6 ms·

Play bank run games, win bank run prizes. Really, I don't love the regulatory arbitrage played by SVB and unhedged duration risk, nor the moral hazard created

• rcme 4y ago
  The Fed’s owns projections have rates coming down in 2 years. Of course the 10Y yield curve is inverted.
• jgilias 4y ago
  You sound like you have a good grasp of this! What’s your take on the observation that historically after the yield curve inversion ends, it’s 3-6 months until a recession?
  • qwytw 4y ago
    It's a correlation. Also a big part of the inversion is the fact the market expects the Fed to start cutting rates in 3-6 months. Of course historically this tends to happen at the beginning of recessions...
    • kurthr 4y ago
      I agree. I'm an amateur, but I'll give my personal answer: There will be a recession.. but when? I think sooner than later, but next year NBER could declare it to have started today or maybe it starts in another 12 months. There are "large and variable lags to monetary policy", which usually means >12mo. Almost a bigger issue than the actual risk free treasury yields is the rate volatility that's forcing lending rates up, because bankers don't know what's going to happen, and that makes loans and hedging expensive. The speed of the rate rise, along with the size and duration of inversion is very large and relative to previous rates it's a huge move (doubling from the 1-2% previously). That's going to break something some time. It could be when the debt ceiling fight gets real or earlier, if the economy actually turns over. You'll know when the Fed starts cutting rates.
• mitthrowaway2 4y ago
  > BTW you can blame the Fed for low interest rates, but it's the yield curve inversion and long rates which caused the liquidity/solvency problem not the short term rate hikes Can't we sort of blame the Fed for that [yield curve inversion and long rates] too? It undertook massive quantitative easing during the pandemic, which depressed the yield of long-term bonds such as those bought by SVB. Perhaps if it hadn't done so much QE, the yield on SVB's long-term bonds would have gone from, say, 3% to 4% instead of 1.56% to 4%. Edit to clarify: I'm not saying that the Fed deserves blame for SVB taking such a risky long-term position, I'm saying the Fed deserves some blame for long-term bonds being risky.
  • wbl 4y ago
    Please explain to me what you think rates would have needed to do given that there was a massive decrease in economic activity due to a pandemic followed by inflation.
    • mitthrowaway2 4y ago
      IMO? Overnight term rates should have been zero from 2020 to about July of 2021, then allowed to rise by 0.25 per month to perhaps about 3.5%. Long-term rates should have been left to float with the market, pricing in the expected risk of inflation. Not that my opinion matters, since I wasn't in charge.
      • wbl 4y ago
        That still leads to losses on long term treasuries. That's what it means to raise rates.
        • mitthrowaway2 4y ago
          Yes, of course it does. But there are losses, and then there are losses. 10-year yields going from 1.56% in 2021 to 4% in 2023 is equivalent to a price drop of 17% (given maturity in 2031). If 10-year yields had only been, say, 3% in 2021 before rising to 4%, the bond price drop would have been closer to 7%.
• LapsangGuzzler 4y ago
  SVB also chose not to prioritize hiring a risk manager for months leading up to their collapse, which is just pure stupidity. Anyone worth their salt would've pointed out that they were at risk due to their investment strategy and homogeneous customer base. We're seeing old lessons from the 80's being retaught in the banking world. Never put all of your cash in one bank. Keep your debt and your liquidity held in separate banks because if you don't and your bank goes under, your bank debt is written off against your balance when the bank liquidates. The hyper-connected world we've created prioritizes efficiency and optimization at the expense of operational redundancy, which leads to people getting caught doing stupid things like putting all of their money into one business bank that had no visibility into their own risk profile.
  • rodgerd 4y ago
    > SVB also chose not to prioritize hiring a risk manager for months leading up to their collapse Closer to a year, I believe. On the back of lobbying for exemptions from Basel III adequacy requirements.
  • htss2013 4y ago
    Maybe they didn't "chose" to not hire one. Maybe every competent risk manager they wanted to hire wanted nothing to do with SVB because a competent risk manager would have been able to foresee what happened.
• 3327 4y ago
  [dead]
• somedudetbh 4y ago
  > unhedged duration risk, nor the moral hazard created by the bailout Just making sure I understand your point here, are you just against "banking"? Unhedged duration risk w/ LOLR backup is essentially "the banking business". What's the moral hazard? The equity is wiped out, most of the debtors are wiped out. The only moral hazard I see is we've disincentivized individual depositors from assessing the financial strength of their banks. To me, this is a good thing. I hope we can bring similar moral hazards to choosing a plane to fly on, bridge to cross, building to enter, restaurant to eat in, hospital to go to, etc.
  • bagacrap 4y ago
    Other banks had less duration risk and some used hedges. No bank is going to profit from this particular environment but it's possible to manage assets such that the entire business doesn't implode.
    • landemva 4y ago
      While it is possible to manage for safety, regulators did not enforce that.
• akira2501 4y ago
  > you can blame the Fed for low interest rates, but it's the yield curve inversion and long rates which caused the liquidity/solvency problem not the short term rate hikes I blame the bank management. They left the risk management position open and spent way too much time, money and effort on marketing during that period of time rather than shoring up their shaky position.
  • MilnerRoute 4y ago
    And even then, this article is about the venture capital industry and private equity firms - which for the most part is entirely different than banks. The problem is apparently that some VCs invested in banks - and banks are about to be more heavily regulated. That's a good thing -- it will get banks implementing the backstops they should've had all along. I really don't mind if some VCs make less money than they'd hoped on their investments in risky banks. Also, this $500B number is spread across the entire VC industry. And even then, most VCs have heavily diversified portfolios. Just for example, one investor in SVB was Insight Partners -- but their web site lists 800 different investments. They're part of that $500B number, but it will have very little effect on their overall portfolio.
    • phoehne 4y ago
      Actually, I think what they're talking about is that VCs have a lot of investments they would not dump more money into. They use these investments as collateral for loans, based on some kind of valuation (which, given the lack of price discovery is arbitrary). Essentially, they're zero value investments. If these investments were repriced, the actual value of the VC fund would fall. As the interviewee indicates, if the uninsured depositors had not been bailed out, the VC firms would have had to decide to put more of their "dry powder" (on a temporary basis) into these startups until their deposit claims were adjudicated. the VCs would not have done that, even if they knew they'd get all of it back when the dissolution was complete. That means that the 10 million investment in a 1 billion dollar unicorn currently on their books would suddenly be worthless. Instead, they're going to continue to carry it at $10 million because there's no price discovery on these highly illiquid assets. Essentially, the last price discovery was when they made the investment which caused the value to be set.
• JumpCrisscross 4y ago
  > have to plan to lose >20% of your deposits in a single day you're not a bank anymore When you sell your liquid portfolio to Goldman at a $2bn loss and then announce non-binding equity commitments, you’re going to lose 20% of your deposits. Now that banks can borrow against the face value of their Treasuries, the same duration problem shouldn’t recur. (That said, we’re stuffing an ungodly amount of crap into the FHL bank.)
• arcticbull 4y ago
  > nor the moral hazard created by the bailout There was no moral hazard created because bank shareholder equity got zeroed out. Bank management and shareholders were not protected against the 'find out' phase.
  • cwilkes 4y ago
    The moral hazard is that there isn’t a a limit to the $250k FDIC insurance so people that put money into the bank don’t have to care what the bank does. So there’s no incentive to work with a bank that took the time and money to pass a stress test — in fact the one that didn’t bother to do any testing can give better terms as they aren’t spending money to be safe.
    • arcticbull 4y ago
      I don't think it should ever be the depositors responsibility to figure out whether a bank is properly managing their risk backing your deposits. That's both intentionally meant to be opaque to depositors - you get dollars in an account, not share in an MMF for instance - and also, it is incredibly difficult for even professionals to evaluate. This is the responsibility of regulators plain and simple. And I'd argue by the FDIC taking this risk on now, moral hazard is still not a factor.
      • tripletao 4y ago
        I'd tend to agree that expecting depositors to police their banks is bad policy. It would be better to make that policy change explicitly though, by insuring all deposits, rather than by slouching into it with ad hoc rescues like here. I agree there's no moral hazard as to the SVB shareholders, since they got zeroed. There is a moral hazard as to the shareholders of other banks, who will benefit from the new lending program in proportion to the amount of bad interest rate risk they took.
        • 8ytecoder 4y ago
          That’s the point IMHO. We want depositors to keep money in the banking system. The inverse of it where depositors don’t trust the system would result in even more bank runs. However well run a bank is, there’ll always be a certain amount of assets in long term that lose value in the short term. We’ll be seeing perpetual bank runs and a new shadow banking system will emerge if this continues unchecked.
• landemva 4y ago
  > not raising short rates would increase inflation Proxy war in eastern Europe, with USA dumping big $ there, is causing price rises.
  • bagacrap 4y ago
    How does that domestic impact service sector wages
    • landemva 4y ago
      Large debt issuance to fund federal government deficit spending causes dollar to devalue. This affects price of labor paid in dollars.
  • nostromo 4y ago
    Inflation was surging well before the Ukraine invasion.
    • landemva 4y ago
      The virus lockdowns ended and manufacturing supply chains have been clearing. Please tell us why price increases are continuing, instead of dramatically dropping.
      • nostromo 4y ago
        Because we printed trillions of dollars and haven’t come close to removing all that excess stimulus yet - nor have any supply side constraints that existed prior to Covid been removed.
        • landemva 4y ago
          I thought the corona payments ended. Computer chips and PCs are largely back in stock. Fertilizer and fuel prices went higher in past year. This drove up food production prices. Molecule flow in pipeline to Germany was sabotaged, preventing sale of molecules from east to west which drove up natural gas prices at export ports in USA. Not a virus issue but a war issue.
• joe_the_user 4y ago
  ...really this was a bank run pure and simple I've been around and around on this question. Insolvent or illiquid, illiquid or Insolvent... etc. It was solvent on paper, by what it had to record on it's books. But it was insolvent by mark-to-market (which it didn't have to use but which the sophisticated but not-that-sophisticated investors, say venture capitalists, would assume is the reality). But hey, you could say it was solvent by the Fed supporting every bank it's size. Then again, the Fed didn't come through in time, so maybe the run was the reality [1]. And you could say "this was a run 'cause picking on SVB in particular was unfair since a significant portion of all bank capital in long term bonds that put them on an "overhang" of unrealized losses. [2] But maybe, the Fed needed to cool the economy so killing a few banks was needed and SVB and signature were the most logical. [1] Matt Levine in Bloomberg https://archive.ph/uIYtY https://archive.ph/uIYtY [2] https://finance.yahoo.com/news/u-banks-sitting-1-7-211212318.html https://finance.yahoo.com/news/u-banks-sitting-1-7-211212318...
  • tripletao 4y ago
    > which it didn't have to use but which the sophisticated but not-that-sophisticated investors, say venture capitalists, would assume is the reality It seems like you're implying that a truly sophisticated investor would view this differently, but I don't see how. By all indications, the MTM price was economically correct--bid/ask spreads and trading volumes were normal, and the price was very close to what a simple NPV model would predict. There are cases where the market price is economically wrong (in a "liquidity crisis", "fire sale", etc.), but there's no evidence of that here. Managers and shareholders of the SVB and other banks with similar losses have strong self-interest in arguing otherwise, and they're doing so quite successfully. It's particularly easy to confuse people about interest rate risk, since it's so abstract--you're still getting the same future cash flows, and it's hard to explain why they're less valuable without a concept of NPV or other bond math, which relatively few people understand. The economic loss is just as real as with any other risk though, regardless of what the accounting says.
    • joe_the_user 4y ago
      I suppose I shouldn't necessarily imply "truly sophisticated" but one group of depositors would just say "of course the Fed will support the depositors, there's no reason to worry" your argument about insolvency is correct.
• grozzle 4y ago
  Last week's David Mcwilliams podcast [1] goes into a good explanation, generally matching your reaction. David is an economist who's regularly been around all the top conferences, worked for a couple of central banks, and he is good at explaining how you can tell when the emperor has no clothes. 1 https://podcastaddict.com/episode/154571052 https://podcastaddict.com/episode/154571052
• ouid 4y ago
  There is a hard line between insolvency and illiquidity. Illiquidity can be solved by borrowing at the federal funds rate.
  • kurthr 4y ago
    I kinda agree here, it's a cash flow question. But that doesn't resolve what the haircut (if any) should be put on the securities in question nor what their future value will be. The BTFP does it for one year without a haircut. Is it long enough? Depends on where long term rates go. If they fall a couple of points in the next year, it could, but if the Fed fails to beat inflation and they rise... then it wouldn't.
    • cinquemb 4y ago
      The other problem with BTFP is that it makes less HQLA collateral available in the swaps market for securities with similar tenors to those locked in BTFP. In one hand you increase liquidity for the most liquid assets on bank books, and on the other hand you decrease liquidity for least liquid assets on bank books (and private holders of said securities, private $ denom debt > us gov debt).
• alxmng 4y ago
  Blame properly rests on the system that is designed to produce bank runs. The system is designed such that bank runs are possible and inevitable. It is not risk management or the Fed’s interest rates, it is the practice of investing deposits while simultaneously promising those deposits are available. Instead of changing this system, the Fed was created to solve a problem the system inherently produces. If deposit accounts were held at the Fed, and investment and loans were kept separate, there would be no possibility of bank run.