Midas79

Wow that is delicious! I take back everything, well most, of what i said about him! I'll enjoy watching that replay :-) Start at 1:42:40 for the exchange. Calabria also defends the capital rule. https://www.banking.senate.gov/hearings/oversight-of-housing-regulators There's also some about the senior prefs from Warner at 1:59:00. Warner wants Calabria to commit to not wiping out the seniors with no compensation, but Calabria wisely and correctly said that that decision is not his to make but instead Treasury's.

I have to admit, this was delicious. Calabria isn't being swaying by Zandi's BS. Here's the quote: "I would disagree with [Zandi's] analysis and certainly note that Mr. Zandi is on the board of a mortgage insurer that has a strong economic interest in not seeing us do this rule, so I certainly would not put him forward as an unbiased expert in this." https://twitter.com/midas79_/status/1270381139673432065

Well done. I especially like the implication that FHFA's mandates extend to making sure the recap actually happens rather than paying lip service to it.

FnF's capital has nothing to do with the market price of the commons. There is also no guarantee of any price appreciation after said events, let alone enough to get the price above $4 for 30 days to re-list. For all the good news FnF shareholders got in the last 16 months, the price is around 40% lower than it was at the end of January 2019. A reverse split is the only realistic way to get an over-$4 share price to stick, and I fully expect it to happen later this year. Somewhere between 1:10 and 1:20 is my guess for now. Moelis 2.0 is completely unrealistic anyway. Why would junior pref shareholders sign up for a plan that involves them converting at a re-IPO price in the teens (a far worse ratio than they could get right now in the open market), and giving existing common holders a much greater return than the junior prefs themselves?

Here is the rest of the post from Tim Howard's blog; he just kept the first 4 points for brevity. I also had forgotten to finish point 5 at the time. 5) I didn't realize that Freddie had so many more off-balance sheet assets than Fannie: $300M worth. Will this have any effect on the capital standards or raises? 6) Table 4 on page 28 is a mess. The numbers $25B, $42B, $209B, and $122B should be $27B, $32B, $145B, and $101B respectively. The first two are in Fannie Mae's 2017 10-K, and the latter two come from Table 3b on page 236 (I multiplied Adjusted Total Assets of $2.524.6T by 4% to get the $101B). Freddie's table (Table 5 on page 30) has similar problems. 7) Normally I would chalk this up to typos and human error, but it actually undermines Calabria's point that his capital standard would have been enough to avoid the peak "losses" that Fannie and Freddie would have sustained after the crisis in 2008. However, his simplistic method of just adding net worth, equity issuance in 2007, and total draws from Treasury ignores that a significant chunk of the Treasury draws were circular, i.e. only taken to pay Treasury its 10% dividend on the seniors. Backing that out might well cancel out the effect of accidentally overstating the capital requirements in Tables 4 and 5 on pages 28 and 30. 8 ) I think Calabria's mistake (or whoever drafted the report) was that they somehow added the PCCBA twice to the Total Capital Requirement in Tables 3a and 3b on pages 235-236. That gives the exact (erroneous) number of $209B for Fannie, and gives $124B for Freddie (close to the $128B shown in Table 5 on page 30).

I posted a bunch of thoughts after my first read-through of the fact sheet and (what I believe to be) the relevant parts of the full capital rule for us shareholders on Tim Howard's blog. While he does occasionally delete posts, and mine is rather verbose, it is entirely aprpros and should stay up. https://howardonmortgagefinance.com/2020/05/05/first-quarter-takeaways/comment-page-1/#comment-15295 I'll post a link to that on Twitter too, but probably not until I'm sure that the post will stay up.

The $19B of junior par value is already included in the $14B. That is, everything else that contributes to the $14B adds up to a combined negative $5B. Also, the refund from Treasury could come in the form of a tax credit. In that case, I believe that Fannie (and Freddie too, of course) would only add their tax savings to capital as they occur. It could very well take 5-6 years to earn all of that refund. The upshot is that if a capital raise is needed to hit a particular number then the full amount of the refund won't count, only the taxes they would have paid. At best I can see Fannie getting $2.5B of it back per year, and that would be on earnings of around $13B. Also, I think Fannie will make more than $6B this year ($10B imo) but for now I'll use your number. If the capital raise happens a year from now, adjusting your numbers for these things leaves a gap of $56B: your $27B plus the $19B of prefs and $10B of unearned tax credits. Another thing to keep in mind is that $81B is Fannie's risk-based capital requirement. For some reason I can't explain, Calabria decided on a minimum capital standard (he calls it the Leverage Capital Requirement) that's higher than the risk-based one: $89B. Now the gap is $64B if that's the milestone. Similar calculations give a gap of $42B between Freddie's current core capital (once the seniors are dealt with) and their minimum capital requirement of $63B: Freddie has $9B of current equity, $7B of earnings between now and the capital raise, and $5B of earned tax credits. So I'm getting core capital shortfalls of $64B for Fannie and $42B for Freddie, or $106B combined. This would have to be the size of the capital raise if it occurs a year from now and is designed to hit the minimum capital standard. Treasury returning the overpayments in cash instead of tax credits would lower each company's numbers by $10B, for a total capital raise size of $86B.

Technically, that's not what I asked. I previously did some calculations on FnF's balance sheets in their 10-K forms, and the numbers they report for core capital were precisely the sum of all equity-section balance sheet entries other than senior pref stock and AOCI. When (if?) FnF get a tax credit from Treasury, they will book it in the asset section of the balance sheet. The offsetting entry has to go in the equity section somewhere. My question is: will it go to a line item that counts towards core capital (tier 1) and/or CET1? If the offsetting entry is in retained earnings then the answer to both is yes, while if it's AOCI (or some new item that they don't count towards core capital) then it's no.

I don't mean anything negative or personal by this, but this sentence made me laugh really hard. The performer of the experiment was an unwitting subject all along!

Footnote 63 on page 101 says "An Enterprise’s “prescribed buffer amount” means, as applicable, its PCCBA or its PLBA." The last two tables on page 10 of the fact sheet show that the PCCBA is $99B and the PLBA is $91B (both numbers are for FnF combined). That puts Table 8 into a whole new light. FnF can pay out up to 20% of its eligible retained income (page 100: "The maximum payout ratio is the percent of eligible retained income that an Enterprise can pay out in the form of distributions and discretionary bonus payments during the current calendar quarter.") when it has between 25% and 50% of those buffers above, which is roughly $25-50B of core capital (if I understand the definition in footnote 62 correctly). 20% of $17B in annual income is $3.4B, which is enough to pay the full $2B per year on the juniors and have $1.4B left over for common dividends and executive bonuses. FnF will already have more than $25B in core capital once the seniors are gone. This whole thing about "no dividends until $175B in common equity" is way, way off. you are right, it's not all or nothing until 175bn. but it's important to look at the leverage ratio also. the minimum leverage ratio excluding buffers is 152bn in tier1 capital. if you assume 25-50bn for preferred, that's around ~115bn common equity. throw on 25bn from the initial 25% buffer = ~140bn common equity before any dividends can be paid. still a lot. No. This isn't even close. I just did the math: FnF can pay distributions of up to 20% of "eligible retained income" when their "capital buffer" hits 25% of $91B or $99B, depending on which of PCCBA and PLBA is used. The payout table doesn't directly depend on common equity at all. I don't know why you keep bringing that up when page 101 is quite clear. I still need to fully understand the meaning of the quoted phrases, but it points to FnF being able to pay dividends at $23-25B of capital. Not $140B of common equity. Edit: I might be calculating this wrong, if the buffer in footnote 62 is meant to be on top of the total capital standard, rather than standalone. If so, the $140B number could be right, but it wouldn't have to all be common equity. Edit 2: Turns out I was wrong. The "capital buffer" in Table 8 really is on top of the other capital requirements, meaning that FnF would need around $160B in capital before making any distributions ($135B total capital plus 25% of $99B). However, a section at the bottom of page 102 says that these restrictions are not set in stone. Divs can turn on at any point Calabria chooses, even if it technically violates the rules in Table 8.

Footnote 63 on page 101 says "An Enterprise’s “prescribed buffer amount” means, as applicable, its PCCBA or its PLBA." The last two tables on page 10 of the fact sheet show that the PCCBA is $99B and the PLBA is $91B (both numbers are for FnF combined). That puts Table 8 into a whole new light. FnF can pay out up to 20% of its eligible retained income (page 100: "The maximum payout ratio is the percent of eligible retained income that an Enterprise can pay out in the form of distributions and discretionary bonus payments during the current calendar quarter.") when it has between 25% and 50% of those buffers above, which is roughly $25-50B of core capital (if I understand the definition in footnote 62 correctly). 20% of $17B in annual income is $3.4B, which is enough to pay the full $2B per year on the juniors and have $1.4B left over for common dividends and executive bonuses. FnF will already have more than $25B in core capital once the seniors are gone. This whole thing about "no dividends until $175B in common equity" is way, way off.

When? This never happened. Calabria mentioned Treasury potentially forgiving some of its investment, but he was clearly talking about the seniors. Treasury forgiving some or all of the warrants doesn't help recap FnF at all. The companies are worth "X"bn. Taking away one of the consumers of that value leaves extra portions for the existing common + Potential jr converters + new equity. Plus no one wants a govt overhang of more secondary sales after the primary ones. That's not how capital works at all. Core capital is common equity plus non-cumulative pref equity plus additional paid-in capital plus retained earnings. You're talking about market cap. Treasury will exercise the warrants before the re-IPO. That removes any concerns about holdings and timing.

When? This never happened. Calabria mentioned Treasury potentially forgiving some of its investment, but he was clearly talking about the seniors. Treasury forgiving some or all of the warrants doesn't help recap FnF at all.

David Metzner, who I believe has reason to potentially know this, seems to think that it could be much sooner. As in this Friday. See this tweet thread: https://twitter.com/HoldenWalker99/status/1262561563489140736

@Wiggins Thanks for the explanation. It makes perfect sense. Yes, Citi's conversion took the market completely by surprise: the pref:common ratio in the exchange offer was more than 3 times that of the day before the announcement was made. Being caught off guard, i.e. with a large common position, when a FnF share exchange hits could be devastating compared to just owning all prefs. In fact, that's essentially the reason I only own the prefs most of the time, and never go below 90% (vis a vis the commons). I wrote a post about Citi's conversion last year, it might answer some of your other questions if you haven't read it yet. https://www.cornerofberkshireandfairfax.ca/forum/general-discussion/fnma-and-fmcc-preferreds-in-search-of-the-elusive-10-bagger/msg382592/#msg382592 @cherzeca I see your point that Washington Federal's claims are a very different animal than those made by basically all of the other cases. I just wonder if Sweeney will paint Washington Federal's claims with the same brush, whether it's out of laziness or that she really does think it's the proper thing to do. I think you have a much better idea than I do about the similarities and differences between the cases. Which part appears to have been written before Collins and not revised since? The Collins decision was in September, while Sweeney's Fairholme decision was in December. So if she failed to update something as a result of Collins, she did so in both the Fairholme and Arrowood cases. It also means she had written her opinion before September, right? 3+ months would be a long time to sit on that.

https://ecf.cofc.uscourts.gov/doc1/01513723106 crappy argument and crappy opinion. given Fairholme and collins, this is no great loss. it is a shame Sweeney was spending her time on this and not on Washington federal. What's crappy about it? At least, is there anything crappy about it that wasn't crappy in the Fairholme opinion? It's mostly a copy-and-paste outside Section V (Standing). I do agree that it's no big loss. She dismissed Arrowood's direct claims for basically the same reason she dismissed Fairholme's direct claims. I would expect her to dismiss all other similar direct claims (breaches of fiduciary duty and implied-in-fact contract). I also think that the Washington Federal case is in grave danger of being dismissed in its entirety. If Sweeney decides that she lacks jurisdiction over any direct claims, doesn't that case just go poof like Arrowood?

@Wiggins That's a great story and strategy, I'm glad you shared it. What I meant by trading the commons and prefs, though, was selling one class to buy the other at what I hope to be advantageous times while being cash-neutral. Basically betting on the direction of the FNMAS:FNMA ratio. It's approaching 5:1, which is around the highest it has been since late 2018. https://stockcharts.com/h-sc/ui?s=FNMAS:FNMA&p=d&yr=2&mn=0&dy=0&id=p6911660031c It seems that the commons are always the first to overreact to good news, so even for a 100%-pref investor like me, it might be worth swapping a bit over to commons to potentially gain on good news like a capital rule actually being released or a positive Supreme Court decision in Selia. The chart I linked to shows two big plunges (the kind that swapping prefs to commons would take advantage of): Watt's term ending and the Fifth Circuit en banc decision. The real money to be made 18 months ago, in hindsight of course, was to swap prefs to commons in late 2018 and swap back in March 2019. Edit: and if you don't mind my asking, why didn't you do the same thing at the end of 2019 as you did in 2018? @beaufort I am 99% sure that the caps' relation to junior pref par value is a coincidence. I just found it interesting and decided to post it.

@Wiggins I have tried to time the pref/common back-and-forth trade in the past, and I have just about broken even. I suppose I shouldn't complain, but it tells me that I'm no good at it. Of course, that doesn't stop me from continuing to try. Right now the commons are underperforming enough that I'm tempted to take another spin at the wheel. Are you still doing this trade, and are you on the common side? @beaufort No, I don't see a redemption happening at all. Exchanging those shares for commons accomplishes the same goal (clearing out the capital structure for the re-IPO commons) while costing nothing instead of dozens of billions. I agree that the juniors will want to be in commons post-release due to the fantastic profitability of the business. I have a different view on the caps. I think they were put in place to put FHFA and Treasury on the clock; they need to get the PSPA amendment done before the caps are hit or the NWS restarts, which is major egg on Calabria's face due to his insistence on FnF building capital. The caps should be hit late this year, and a PSPA amendment finally killing the NWS at that time coincides with the timelines of ACG Analytics and other investors. I think the cap levels relative to the par value of the juniors is a coincidence, but I don't feel the same way about those levels as a function of FnF's earnings.

I posted this elsewhere but I'll do so here too to see if anyone has ideas or theories about this. Last September, FHFA and Treasury agreed to lift the caps above which FnF have to pay all net worth to Treasury, from $3B for each company before that date, to $25B for Fannie and $20B for Freddie. https://www.fhfa.gov/Conservatorship/Documents/Senior-Preferred-Stock-Agree/FNM/LTR-Agree/FNM-2019-Ltr-Agreement_09-27-2019.pdf https://www.fhfa.gov/Media/PublicAffairs/PublicAffairsDocuments/9-27-19_FRE-Capital-Agreement.pdf At the time I couldn't figure out why they used those numbers. Freddie is around 2/3 the size of Fannie, but its cap is 4/5 as much. That's enough of a difference to make one pause. Also, Fannie's total par value of its junior prefs is $19.130B, and Freddie's is $14.109B. Now I have a theory: the fact that the difference between the letter agreement cap for each company and its total junior pref par value is almost identical is no coincidence. Said difference is $5.870B for Fannie and $5.991B for Freddie; the $121M difference between those amounts to a rounding error given the round $25B and $20B numbers in the letter agreements. Of course, I don't really know what this could mean. It could very well just be coincidence.

@Wiggins Citi's pref conversion was an offer made to all shareholders, of which most accepted. If the same happens with FnF, no 2/3 vote will be necessary because each shareholder would be free to convert or not. However, if the re-IPO investors really do want the capital structure flattened before investing, that gives FHFA urgency to get all of them converted. I think the easiest (and cheapest) way to do that is just offer more and more commons until everyone accepts. If enough do, the rest (outside of FNMAS and FMCKJ) can be redeemed. Redemption on a large scale, though, makes no sense at all, even if it's financed with more prefs. Re-IPO investors that want to buy commons want there to be as few prefs as possible in front of them in the capital structure. Conversion to common accomplishes the same goal while not costing any money. As far as whether div rates will matter in the conversion, I can see it going either way. A holder of FNMAT will scoff at getting the same percentage of par as FNMAO, but if they get a good enough deal in absolute terms it might not matter. To prepare for this, I choose to own the middle of the div structure: I don't own any FNMAT/FNMAS/FMCKJ, but I also am staying away from the low-div floaters because the 4-6% fixed-div series aren't much more expensive. My main holding is FNMAM (same price as other Fannie 50s with a higher div rate), and I have started picking up chunks of FMCKK at what I perceive to be good prices. FNMAG, FMCKL, and FNMAH are overpriced compared to the rest of the prefs in my opinion. So are FNMAT/FNMAS/FMCKJ if things happen the way I think they will. I do agree that FNMAS should gain a good chunk of value if it still exists on 1/1/2021.

Doesn't liquidity only matter if you want to get in or out quickly? If so, what event(s) would make you want to run to the exit?

I used to think the same thing, but I've changed my opinion. I previously owned FMCCL but switched to the more expensive (based on percentage of par) FNMAH about 6 months ago. Currently, I own FNMAH and FNMFN in roughly equal dollar amounts. I'm not necessarily saying any of the following are going to happen, but the real possibility is enough to at least think twice (especially points 1 and 2 below). My reasons for changing my opinion are below... (1) If nothing was done other than turning on the dividends for lower dividend prefs, like FMCCL with it's 1.54% coupon, where do you think they'd trade in relation to par? Let's say the yield is 4%, that puts the price of FMCCL at $20 (40% of par). That would be awful if an investor is expecting at or near par. (2) There has been talk of converting at 60% of par plus back-interest paid for 6 years. Yes, I'm aware they are non-cumulative but anything can happen in a negotiated settlement. FMCCL (1.54%): ($50 * 60%) + (1.54% * $50 * 6 years) = $34.62/share... or 69.2% of par. FNMAH (4.5%): ($25 * 60%) + (4.5% * $25 * 6 years) = $21.75/share... or 87% of par. FNMFN (7%): ($50 * 60%) + (7% * $50 * 6 years) = $51.00/share... or 102% of par. (3) Fannie is expected to exit conservatorship via consent decree up to 6 months before Freddie exits because they have more capital built up. Therefore, people might be more willing to buy Fannie prefs. (4) Some investors don't want to switch back and forth within the preferred issues due to long-term capital gains tax status. It's a tough pill to swallow to start that year-long clock over again. Good points, particularly 2 and 4. That said, why would you own FNMAH (variable div, 4.5% right now) over FNMAJ (7.675% fixed div) right now when you could trade the former for the latter at parity? I just checked: the bid on FNMAH and the ask on FNMAJ are both at $6.15.

As cherzeca said not sure why someone who was willing to contribute to 100 million dollar fund wouldn't just buy the preferred on their own and save the management fee. That being said 2/20 maybe a small price to pay if your unfamiliar with the story/possible outcomes and probabilities and don't want to put the time in to learn it all on your own. If things go the way we think they will that an easy 2/20 for them. I don't see a meaningful threshold in required learning between buying prefs oneself or investing in a fund that's going to buy them. Just a strange story all around. If it ends up lacking credibility it could actually be a pump-and-dump. I assume that this fund will buy mostly if not all FNMAS and FMCKJ (maybe FNMAT) because the other series just don't have the liquidity to deploy $100M in any reasonable timeframe, and the other series stand to gain more than 200% if they hit par. Even FNMAS and FMCKJ will make more than 200% in a return to par, but the 2/20 might account for it coming out at 200% in the end. Monday was a good day for the prefs but nowhere close to $100M in volume. Only around $11M for FNMAS and $4M for FMCKJ. Either that's when this fund decided to deploy some of their money (a rumor I saw said that they had already raised $20M) or someone else decided to pile into FNMAS.

There is nothing in HERA, that I am aware of, that sets out the conditions for release from conservatorship. While some, like ACG Analytics, have speculated that FnF will be released when they hit the statutory minimum capital requirement (which I believe was around $41B combined), I don't see anything that holds Calabria to that. In fact, Calabria said "It has always been my view that an exit from conservatorship is going to require a large capital raise by Fannie and Freddie." a month ago. To me, that signals release at or near the full minimum capital standard. Watt's two proposed alternatives for that were $103.5B and $139.5B, and I would be shocked if Calabria's number comes in lower than that range. I would be much less surprised if it's higher, even though that would hamper release by requiring a $100B capital raise all at once. Tim Howard said "I think both FHFA and Treasury will be more focused on where Fannie and Freddie’s core capital is versus the new minimum capital requirement in deciding on the timing of their release from conservatorship", and I completely agree.