over common shares. Even if the Ministry of Finance transfers shares, preferred shares will perform better than common shares.

In other words, Burry did not give up the privatization option for the two houses, but instead adjusted the position level. Common stocks bear the greatest risk of policy uncertainty, while preferred stocks provide downside protection.

Policy Contradiction: Buying MBS Conflicts with Capital Accumulation

Pulte's policy orientation is another key background for Mizuho's downgrade.

In January 2026, Trump signed an executive order authorizing the purchase of up to 200 billion US dollars of MBS for two houses. Pulte then raised each company's holding limit from $40 billion to $225 billion. Pulte called the move “an immediate reduction in mortgage interest rates,” and Freddie Mac Weekly showed that mortgage interest rates did fall by about 10 basis points to a three-year low of 6.06%.

But the immediate consequence of expanding MBS purchases is capital consumption. As of August 2026, Fannie Mae's retirement portfolio had shrunk from $173.4 billion in July to $172.6 billion, according to KBW data. The two houses are using capital in exchange for falling short-term interest rates. This is the opposite direction of capital accumulation required for privatization.

Is Pulte's priority housing affordability or privatization? Judging from the action, the former clearly has the upper hand. And if capital accumulation continues to slow, the privatization schedule will only move further backwards.

G-fee dilemma and valuation pressure

PIMCO's previous analysis indicated that simply announcing “no reduction in g-fees” can reduce mortgage interest rates by 5 to 10 basis points; however, if G-fees are cut across the board, it may push up interest rates because it will increase the risk of early repayment, and MBS investors will demand higher yield compensation.

This has left the two houses in a dilemma: raising the g-fee is politically unworkable, lowering the g-fee is not financially cost-effective, and maintaining the status quo is slow to accumulate capital. Dolev's judgment — “losses won't be covered by increasing g-fees” — actually acknowledged this predicament.

In terms of valuation, KBW has lowered the target price for two houses to $3.50 per share due to the combination of declining privatization probability and dilution risk. Burry warned that Fannie Mae might fall to around $3 to find support. As of early October, Fannie Mae and Freddie Mac's stock prices were reported at $4.04 and $3.68 respectively, both falling by more than 60% during the year.

Mizuho's rating shift, Burry's position adjustment, and Pulte's MBS purchase plan all point to a reality: the privatization of two houses is degenerating from “when will it happen” to “whether it will happen again.” Capital accumulation is a slow variable, and the political window is narrowing. KBW analyst Bose George wrote in a September report: “The timeline is tight because the government's attention will shift to the presidential election starting in 2028.” Until then, the two houses are more likely to remain in the “news desert,” awaiting a ruling in the next political cycle.