[BidClub_]
Sohn Conference Foundation · · 10 分钟

Victor Bonilla 在 Sohn 2025 推介 Main Street Capital

Victor Bonilla

YouTube
TL;DR
  • Jehosaphat Research 的 Victor Bonilla 推介做空 Main Street Capital(MAIN),认为短期催化剂是公司可能削减股息,股价下行空间“可能在60%左右”。 从更长期看,他认为股价走弱本身会催化业务恶化,“因为商业模式就是如此运作的”。

  • 核心会计论点是:MAIN 对组合资产的估值远高于成本,且明显高于由33家BDC组成的 Oppenheimer 同业样本;但作为上市公司22年来,公司在已实现退出、到期和出售交易上累计亏损。 “这两件事怎么能同时存在?”按模型计价持续赚钱,按市值计价合计亏损。

  • Bonilla 认为,虚高估值集中在控制类组合,这类投资通常不会与其他BDC共同持有;非控制类组合则经常共同放贷、彼此比较估值,“看起来完全正常”。 最典型的是 Cody Pools:这笔受控优先股权益自2021年第二季度以来的指数化估值上涨200%,而上市泳池公司竞争对手的估值已“被彻底摧毁”——“他们是在泳池里用AI吗?”

  • MAIN是33家BDC中唯一使用 Grant Thornton 的公司,而GT休斯敦办公室有大量MAIN高管在 Arthur Andersen 时期的旧同事——过去11次审计中有6次由曾与他们共事或任职重叠的合伙人执行,且2025年GT停止纳入若干实质性审计程序。 Bonilla认为这是“一次宽松的审计”,并称这种“过于亲密的关系”可能起到帮助作用;他指出的是潜在利益冲突,而非笼统批评GT或Arthur Andersen。

  • PIK利息正在飙升,但当前既没有经济灾难,也没有更广泛的信贷危机;此前两次峰值分别与油价崩盘和疫情封控相伴,而商业银行逾期率虽在上升,却没有呈现危机。 Bonilla认为这更具MAIN特异性,反映的是“一场资产狂欢……把账簿塞满垃圾资产”。他的组合现金流覆盖率为0.8x;MAIN尚未削减股息,但他认为公司最终不得不这么做。

  • 那些“知道尸体埋在哪里”的内部人已经从买入转向卖出,而MAIN的市价/NAV倍数是行业最高——这还是建立在Bonilla认为本身已被高估的NAV之上,因此下行将是一个“杠杆化的问题”。

摘要 · 为研究而整理的核心内容

1. 按模型计价与按市值计价的矛盾

  • Bonilla的基本判断是:BDC必须用自有模型为非流动性私人投资计价——“这本身没有什么问题”——但MAIN的总公允价值相对成本的溢价,远高于Oppenheimer选取的33家公司同业样本。

  • 他希望你正视的冲突是:作为上市公司22年来,MAIN账面上持续累积巨额未实现升值,却在退出、到期和出售交易中累计实现亏损。按模型计价不断获胜,按市值计价却整体落败。

2. 控制类组合虚高,共同投资组合正常

  • 关键线索在于:非控制类投资经常与其他会披露自身估值的BDC共同持有——“如果我的估值是100、你的估值是50,就会引发问题”——而MAIN在这里看起来并无异常。Bonilla认为,控制类组合几乎没有其他投资与其他公司共同持有、可供交叉比较估值,但其估值却“高得离谱”。“你必须追问原因。”

  • Cody Pools图表显示,这家受控泳池建造商的优先股权益指数化估值自2021年第二季度以来上涨200%;当时大致正值疫情期间住宅和泳池建设的高峰,而上市泳池公司竞争对手的估值“已经被彻底摧毁”。

3. 他认为独特且令人担忧的审计师关系

  • MAIN是33家BDC中唯一一家由 Grant Thornton 审计的公司;GT在10-K中披露的休斯敦办公室,有大量MAIN管理层在Arthur Andersen时期的旧同事——那也是Arthur Andersen负责审计 Enron 的时期。过去11次审计中,有6次由与MAIN管理层存在任职重叠经历的前同事执行。2025年,GT“停止纳入若干实质性审计程序”。

  • Bonilla认为这是“一次宽松的审计”,并称这种“过于亲密的关系”可能提供便利。他指出的是潜在利益冲突,同时明确表示自己对 Grant Thornton 没有一般性意见,也不会笼统质疑 Arthur Andersen。

4. 股息为何会在此时削减

  • PIK利息占利息收入的比例过去曾两次在衰退相关时期见顶——一次是油价崩盘,一次是疫情封控。如今这一比例正在没有经济灾难或更广泛信贷危机的情况下暴增;商业银行逾期率虽在上升,但既没有衰退,也没有信贷危机。Bonilla认为,MAIN特有的恶化源于一场资产狂欢:“激进放贷,把账簿塞满垃圾资产。”

  • 他的“组合现金流”指标过去2年仅下滑过2次:一次是在MAIN削减股息的疫情期间,另一次就是现在。覆盖率为0.8x:“已经不可持续。”MAIN尚未削减股息,但Bonilla认为公司最终不得不这么做。

5. 内部人卖出与估值脱节

  • 他认为内部人卖出在这里尤其值得重视,因为在久期很长的借贷业务中,“这些人提前数年就知道尸体埋在哪里”——而一些曾经买入的人如今已经转为卖出,部分减持幅度很大。

  • 估值层面的关键在于:MAIN拥有行业最高的市价/NAV倍数,投资者看中的是它的EPS和更高的ROE。Bonilla认为,这些指标很大程度上都是公允价值会计的产物,因为未实现升值会计入ROE。若同时压低NAV和估值倍数,“这里的下行空间很大——取决于假设,可能在60%左右”。

Victor Bonilla

As Jim mentioned, I’m Victor Bonilla. I’m the author of Jehoshaphat Research. I’m also the chief investment officer of Carroway Capital Management. Here’s our track record as Jehoshaphat Research on the screen. We’ve been publishing short ideas for 4 years.

1. The Main Street Short Thesis

Our short thesis on Main Street Capital, ticker MAIN: Main Street is a business development company. This is coincidentally a timely topic, given that Bloomberg had an article today about the emerging cracks in private credit. If you’re paying attention to the golden age of private credit, this one is going to be particularly interesting to you, I think.

We’re short Main for 5 key reasons, or these are our key opinions about the company. I’ll leave this screen up for a minute so you can get a look at it. I’m going to be talking about each of these opinions in more detail.

In the short term, the proximal catalyst we envision for this stock is a dividend cut. Longer term, this business has a lot of problems. We think that weakness in the stock price will catalyze weakness in the business because of how the business model works, and ultimately it’ll hurt the business. There’s a lot of downside here, and we’re short.

2. Fair Value Marks Hide Losses

Key opinion 1 relates to fair values. This is the most complex part of our investment thesis, but I can distill it down into a pretty simple concept. Because Main is a business development company and owns a portfolio of private, illiquid investments, they have to mark those investments to their own models.

These are not investments that trade on a stock index, so you can’t look and say, “That’s the price of the stock, so this is what it’s worth.” They have to use judgment. There’s nothing inherently wrong with that. That’s what all BDCs do; that is the nature of accounting.

We believe Main is using extremely aggressive fair-value accounting to generate unrealized gains, or unrealized appreciation, on that portfolio. The problem is that this conflicts with a track record of realized losses, as I’ll take you through. We have to ask the question: How can these 2 things exist together?

First, Main is compared with its peer set. This is a peer set of 33 companies, and we took it from Oppenheimer—a standard BDC peer set. These are all BDCs. There are a lot of them. You can see Main on the right stands out as having a total-portfolio fair-value mark-up over cost far in excess of its peers.

Remember, fair value is largely subjective. The accounting for fair value requires a lot of judgment. Main marks its portfolio far higher than cost. The other peers generally don’t do this.

But what happens at realization? This is the conflict. While Main’s mark-to-model fair values show enormous unrealized appreciation embedded in its portfolio—that’s why the fair values are so high—at realization, Main has actually lost money in aggregate over its 22 years as a public company.

Now think about that. Why is a company showing, by far, the best unrealized appreciation? They’re winning all the time on mark-to-model investments that they hold on their own book. But when that’s marked to market, they’re actually losing money.

This chart is cumulative. Going back as far as we can find, Main has been public for about 20 years and has cumulatively lost money on exits, maturities, sales, and so on.

Where this thesis starts to get really more interesting is when you look at the difference in marks between the control portfolio, where Main has a controlling stake or controlling influence on the board, and the non-control portfolio. The control portfolio is where all the inflated marks live. We believe the control-portfolio fair-value mark-up over cost is much higher, and it’s also much more dislocated from the rest of the peer set.

That’s the control portfolio. But in the non-control portfolio, by contrast, everything looks completely normal. You can see Main sort of in the middle there. Its marks on the non-control book are not remarkable. They don’t stand out at all. There’s nothing to report; there’s nothing crazy about that.

What’s the difference, for our purposes of analyzing this, between the control book and the non-control book? Yes, we have control of these companies. We don’t have control of these. But the key difference, for the purposes of thinking about fair value, is that the non-control investments are frequently shared with other BDCs.

Which means if I have a loan to Company X, another BDC down the street might also have a loan to Company X, and they report their own fair values. If my mark is 100 and your mark is 50, it’s going to raise questions.

So, like I said, in the control portfolio, where we don’t believe there are any other investments that share marks with other companies, those mark-ups are astronomically high. They stand out against the whole peer set. But in the non-control portfolio, everything looks totally normal. You have to ask the question: Why?

Which brings us to stories about the individual companies. I’m not going to go through all of these in detail. They will be in the write-up that we’re going to post on our website shortly. We’ll go through individual crazy mark-up stories.

One story that lends itself well to graphical interpretation, just to sort of paint the picture of how this might work, is Cody Pools. Cody Pools is a controlled investment by Main. They invested in the preferred equity of Cody Pools.

That red line is the indexed marks of Cody Pools. Everything starts at 100 on this chart. The chart begins—I don’t know if you can see this—it begins at Q2 2021. That was the peak of COVID, approximately the peak of home construction, and the peak of pool construction, which is the business that Cody Pools is in.

All those other lines at the bottom—you may have already guessed this—are the tickers of publicly traded stocks of pool companies. These are Cody Pools’ competitors. Their valuations have gotten annihilated since the peak of COVID. Cody’s up 200%.

Are they using AI in their pool? I couldn’t tell you that, but that’s a fair question.

3. Auditor Ties Raise Concerns

This brings us to key opinion 2, which is that we think Main’s auditor-client relationship is rather unique and concerning, and it may be enabling the aggressiveness that we believe is happening in these fair values.

Main is the only BDC to use Grant Thornton as its auditor. That’s okay. We don’t have any problem with Grant Thornton. The problem we have is that Grant Thornton’s Houston office, which is auditing Main, is heavily staffed by people who used to work at the same company that a lot of Main’s executives came out of.

A huge number of Main executives—most of the senior leadership—worked at Arthur Andersen, which was Enron’s auditor. I’m not going to cast aspersions on Arthur Andersen. I’m not going to do the Enron thing. You can cast aspersions on Enron. I can’t believe anybody here saw Enron coming at the time, so who would have known?

The Enron piece of this is interesting from a storytelling perspective, but what’s really interesting is the fact that Grant Thornton’s Houston office is auditing Main. This is reported in the 10-K. These are former Arthur Andersen colleagues auditing their former colleagues’ books at Main. You see the conflict—the potential conflict of interest—there.

6 of the past 11 audits of Main were actually conducted by Grant Thornton audit partners who worked with or overlapped with Main leaders at Arthur Andersen in Houston, back when these guys were auditing Enron.

In 2025, Grant Thornton stopped including a number of substantive procedures in Main’s audit. We think this is a lax audit that’s taking place, and the cozy relationship here may help. I forgot to mention that, of the 33 BDCs in the peer group, Main is the only one using Grant Thornton as its auditor. Again, nothing against Grant Thornton at large.

4. The Dividend Faces Pressure

This brings us to the timing of the short. I call it key opinion 3: The dividend is now under threat.

This chart shows PIK interest as a percentage of total interest, for all interest income collected by Main. You can see that there are peaks and valleys. The prior 2 peaks correlated to recessions. They correlated to the oil crash and COVID lockdowns, which directly affected Main’s portfolio. Main does a lot of business in the Southwest and was heavily exposed to oil and gas.

What is the economic catastrophe that’s been happening through 2024 or Q1 2025? There isn’t one. And to prove it, this chart shows commercial-loan delinquencies across the U.S. for all commercial banks.

There’s no recession here. There’s no crisis happening in credit. I mean, the numbers are ticking up, but there’s not a recession happening. This is idiosyncratic to Main, and we believe it’s the result of an asset binge: aggressively lending, filling the book with garbage, and now you’re starting to see PIK explode despite this not happening in the broader economy.

This is affecting Main’s cash flows. This is a number that we calculate—a metric that we calculate—in our own report. We explain it. It’s not that complicated, but we call it portfolio cash flows. It’s essentially the amount of cash flow that the portfolio is kicking off, excluding exits and new investments.

It’s only declined twice in the last 2 years. One was during COVID, when Main cut the dividend, and it’s happening now—not because of a recession. Main hasn’t cut the dividend, at least not yet. We think they’re going to. We think they’re going to have to.

Part of the reason we think that is because the cash-flow coverage is below 1.0 times. It’s 0.8 times. As you can see, this is the cash-flow coverage. It’s no longer sustainable. This can’t go on.

Key opinion 4: insider selling.

5. Insiders Start Selling

This sort of speaks for itself. I'll make it very quick. I just think that this is more predictive in a lending business with long-duration assets because insiders know what's in the book. They know the problems, typically in many cases years in advance. It's not like a widget company where you're guessing about demand in the future. These guys know where the bodies are buried.

They're sellers now. They used to be buyers. Some of them are selling in large percentages.

6. Valuation Disconnect Creates Downside

And then, to bring this to the final slide, there's a valuation disconnect here. Main Street Capital's stock is valued at the highest price-to-NAV in the space. Everyone thinks it's the best company because it's got all the EPS and a higher ROE. These are products of fair-value accounting in large part. Unrealized appreciation flows through ROE.

This is a double, sort of a leveraged problem, because Main Street Capital's NAV, in our view, is heavily inflated. This is a multiple of that NAV, and that multiple is the highest in the space. You deflate both appropriately. We go through this more in our report, but it's not exact. This part is not rocket science—just valuation. There's a lot of downside here, maybe around 60%, depending on the assumptions you want to use.

And with that, that's it. Thank you for listening.