Comparing Treasury's IRR on GFC Bailouts of Banks/AIG to Treasury's IRR on GSEs Investment, on Assumption that the GSE Senior Preferred Stock is Cancelled in a GSE Recap/ReleaseEven should the GSE senior preferred stock be cancelled as economically repaid, the investment return for taxpayers is much greater than that obtained from bailing out the money center banks and AIG.
Rule Of Law Guy - Jan 5
Suppose Trump 47 asks Treasury Secretary Bessent whether the taxpayers’ return on Treasury’s investment in GSEs’ Senior Preferred Stock (SPS) and Warrants is comparable to the taxpayers’ return on Treasury’s investments in money center banks and AIG during the Great Financial Crisis of 2008-2009, on the assumption that Treasury cancels the SPS as (more than) fully repaid in a GSE recap/release.
What would Bessent’s reply be?
One would assume Bessent would be well-prepared and pull out a spreadsheet in order to answer this question.
Interestingly, as far as I can tell, there has been no public disclosure by Treasury as to what Treasury’s internal rate of return (IRR) was on those GFC investments.
Bessent’s spreadsheet would necessarily have to be prepared bespoke to answer this question.
I posed the question as to what was Treasury’s GFC IRR to an AI agent. Here is the reply:
“The US Treasury’s TARP investments in AIG resulted in a positive return of $5 billion on $67.8 billion disbursed, over a period of roughly 4 years from late 2008 to late 2012. Using approximate cash flow timing (investments in 2008-2009 and repayments/sales primarily in 2011-2012), this equates to an estimated IRR of around 1.8-2%.
For the money center banks (Bank of America, Citigroup, JPMorgan Chase, Wells Fargo), a 2024 study in the Journal of Finance (“Did Banks Pay Fair Returns to Taxpayers on TARP?”) calculated an annualized return of 11% to taxpayers on TARP investments (primarily through the Capital Purchase Program or CPP), compared to a market benchmark of 39% for similar preferred equity.
This IRR is the effective compounded rate based on investments, repayments, dividends, and warrant proceeds. The large banks generally repaid within 1-1.5 years, leading to higher IRRs in some cases (e.g., 15.3% for Bank of America’s $20 billion TIP investment, and similar for Citigroup’s TIP based on comparable terms and timing).
Smaller institutions that held TARP funds longer pulled the overall average down to 11%.”
What would be Treasury’s IRR for taxpayers from a GSE recap/release, on the assumption that the SPS is cancelled as (more than) fully repaid?
I answered this question in GSE Recap/Release: Considering Treasury's Incentives, What Is a "Good Price" for "Government's Stake" and Keefe Bruyette's GSE Common Stock Valuation Is Half of Deutsche Bank's Valuation, where I wrote (in the context of Bessent’s statement that Treasury will seek a “good price” for Treasury’s GSE common stock in a GSE recap/release):
Treasury’s investment in Fannie Mae (“government’s stake”) began in 2008 with its $121 billion SPS investment, for which Treasury received, in addition to the SPS, warrants to purchase 79.9% of Fannie Mae’s common stock for a de minimus exercise price.
This “government stake” generates a total return, including payments made in respect of the SPS in the past and payments Treasury will receive in connection with the sale of its common stock. Economically, it is nonsensical to ignore past payments made in respect of the SPS.
If Treasury looks to this total return, this would require Treasury to recognize that its SPS has been economically fully repaid (and more) in accordance with the SPS’s original terms. These payments would include the notorious net worth sweep instituted by the Obama administration (Obama NWS) in 2012, which produced a 10% internal rate of return (IRR) to Treasury in the case of Fannie Mae (the 10% Moment) by 2018, and eventually a 11.9% IRR until the time when SPS dividends were halted in 2020 (resulting in Treasury’s receipt from Fannie Mae of $181 billion of “dividends” on a total draw of $121 billion of funds under the SPS line).
This would mean that in determining a “good price” arising from Treasury’s “government stake” in Fannie Mae, Treasury’s return on its common stock ownership interest to be sold in a GSE recap/release should not be viewed in isolation, but taken together with its 11.9% return on, and the complete repayment of, Treasury’s SPS from Fannie Mae.
Let’s calculate an all-in IRR for Treasury’s SPS investment of $121 billion in Fannie Mae that includes Treasury’s 79.9% warrants interest to see if it is getting a “good price”.
Assume that Fannie Mae’s common stock valuation is approximately $220 billion at the time of the GSE recap/release ($17 billion annual net income multiplied by a 14X P/E ratio, less $19 billion of outstanding junior preferred stock (JPS)). Treasury’s 79.9% common stock stake would be valued at $176 billion.
Treasury has already received a 11.9% IRR in respect of its SPS through 2020. Treasury’s valuation of its 79.9% common stock in 2025 would add a 2.3% IRR to Treasury’s return in respect of its SPS since 2008, resulting in a total IRR of 14.2%.
So, Treasury’s all-in IRR in a GSE recap/release from its GSE investment where the SPS is cancelled as (more than) fully repaid far exceeds Treasury’s IRR from its AIG investment, and substantially exceeds Treasury’s IRR from its various bank investments.
Do you have the spreadsheet handy, Treasury Secretary Bessent?
Market Data 
Markets 
AI
