Kathryn Judge On Financial Fragility In The Shadow Banking System, Interbank Lending, & Overlooked Lenders Of Last Resort During The 2008 Financial Crisis
Kathryn Judge is Professor of Law at Columbia University School of Law.
She is Chair of the Research Committee of the European Corporate Governance Institute, has served on the Financial Advisory Committee of the U.S. Department of Treasury’s Office of Financial Research, and was a Law Clerk to Supreme Court Justice Stephen Breyer.
By Aiden Singh, July x, 2025
The 2008 Financial Crisis As Origin Story
Aiden Singh: You're a lawyer by training - you studied at Stanford Law and you teach at Columbia Law School.
But your research is heavily focused on finance and financial regulation. How did that interest, come about? Was it something that preceded your time in law school? Did it develop after law school?
Kathryn Judge: That’s a great question because I actually think there should be more people at this nexus between finance and law.
I came to it the same way that many peers in legal academia came into this field: in the wake of the 2008 financial crisis.
I'd gone to law school, clerked in Chicago, I clerked in DC, and then I was out in San Francisco working on mergers, securities, and corporate governance.
And one of the clients for whom I had done a lot of securities work had engaged in a number of different securitizations. And suddenly, their balance sheet didn’t make any sense: no matter what you did, if you were trying to use Generally Accepted Accounting Principles (GAAP), the two different sides just didn't match up.
We were experiencing market dysfunction in so many different places. And so suddenly, you're in this world where there's all of these things happening that theory says shouldn’t be. So all the theories that we had about the efficiency of capital markets were very quickly being proved inaccurate and, actually, were being proved harmful, because they left people ill-prepared for the degree of dysfunction that we saw.
So that really catapulted me into academia. I'd always played with the idea of being an academic. I always liked the idea of teaching. I liked the idea of research. But suddenly, I had these pressing questions that nobody seemed to have good answers for.
And it was the desire to better understand how this could happen and how we could prevent this from happening that really drove me into academia. And that's where I've been ever since.
- - - - - - -
The Shadow Banking System
Aiden Singh: In your analysis of the systems of financial intermediation, you make a distinction between capital markets, the banking system, and the so-called ‘shadow banking’ system (which overlaps with the capital markets).
Can you describe for our readers what the shadow banking system is?
Kathryn Judge: The shadow banking system is oftentimes referred to it as non-bank financial intermediation. That has become the the umbrella term of choice.
And the core idea is that, historically, when we thought about financial intermediation, we thought about banks because banks were the primary financial intermediaries.
And by intermediaries, we mean financial institutions or structures that engage in maturity transformation and liquidity transformation.
So one of the ways banks really create social value is by extending loans to households and to businesses and helping credit flow to where it can enhance productivity or the economy.
At the same time, a second function they play is providing a safe place for people to put their money. And then along with that, they play a really critical role in the payment system.
So that coupling of the creation of loans with the creation of money is a defining feature of banks that is incredibly useful - but also fragile.
And so we have a very robust system of banking regulation and deposit insurance to deal with this fragility.
What has happened over the last half century is there's been a significant growth in the provision of both of these types of services outside of the banking system; functions that were traditionally taken on by the banking system are now also taking place outside of the banking system.
And there several reasons for this. Part of it is technology and innovation enabling the rise of financial intermediation outside the traditional banking system. And part of it is regulatory arbitrage because the shadow banking system is not subject to the same type of prudential regulations as the traditional banking system.
Aiden Singh: What kinds of institutions are engaged in this sort of shadow banking?
Kathryn Judge: So there’s multiple definitions out there of what actually constitutes shadow banking.
But things like the private credit funds that are currently becoming popular are certainly among the types of things that would fall under the definition - they’re non-bank funds that create credit.
I think we also see a lot of it at the the low the the shorter end of the maturity curve - things like money market mutual funds.
What you're really thinking about is situations where somebody who traditionally would go to a bank for a particular financial service that is now going to a non-bank institution.
And then the key issues for the purposes of regulation are: what function is that entity serving and are there reasons to be worried about it for purposes of financial stability?
So, for example, money market mutual funds provide a very nice alternative to a bank deposit. And people put their money in them because they think money market mutual funds are really safe places to put their money. But as we saw both in 2008 and 2020, as soon as a shock hits the economy that suggests that maybe they're not so safe, the government has to come in with a significant backstop.
Now, that doesn't mean that these instruments shouldn't exist: they can play an important role. But we don't yet have the the optimal system in place for trying to make sure that these new types of of non-bank alternatives are appropriately regulated given the threats they might pose to the stability of the broader financial system or the risk of runs on those institutions that might exist in situations where there's a first-mover advantage.
Aiden Singh: The shadowing banking system is estimated to be larger than the traditional banking system.
And you’ve argued that it is more fragile - and that it was a major contributor to the 2008 financial crisis.
Why is it more fragile?
Kathryn Judge: The difference in the regulatory paradigm can make it more fragile.
We’ve seen that banks are fragile entities at times. But with banks we have structures in place that, while flawed, are at least conceptually sound. For example, most notably, the government comes in and says, if you're a depositor up to a certain amount, you're fully insured. So rather than actually having to trust or know anything about the the quality of the assets backing your deposit or the quality of the management running the financial institution, you know that the government has a very robust regulatory system in place to protect your deposit. And the government limits the types of activities banks can engage in. Plus there is ongoing supervision. So that all makes deposits a lot stickier.
The challenge is, when you get outside of the banking system the government usually does not provide those types of explicit backstops. And there's good reasons it doesn't provide that type of explicit backstop. But the absence of such a backstop is part of what can give rise to a first-mover advantage for withdrawals.
Money market mutual funds are the classic example of this.
And it’s also worth thinking about how this issue arises for things like open-end bond funds. An open-end bond fund takes corporate bonds - which are historically not very liquid instruments - and then offers investors daily liquidity against those instruments, based on what the value of those instruments are today. The challenge is that, if you're an individual investor in a bond fund, if you issue a sell order during a period of distress, you're putting pressure on the fund to sell a bunch of bonds into what's potentially a distressed market where it might get fire sale prices.
When a shock hits, you get something analogous to an old school bank run where people used to line up at the door to get their money back. Investors are really pulling their money out in mass. And that creates dysfunction in the secondary market, for corporate bonds, which in turn can also make it harder for companies that are trying to issue new bonds and raise new money to do so because there's a lack of liquidity.
So you can see the dysfunction spreading.
And we don’t yet have a perfect regulatory paradigm for how to respond when there is bank like activity that's happening outside of the banking system which has the potential to give rise to the risk of runs.
Aiden Singh: So existing financial regulations are ill-suited to regulating the shadow banking system. Is this due to the fact that regulation of the shadow banking doesn't fall under the purview of existing regulation, or is it that the existing regulation, even if applied to shadow banking, is just not appropriate for this new way of creating credit?
Kathryn Judge: It's both.
The potential regulators don't have oversight authority over key components of the non-bank financial system.
But there's also a lot of reason to think that we actually need a different type of regulatory scheme to deal with the shadow banking system, one which is something in between the classic paradigm for capital markets and the classic paradigm for banks.
So for banks, there's deposit insurance. And they can also go to the Fed at any time and get access to loans through the discount window. It’s a situation where there's prudential regulation, access to liquidity, and a government guarantee that's provided in exchange for that oversight.
When we move into the capital markets, market discipline as opposed to government oversight is playing more of a role. But as we just discussed, that's imperfect in a variety of different settings, because there can still be significant collective action challenges. And when those collective action challenges arise, there can be spillover effects that really give rise to dysfunction.
And so what we need to do is try to understand what is the right framework - the right set of tools - for dealing with the risks and benefits and distinct roles played by different components of the non-bank financial system.
So we need a better paradigm. And that's one of the significant shortcomings of the reforms after 2008, even though the shadow banking system was a huge part of that crisis. We tried to address it through the Financial Stability Oversight Council. But it was a imperfect effort at trying to get at the significant risk that this sector poses.
Aiden Singh: In a Virginia Law Review article, you laid out why existing financial regulations are ill-suited to regulating the shadow banking system.
You develop the notion of ‘information gaps’ that are present in the shadow banking system, but aren’t present in the traditional banking system. What are these information gaps and how do they contribute to financial fragility?
Kathryn Judge: So information gaps refer to information that is knowable - it’s not situations of Knightian uncertainty - but takes a lot of work to figure out. And part of what we've seen with information gaps is the development of structures under which nobody has the right incentives or capacity to make sure that they have all of the relevant information.
So consider traditional banks. First, they have bank investors, equity holders who are meant to engage in due diligence into the banks. And then there’s also the government which engages in supervision and regulation. And the government does a lot of the heavy lifting to understand and respond to the risks that banks might be taking on and to correct for those risks when they're excessive.
In the capital markets, we require investors to engage in this process. So we say, ‘O.K., we're going to require a lot of information disclosure from issuers of assets. And then investors are responsible for using that information to shift the prices up and down by virtue of their buying or selling activities. And so the prices actually contain information about what disparate investors across the market think about the riskiness of a particular financial asset.
But the challenge we saw in 2008 was that there were these very layered securitization structures. There were layered situations like having a secured item that's going into another securitized asset, that would go into an off balance sheet investment vehicle which was sponsored by a bank, which would then issue a short-term commercial paper, which might then get bought up by a money market mutual fund.
And when everything's working well, this layered complex arrangement can actually be very stable for a long period of time.
But when something happens - like a series of downgrades of securitized assets as there was starting in February 2008 - then people suddenly realize that they’ve been relying on a AAA rating moniker to tell them something about the quality of the assets, but they hadn't actually been doing due diligence; they didn't actually have the information necessary to understand the risks of these assets. And so when investors realized that they'd been relying on proxies rather than actually doing due diligence, and they learned that those proxies were not reliable, their instinct was to withdraw. And that rapid withdrawal created a lack of liquidity and dysfunction which then had significant spillover effects.
Aiden Singh: Given the presence of these information gaps that increase financial fragility in the shadow banking system, and given the inability of existing regulations to address this, how would you recommend the shadow banking system be regulated?
Kathryn Judge: One piece of it is making sure that there is some entity out there that is responsible for responding to the changing structures in the financial system. Currently, there’s the Financial Stability Oversight Council (FSOC), but it needs to be something more powerful than the FSOC, which works to identify the sources of systemic risk ahead of time.
You could also have - and this is something I’ve proposed along with my colleague Dan Arwey - outside commissions organized every five or ten years that make specific recommendations to Congress about how the financial system is changing.
The financial system is dynamic. And we want it to be dynamic because that creates a lot of benefits. But it is also going to create new risks. So you can have a new commission that comes in every five or ten years and takes a fresh look into the risks developing in the financial system.
So we need to build structures whose job is to look for potential sources of systemic risk and look into how those should be addressed.
We also need to do a better job of making sure that the regulatory system is able to adjust to the threats posed by different types of financial institutions. This was actually the idea for the Financial Stability Oversight Council. It was supposed to designate non-bank financial intermediaries as systemically significant where appropriate. And in the past, we’ve even seen it use this power to designate non-bank financial institutions such as large insurance companies and GE Capital as systemically significant. But today there's zero non-bank financial institutions that have this designation, and that's been true for a considerable amount of time since the first Trump administration and all the way through the Biden administration. And this is despite the fact that the non-bank financial intermediation system has grown and its complexity has increased.
And the way the FSOC was designed, I think, doesn't provide as much flexibility as would be optimal to allow regulators to really respond in a nuanced way to the type of threats posed by different types of entities. But I think the core idea of saying, ‘If you're starting to pose some kind of threat to system flexibility, there has to be a way where we can responsibly, in light of those risks, bring you within the purview of the federal prudential regulatory system’ is the key.
The last thing I would note is that the US is one of the only places where insurance companies are not subject to meaningful federal oversight, and that's been a significant and chronic challenge because state regulators just don't have the capacity to really understand and address the systemic stability concerns that can arise from the type of activities insurance companies are undertaking.
- - - - - - -
Interbank Interest Rates
Aiden Singh: The 2008 financial crisis inspired your move into legal academia. Just four years after that crisis, the London Interbank Offered Rate (LIBOR) scandal broke out.
In the wake of that scandal you co-chaired the working group at the U.S. Treasury Department’s Office of Financial Regulation (OFR) on transitioning away from LIBOR.
Can you tell us a bit about the OFR and its role?
Kathryn Judge: The OFR was originally meant to be independent, but located within the Treasury. I think now it's clearly not an independent entity, which is unfortunate.
It was meant to be independent precisely because if regulators are falling short, we wanted somebody else to sound the alarm bells. We wanted some other entity to be able to say that there's systemic risk building in the financial system. That was one of the core reasons the OFR was given some independence, including some control over its budget, and some other ways in which it could operate somewhat independent of the executive branch. But this level of independence has dissolved over time and is not likely to survive the overall shift towards a unitary executive.
So I was part of their advisory committee, made up of academics and industry professionals who work in different areas of the financial system. And our role was trying to help OFR understand how it could best fulfill its role of monitoring where systemic risk was arising in the financial system.
One of the things that was so interesting about the LIBOR situation was that it's a great example of why you need sometimes the government to intervene - it was a situation where market participants were not going to move away from an outcome that's suboptimal, even when they know it's suboptimal for them so long as everybody else is continuing to use it.
So LIBOR was used initially to create what was effectively an interest rate that was meant to reflect the current interest rate environment. The idea was banks are all going to be relatively safe. They're making these very short-term overnight loans to one another and understanding the rate that they are charging one another for those loans is going to be a byproduct of the current interest rate environment, more than of credit risk.
And the LIBOR became a very good anchor that could be used for creating variable rate interest products. So you could do LIBOR plus, say, 200 basis points. And so that way, as interest rates moved up or down, the two parties to that transaction were each bearing the changes in the interest rate environment as opposed to one benefiting or being harmed as a result of changes in in the overall interest rate environment.
The challenge that arose after 2008 was that we realized unsecured lending was actually risky even if it's between banks, and it can add to fragility. So we had this incredible decline in the amount of interbank lending.
At the same time, there had been an incredible proliferation in derivatives that were interest rate swapped using LIBOR.
And there was abuse: there was actual dishonest behavior trying to game the system.
And everybody knew this, yet market participants did not move away from using LIBOR because they didn't have a readily available alternative. Everybody was used to it.
It took a lot of government work and a lot of pressure to force financial institutions and investors in the financial system to adopt a non-LIBOR alternative. And so it's an example of a situation where regulatory pressure is needed.
Aiden Singh: In addition to your work advising on interbank interest rates at OFR, you’ve written about how interbank discipline, via the rates at which banks lend to each other, affects bank risk taking - how it discourages banks from taking some types of risk while potentially encouraging them to take other kinds of risk. And you’ve written about how this might lead to inefficiencies and what the regulatory consequences might be.
What risks are encouraged and discouraged, how might these lead to inefficiencies, and what are the implications for regulators?
Kathryn Judge: One of the things that's striking is that unsecured exposures have actually declined a lot, in a whole variety of ways. And a lot more derivatives are now centrally cleared, and there's different types of collateral requirements. So we see very different structures in place today than we saw pre-2008.
But that being said, that creates different sets of challenges.
So collateral reduces your need to know as much about the counterparty, but that also reduces that type of discipline.
And then the question is, if the discipline of sophisticated financial market participants monitoring one another because they have exposures to one another is lost, what serves as a substitute?
And there's also challenges when relying more on collateral because collateral squeezes can create a lack of liquidity.
So the near failure of AIG is a good example of this. AIG was not engaged in significant maturity transformation like banks. It wasn’t subject to a run. Instead, it had a lot of counterparties who said, ‘look, heightened volatility reduces what we can bank on your assets that you posted as collateral being worth, and you've lost your AAA rating. And, therefore, you need to post a lot more collateral’. And posting collateral can pull liquidity out of a financial institution very, very quickly in ways that can give rise to fire sales and have disruptive systemic effects.
So a lot of the work OFR was meant to do is mapping out the financial system to understand where there are exposures, where there's due-diligence applied in an ongoing way, and where is there actually not diligence being applied in an ongoing way which could lead to instability.
So if we want to understand financial stability and how to build a resilient financial system, a lot of it is about capital requirements and the like. But it's also really going to require us, given the rise of non-bank financial intermediation, to have a much more sophisticated understanding of how those markets are actually structured, who's relying on what type of funding to provide what type of credit, and what are going to be the spillover effects.
- - - - - - -
Overlooked Lenders of Last Resort During The 2008 Financial Crisis
Aiden Singh: When banking crises do erupt and we think about lenders of last resort who can step in to provide liquidity to troubled banks, we tend to immediately think of central banks.
But you’ve looked into the 2008 financial crisis and found that, in addition to central banks, banks actually turned to two other sources of liquidity as the financial system seized up.
What were those two other sources of liquidity?
Kathryn Judge: One is insured deposits. So one of the things that we've seen is that when some banks get into trouble, they start to lose deposits. And so they have to find a way to keep funding themselves, and what they do is they offer a higher rate of interest on insured deposits. And people are willing to accept that because, even though there’s a risk of failure, the government is backstopping those claims.
And this can actually have significant effects for banks that compete with the institutions that are offering these higher rates. So for example, during the savings and loan crisis, Texas had the most troubled S&Ls, so they were competing for deposits by offering much higher rates on insured deposits. And that caused banks in Texas to offer higher rates to their depositors than banks in other parts of the country even though they didn't have more attractive loan opportunities. So it really was unhealthy for the banking system.
The other source of liquidity banks tapped during the 2008 crisis is called the Federal Home Loan Bank (FLHB) system, and it’s one of the most troubling parts, I think, of the financial system and the financial regulatory regime that exists.
The Federal Home Loan Banks actually have an interesting origin story and they initially made a lot of sense. So historically, the United States had two very different types of banks.
You had commercial banks that worked with businesses.
And then there were these smaller organizations called thrifts. They often had limits on where they could make loans and they really focused on home loans and working individuals.
And when the Federal Reserve was originally created, it could only make loans to banks, and it couldn't accept mortgages as collateral. And so thrifts were outside of the Fed’s lender of last resort system. And that wasn't good for thrifts.
Nor was it good for housing. When they were created in the 1930s, the average mortgage was five years. They often had balloon features. And they just weren’t a good product for a middle class family.
So the FHLBs came along and served as kind of a quasi-lender of last resort to the thrift system.
But they also wanted to really encourage thrifts to make the type of home loans that people really needed. So they said, ‘look, if you're making a slightly longer-term loan, if it's an amortizing loan rather than a balloon loan, so if somebody makes monthly payments that actually pay off their principal over the life of that loan, we're going to allow you to borrow more money from us when you post that loan as as collateral’.
So the FHLBs put this really important rule in place helping to facilitate housing finance when it didn't exist and serving as a lender of us as ordered to thrifts. And that made sense for the first 50 years of their existence.
The challenge starts around 1980.
Thrifts were given access to the Federal Reserve, so they no longer needed the separate lender of last resort. And there were a bunch of changes in the rules governing thrifts, so they now look a lot more like commercial banks.
So the challenge is we still have this thing called the Federal Home Loan Banks, and they are able to issue debt cheaply into the capital markets because, even though they don't have an official government backstop, everybody thinks that they would be government backstopped if need be.
So they issue cheap debt and they use it to make cheap loans to banks. And so what we see quite often is, before banks get into trouble, they turn to the Federal Home Loan Banks. So Silicon Valley Bank (SVB), was the number one borrower from the Federal Home Loan Bank of San Francisco when it failed. Even though a year and a half earlier, it didn't have any borrowings outstanding. First Republic Bank was the second biggest borrower at that time.
So what we see is, when banks start to get in trouble, we want them to go to Fed. The Fed knows that there's a liquidity problem, both with those individual banks, but also potentially with a broader system. They're paying attention to it, and they have the capacity to provide liquidity if needed, but also the monitoring that's needed.
But instead, we have this FHLB system, which really is providing a significant subsidy to banks, providing liquidity backstops to banks, and oftentimes really helping weaker banks move along when they should be closed earlier and in a more timely way.
Aiden Singh: So you started to allude to one of the reasons why this could be a problem, which is that the Fed may not know that there's stress in the banking system because these banks aren't going to the Fed for emergency liquidity.
What are some of the other implications of banks being able to obtain government backed liquidity without going directly to the Fed?
Kathryn Judge: Yeah, so one is, as you said, the Federal Reserve can be a little behind the ball in terms of access to information.
Another challenge that arises is moral hazard: there is an expectation that liquidity from the FHLBs will be forthcoming when it might not. The reason central banks around the world operate as lenders of last resort is that they alone are well positioned to create as much liquidity as the system might need during periods of systemic distress. The FHLBs actually have to issue debt to raise the funds that they then pass along to their bank and non-bank clients. And so there's a possibility that they won't be able to raise that money during a period of distress.
And it’s not accounted for properly. A primary reason this overall system still exists is because the government backstop to the FHLBs is implicit rather than explicit, and so it's not consolidated onto the government's balance sheet.
So it seems like this free piggy bank, but it's a free piggy bank that exists precisely because the government's actually taking a tail risk. And and those situations tend to end badly.
——————