Welcome back to the last chapter of module 6 which is querying, settlement, and regulation. Regulation is a vast topic, one that is close to my heart. I'm going to do a 30-minute, very high level overview and then save time in the live session to do Q&A, and share more stories, and anecdotes, and thoughts on what's next for regulation. Let's jump in. Finance is something unique in the world. Of course, every business would say that. Finance along with health care and maybe few other areas of human endeavor have some specific features. It requires the ongoing participation, communication, and consent of governments and regulators, that's built into finance. You loose your banking license and it's game over. Finance is a type of technology, but because of the role it occupies, foundational role in society economy, the idea of moving fast and breaking things doesn't work well in finance. When we move slowly, things break as we've discussed. You need something in finance that's well beyond revenue, and cost, and return on equity, so this is the role of regulation. You need a safe, stable, and fair financial system. It's also important that financial innovation be not only faster, and cheaper, and a better user experience, but also safer. This is hard in particular in finance because markets are complex, unpredictable, risky and time is money, market prices move quickly. You want to make it cheaper, that's going to raise returns, but at what cost, what kinds of risks do you create? If you can do something faster and more efficiently, that's great. But if it's less safe at the same time, and often it will be left safe in ways that aren't obvious and that's something that's not going to work for regulators. Let's talk about the background of regulation for Steagall right libraries on this topic. But let's go back to the Glass Steagall Act, which is part of the Securities Exchange Acts of 1933 and 1934. There were four provisions of the United States Banking Act of 1933 that prohibited securities firms and investment banks from taking deposits. That's the Pecora of so-called Glass Steagall Act. A number of things were prohibited to banks that were part of the Federal Reserve's system or regulated by federal reserve bank and have a master account holding federal reserve deposits with the Federal Reserve Bank. These activities were prohibited by the Act; dealing in non-government securities for customers, investing in non-investment-grade securities for themselves, underwriting or distributing non-government securities, and affiliating with or sharing employees with companies engaging in such activities. Now, it's widely assumed that the Pecora Commission which investigated the causes of the 1929 Crash, actually recommended that separation of investment from commercial banking let alone in this way. If you are often connected, I've got non compelling evidence that they actually work. That's Glass-Steagall. Let's talk about the Securities Acts of '33 and '34. I'm sorry, that's a typo. Securities Act of 1933 not 1993. Both pieces of legislation arose from the financial panic of the Great Depression. I think of the two acts together, but let's keep them separate for now. Here's what the '33 Act said. It's usually just called the '33 Act. It said, every offer or sale of securities, and of course, as we've discussed that's primary issuance, using the "mean and instrumentalities of interstate commerce", that goes back to the US Constitution, must be registered with the SEC, unless exempt. There are some exemptions. There's intrastate offerings, securities of municipal, state, and federal governments, certain kinds of private offerings that are offered to specific kind of person or institution, those of you who have invested in startups will know about some of these provisions of the Act. The '34 Act, and I think about that as relating to the secondary side of things, it regulates the secondary trading of securities on venues generally, which could be exchanges or it could be Alternative Trading Systems, ECNs, or electronic communication networks, or particular subclass of ATS, Alternative Trading Systems. We've also talked in module two about REG NMS, the Regulation National Market System. The Reg NMS rule of 2005 is really the culmination of decades of work starting in the '70s, where the SEC worked to create a national market system that would break the duopoly of NYSE and NASDAQ. We've discussed that rule that requires that exchanges make bids and offers, sometimes called Ask prices. They're synonyms, visible to retail and institutional participants. The most important part of REG NMS is a rule that has a bunch of different names. Sometimes people call it, The Order of Protection Rule or The Trade-Through Rule. Technically, if you look in the Federal Register, it's Rule 611, and it requires trading centers to establish, maintain, and enforce written policies and procedures reasonably designed to prevent the execution of trade at prices that are worse than the protected quotations and that's specifically defined displayed by all the other trading centers. A laudable goal. There are multiple criticisms of REG NMS. One that I've been known to say, is well, we didn't like it when there were two venues and we liked it even less when they're 57 venues, which is about the current number without 13 exchanges the rest non exchanged venues. What's the right number? I want to say five. Certainly, I would be thinking in single digits, not 57. This idea of making sure that everyone has to portray that the National Best Bid Offer better is again, a wonderful idea, but it greatly increases the complexity and gets into issues almost of general relativity, like what is simultaneity and the speed of light becomes super important factor. Anybody who does high-speed trading and knows about NBBO is going to be able to tell you that it takes light one nanosecond to travel a foot and that's in a vacuum and it takes light a little bit longer if it's traveling in fiber-optic tube. It creates latency arbitrage and therefore the opportunity for HFTs or high-frequency traders to trade on the information than the short-term differences between prices for the same security on different exchanges. Then you get all kinds of practices, like quote flickering, putting it in order, and then making sure that that order is canceled before anybody actually trades on it, to see what effect putting in that order has in market. Most importantly it prioritizes speed and low fees rather than stability and liquidity and therefore you have the increase in dark trading undisplayed liquidity. Dark trading, dark has all kinds of connotations. But if you're specifically prioritizing speed and low fees, there will be market participants who have different priorities, stability, liquidity, getting large trades done, and REG NMS makes life incredibly complicated for them, as we've discussed. Because these institutional traders of large blocks have to access according to REG NMS and The Trade-Through Rule 611, they have to access these small quotations, which of course is going to provide information to these short term prop traders or principal trading firms. What caused the financial crisis of 2007 and 2008, just a couple of years after REG NMS? Well, I went to Wikipedia, which did a pretty good job of it. Wikipedia has all kinds of ideas. It was sub-prime lending, it was a housing bubble with easy credit conditions, weak fraudulent underwriting practices, predatory lending, banker agree, deregulation, too much leverage, financial innovation, complexity, CDO's, CDO's set squared, CDO Cubes, derivative, incorrect pricing at risk, boom and collapse of the shadow banking system, a systemic crisis of capitalism itself, something deeply broken in the banking model, or a hypothesis that is really pedestrian, but maybe that's just as good a job as any of these other things that are explaining what happened. People who shouldn't have been lending money, we're lending it to people who shouldn't have been borrowing it and then those people didn't pay it back and all kinds of consequences ensued. Then let's talk about the Dodd-Frank regulation. Dodd-Frank is the short name for the Wall Street Reform and Consumer Protection Act of 2010. There are 15 titles to that act. While I was co-head of the global Goldman equities business, I had the particular assignment from the firm, an extra job of implementing Dodd-Frank for the trading business and I can tell you that I got more gray hairs from that period than I did from the financial crisis itself. It would take me forever to tell you all about Dodd-Frank. Here's my very brief summary and you can find this in the statement of the act itself. To promote the financial stability of the United States by improving accountability and transparency in the financial system, to end 'too big to fail,' to protect the American taxpayer by ending bailouts, to protect consumers from abusive financial services practices and for other purposes. I'll just observe the protecting the American taxpayer by ending bailouts. You can look at the investments that were made in the banks and the [inaudible] considering them in the light of what's happening in the COVID-19 crisis. The banks were required to take these investments, they restructured its preferred securities. They were warned and there's an oversight group that looks at the entire tax payer journey; how much money did the taxpayer put into these banks to receive and return these preferred shares and these warrants. Then what were the net proceeds? The taxpayer did in fact make significant money on the bank. [inaudible] did not make money net on the injections into the automakers. But nevertheless that the injections happened led to a very public discussion in processes in food for at least for next ten years. The Volcker Rule gets a lot of air time. That's title six of the 15 titles of Dodd-Frank. The Volcker Rule amends the BHC or bank holding company Act of 1956. Famously, it prohibits proprietary trading but it exempts principal trading required to support the Reasonably Expected Near-Term Demand or RENTD, it takes a long time to say Reasonably Expected Near-Term Demand. [inaudible] says RENTD of the BHC's clients. This requires us to know the difference between principal trading and proprietary trading. In both cases, the bank is taking on an exposure or risk using its own capital and its own balance sheet. It is specifically not trading as the agent for a client. So the bank shareholders are going to make or lose money depending on what happens after you put the trade on. The difference is that proprietary trading is directly described as trading that is not in the service of what a particular client wants to do. Proprietary trading desk, you have a group of traders who are using analytics, intuition, gut instinct, whatever tool is at their disposal to buy things that they have the view are going to go up in value and they'll think that they have view are going to go down in value. It is not informed by customer flow and they're not putting on these trades that could be [inaudible] customers, where the customer wants to buy so we sell, the customer wants to sell so we buy. We're just doing it essentially in a vacuum away from customer flow. That's proprietary trading, then I will know that banks such as Goldman Sachs had already found that those activities were not profitable. Well before the Volcker rule that [inaudible] eliminated them and the Volcker Rule eliminated the rest of them. But it was also clear that you couldn't ban all forms of risk-taking from banks originating loans as a kind of risk-taking making markets seem important. So where the lawmakers ended up was allowing principle trading but being very descriptive and prescriptive about what kind of principal trading namely the RENTD. The challenge with RENTD, is it the state of mind? I would often say with regulators, what's reasonable? Who's doing the expecting? How near is near term? Specifically who exactly is the client? In a complicated world there will be some entities that have features of both the client and the competitor. There was another part of the Volcker Rule, that limits bank holding companies downing no more than 3 percent in a hedge fund or private equity fund of the total ownership interest in that fund. Separately, all of the BHC's interest in hedge funds and private equity funds taken together, cannot exceed 3 percent of the BHC's Tier one Capital. This is way breaking and January 2020, the fed put forward a proposal to roll back some of the rule. Specifically, rules that limit the BHC's is investment in venture capital and then securitized block. Another important part of the Dodd-Frank Act was the creation of the CFPB, the Consumer Financial Protection Bureau was proposed by Elizabeth Warren. She was not the first leader of it. The mission is to protect consumers in financial markets. It's got extremely broad jurisdiction of the banks, credit union, securities firms, payday lenders, mortgage-servicing operators, debt collectors and others. The main emphasis of the CFPB is mortgages, credit cards, student loans. It's imposed nearly the end of last year, a little over a billion dollars of civil penalties caused the return of over $12 billion to consumers. It's published on its website, nearly a million consumer complaints. Of course, as with everything important, there's multiple points of view. The consumers union called CFPB a vital tool that helps consumers make good decisions, informed decisions. The US Chamber of Commerce called it a GOTCHA game, two very different perspectives. The Dodd-Frank Act did many things to my mind. Here's really the core of it. This is the part that made a huge difference in the faith being founders, I applaud the regulators on their implementation of all of this and other parts of the Dodd-Frank Act I would say. We're in my mind and of course, this was just the personal opinion. Less necessary and less supportive of systemic PHAP and founder. There the voluntary international framework called the Basel Three framework, and the Federal Reserve in the US has done significant amount of work to reinforce, some would say gold plate or the boggled three framework. Under the voluntary framework of banks common equity peer one must be at least 4.5 percent of its risk-weighted assets, 4.5 percent is the magic number. Our common equity pure one is if the long-winded name named for common shareholders equity plus retained earnings, the RWA, Risk-Weighted Assets is just the weighted sum of the V8 fees asset since from each asset, there is a particular weight. There's a lookup table which I've given up on the right. You multiply treasuries by zero, because they are considered to be risk-free for highly rated debt, there's a 20 percent risk rate. You can see that in some cases, the risk weight, if a multiple 300 percent in the case of publicly traded equity, that's the standardized framework, there's an alternate framework called the advanced method. Recognizing that you don't really just want to look at the linear superposition looking things up in a standardized table. Because these combined in a non-linear way. You want to have banks models, but those models must be approved, and of course, one can always look at what the standardized calculation would be. Banks do the calculation in both ways. Now, here's really where the rubber meets the road this is in my mind, get a personal opinion. The most crucial and important part of the Dodd-Frank Act, which is called DFAST, together with the CCAR, they're separate properties, but they're linked in a way that all describe these two things more than anything. I've completely changed on the the risk characteristics for the operating profit fees of bank holding companies. Dodd-Frank requires the Federal Reserve to conduct an annual stress task, which are called the DFACT of bank holding companies with more than a $100 billion in consolidated assets. Let me tell you that exercise is rigorous. In conducting these tests. The Federal Reserve is using bank systems in result to project the banks balance sheets, risk-weighted assets, net income, and post stress capital levels and regulatory capital ratios over a nine quarter planning horizon. Imagine what banks had to do to become facile at modeling every aspect of their business and projecting it forward nine quarters and tie, tying together all of the financial system in adverse scenarios of various kinds, from creative by bank and some created by the bank holding companies regulator, which is the Federal Reserve. Now, there's a separate process intertwined and the reporting of results for DFACT and the CCAR. The Comprehensive Capital Analysis and Review happens over other thing to two-week period. Here, the bank is evaluating each bank holding companies capital adequacy platinum and it individual plan to make capital distributions, paying dividend, buying back shares, and it evaluates the ability of the bank to make those distributions, and still maintain the minimum capital level so it can continue performing its function in an adverse scenario. It's a barely adverse scenario published each year by the Fed in which these scenarios can change from year to year, and there's relatively little understanding that the Fed is sharing good banks F2 and the envelope and which is sampling the scenarios where the scenarios are coming from there, give it to the bank. The important thing is that the banks have to be able to run these tests, and to run these tests, there's no better example I can think of, of banks actually becoming software. Everything about the bank has to be handled in software in order for banks to be able to perform these tests in the first place. I'll pause there. Let's continue. I'll share the screen. To pick up where I left off DFAST is built into the Dodd-Frank Act. CCAR is something that the regulators, the federal reserve under the leadership of Governor Dan [inaudible] created in the two processes are intertwined. They rely on almost the same simulations, processes, data to provide three exercises, internal business processes, challenging of the assumption and the requirement of the unveiling of the results for the slightly different and the consequences of passing or not passing are slightly different. But really to step back away from the details, I want you to carry away two thoughts. First, the amount of work that went into creating these simulations across the financial system, across crop to be themes across, the V8 feed was staggering and it immensely upgraded. The convergence of software with business process across the industry. This is an area where Goldman like to think of ourselves in the lead for many years. The Fed under the leadership of Governor Tarullo really brought that undestanding and that rigor to absolutely everybody. Then showing that you're going to continue to have capital above the regulatory minimum so you can continue to perform all the functions of the bank even in an intense and severely adverse scenario, and you've got all of the system and process in data to demonstrate it, and that's amazing. That in itself creates to my mind an incredible resiliency in banks buffers that really weren't there. A bit of an editorial comment to the time of COVID-19, it was in the interests of many companies, regardless of their ownership private structure, private equity, or otherwise, to run just in time to run really close to wire. You could think of the analogy being banks running with a lot of leverage, having a relatively small capital cushion, and that works great in good times, and it's fantastic for returns in good times, but there is no resiliency. I would call the CCAR review of the ultimate anti-fragility medicine for the financial system. Again, an editorial if you took away much of Dodd-Frank, but you left this core in place, you'd have a DFAST and family financial system. I mentioned the two processes; the DFAST and CCAR together radically upgraded the risk management processes and capabilities across all of the BHCs in completing and taking to the next level this convergence of business and software. Again to BHCs must demonstrate that their capital ratios exceed the minimum at all times over the nine-quarter horizon, even after experiencing a sudden severely adverse scenario at the beginning of the simulation. These scenarios play out, for instance, unemployment. The scenario will play out over multiple quarters. To get very specific in ordinary times, BHCs must have a CET1 common equity to a one ratio equal to the bottled three minimum, as I mentioned before, that 4.5 percent. But then the Federal Reserve is put on top of that, a capital conservation buffer, which is 2.5 percent, and then on top of that, a GSIB buffer. GSIB is an immensely complicated topic. The short form is that GSIB attempts to measure the complexity, risk, and interconnectedness of a BHC as well as its size and its intricate calculation. One could dispute the way the calculation is done. It's discrete high, so if your GSIB score is within a particular range. There's a particular capital buffer that attaches to it when you're in the range and this means you go one point over there range, you jump by 50 basis point to another capital buffer. As a computer scientist and mathematician, I find the discontinuous jumps to be incredibly problematic and risk accumulate at those discontinuous and sharp boundary. Of course, I would have it scale in to me that seems infinitely more sensible. The GSIB's also calculated at one point at the end of the year, as opposed to multiple sample points are averaging and over the year, I think that's problematic. Of course gives everybody an incentive to optimize, to GSIB score as of one particular date. Leaving all that aside, there is a GSIB ratio, a GSIB buffer, in the case of Goldman Sachs, as of March 2019, the GSIB score for Goldman Sachs translated into a 2.5 percent extra buffer. Again, to review that 4.5 percent bottled 3 minimum, the Federal Reserve impose capital conservation buffer, the CCB, that's 2.5 percent. Then on top of that, specifically for Goldman Sachs given its GSIB score at the end of the prior year, that was at 2.5 percent additional buffer, adding that all up, you get 9.5 percent as the minimum. But notice that Goldman Sachs is operating nowhere near that regulatory minimum and its actual ratio in March 2019, given the advanced method was 13.4 percent. There's another creation of the Dodd-Frank Act called the FSOC Financial Stability Oversight Council, it must meet quarterly. The Treasury Secretary chairs the council. Includes a lot of incredibly important influential people. The chairs of the Fed, the SEC, the CFTC, the FDIC, and others, and identifies risks that the financial stability promotes market discipline in response to emerging threats. I will say that for me and my role at Goldman Sachs, the regulators that really are a regular part of my reality and ours reality where the FFP, the CFTC, and the Federal Reserve, especially, and there was relatively little interaction with the FSOC. Let's talk a little bit about derivatives regulation. I won't go into it in great detail, but under title seven, we have the Wall Street transparency and accountability act of 2010. In that part of the act, the CFTC, the Commodities Futures Trading Commission, regulates most security swaps, but the SEC regulates securities-based swaps. That's an important distinction, though today the CFTC regulates most swaps, generally interests rate swaps, equity index swaps. But the SEC specifically regulates the security-based swaps. I should really strike out the security in CFTC regulates not security swaps. That ought to read, the CFTC regulates most swaps and the SEC regulates securities-based swaps. Apologies for that. For instance, the swap on a single stock, that would be regulated by the SEC, a swap on the S&P 500 would be regulated by the CFTC. Obviously, it's complicated to have two different regulators, regulating instruments that are extremely similar, and one is the super position of the others. Part of the act promoted trading on swap execution facilities or staff. I think there was a lot of excitement about what it would mean to have derivatives trade on theft, but also a lot of concern that swaps would be required to be traded on the theft before it really made sense, before there was the liquidity. I don't think any of these fears played out, but neither can I mind that the swap execution facilities created some immense transformation and the derivatives business. The Act also upgraded margin requirements in complex and important way. Let's talk for a moment about the evolving COVID-19 crisis. In 2008 concerns about the financial system itself were the cause of market volatility, and then eventually they reflected themselves in the real economy. 2020 was really different, the pandemic created immediate and deep economic problems, we're still, at the time I'm recording this in late April. Don't know our way out of it or the path out of it, we have thought that nobody knows. Those economic problems in turn caused market volatility. There were radical drawdown in stock prices, widening bonds, and loan spreads. There was insufficient cash lending versus Treasury collateral in the repo market, and whenever the repo market show fragility or walk up, that's a matter of great concern. But it does, it is the case that we got through all of those instabilities, and we don't know what as I mentioned, is going to occur in the real economy as a result of the pandemic. It is possible that dislocations, disruptions, shocks to the real economy will manifest themselves in financial instability. But after the initial intense couple of weeks, we really haven't seen any evidence of that, and it's too early to take a victory lap for the regulators. But I would applaud the regulators for the job they did with respect to the banks and the plumbing of the financial system, by requiring more capital and more liquidity and standardizing processes. They indeed pushed the failure boundary very, very far out. This time around you're little to nothing in the way of concerned about the banks themselves as part of the problem, very different from 2008. In the limit you can make a bank completely safe by requiring the asset side of the balance sheet to the cash and cash equivalents. You could tell banks sell out essentially all of your assets and Federal Reserve deposits. That would totally make a very safe bank, but also make for a completely pointless bank. It would just be a path through for the Federal Reserve, and we need banks to lend and make market and transform maturities. That's what banks do. Take risks and transmit into other forms. Take risks that clients don't want, and transmit into profit manufactured, distributed, and to risks that other clients do want and you can't do that with a perfectly safe balance sheet. The march 2020 dislocations in the Treasury and the repo market actually began in September 2019, and very few people knew about the virus back then. Of course nobody knows but I have a hypothesis which is that going all the way back to September 2019, we were seeing evidence that regulations had veered too far in the direction of safety and this notion of making the banks perfectly safe. There's a particular ratio I won't get into all the details of supplementary leverage ratio that requires banks to hold substantial capital on the order of three percent. Again, activities that almost everybody would regard as completely risky less, holding treasuries on the balance sheet and there's many debates about that rule. But with that ratio as a regulatory minimum prevents banks from performing their functions, especially in extremist and a time of extreme fail risk. Not too much of a surprise to anybody that on March 22nd, 2020, the Fed and the BHCs began discussions about relaxing the Fed's rules including supplementary leverage ratio or the FSLR and a few others. I would take that as evidence that the rules have been calibrated too tightly, especially for an extreme time, and when you want the banks to be there functioning in the repo market, and lending to clients. Will the market-
