The Whistleblower Trying to Stop the Next Financial Crisis

One insider says that big banks have been quietly engaging in the same behavior that precipitated the crisis of 2008.

Photo illustration: Soohee Cho/The Intercept, Getty Images


If you were reading the news back in 2008, then you probably remember how residential mortgage backed securities fueled by subprime mortgages tanked the global economy. Now John Flynn, a veteran of the mortgage securities market, says it’s happening all over again — this time in the commercial real estate market. Flynn joins Ryan Grim and The Intercept’s Jon Schwarz to discuss.

Ryan Grim: This is Deconstructed. I’m Ryan Grim.

If you’re listening to this podcast, chances are you’ve seen the movie or read the book “The Big Short.” If you haven’t, the fact that you were alive in 2008 means you know the gist of the story: Wall Street bankers packaged and sold garbage loans that eventually blew up the global economy, ruining millions of lives in the process and fueling the rise of right-wing authoritarian populism around the world.

Other than Teresa Giudice, star of the Bravo TV show “The Real Housewives of New Jersey,” and her husband Joe, basically nobody went to prison for any of it.

Newscaster: Real housewives getting a bitter taste of reality: a judge sentencing both Teresa and Joe Giudice to prison for fraud.

RG: In announcing their indictment, U.S. Attorney Paul Fishman got cute: “Everyone has an obligation to tell the truth when dealing with the courts, paying their taxes and applying for loans or mortgages. That’s reality,” he said.

Pathetic as that joke at their expense was, his claim was also a lie. Everyone did not, in fact, have an obligation to tell the truth when dealing with loans or mortgages. Wall Street executives had gotten fabulously wealthy in the run-up to the financial crisis by lying about those very things, and got bailed out when it all went south. Going after the gaudy Giudices was about as much as federal prosecutors bothered with, catching them fudging some paperwork the same way millions of others did.

[Clip of “The Real Housewives of New Jersey” cursing & yelling.]

RG: After locking Joe away in solitary confinement, they even deported him back to Italy, where he hadn’t lived since he was a little kid. And since nobody but one real housewife and her now ex-husband paid the price, it’s happening again.

Today’s show is a bit more technical than I usually do, but even if you get lost in acronyms here and there, the broad outline of the story is, sadly, quite easy to follow, and it’s a fascinating one.

The one guide I’ll offer is this one: CMBS stands for “commercial mortgage-backed securities.” Those are similar to RMBS, which stands for, yes, “residential mortgage-backed securities.”

During the 2008 crisis, banks bundled together a ton of bad mortgages and created a new asset out of them, and claimed that since they were now all bundled together, they were no longer risky, and the ratings agencies went along. Those new assets are called securities, and they were made up of residential mortgages. Hence the name: residential mortgage-backed securities, or RMBS.

Commercial mortgage-backed securities are exactly what they sound like. They are financial products that are backed not by home loans, but by commercial loans. The problem in 2008 was the bankers and brokers were inflating the incomes of people taking out mortgages.

This week, in a story in The Intercept, a financial analyst-turned-whistleblower provided data that bankers appear now to be doing the same for commercial real estate, inflating income and packaging it together into risky assets.

So, to recap: RMBS are residential securities, the ones you remember from 2008; CMBS are commercial securities, and that’s what we’re talking about today. It’s an abbreviation that you may end up becoming very familiar with in the not-too-distant future.

I’m joined today by Jon Schwarz, my colleague who reported this story with me, and John Flynn, the CMBS analyst-turned-whistleblower who served as our primary source.

What John found, through breathtakingly laborious detective work, is an underbelly of the commercial mortgage market that is a lot softer than we might have thought.

[Musical interlude.]

RG: Well, John Flynn, welcome to Deconstructed.

John Flynn: Thank you.

RG: And Jon Schwarz, thank you for joining us again here on Deconstructed.

Jon Schwarz: Well, it’s great to be here to talk about this extremely complicated but fascinating issue.

RG: It is fascinating. So John Flynn, tell us a little bit about how you got here. How did you wind up at a credit-rating agency to begin with?

JF: Right. My career started in Tokyo, actually.

My clients were Fortune 500 firms looking to enter into the Japanese market. After working there for about four years, I joined GMAC Commercial Mortgage, the predecessor to Berkadia, and that’s when the Japanese banks were all suffering from non-performing loans, bad credits, and we went around with joint ventures, with Morgan Stanley and Goldman Sachs and the likes, purchasing non-performing loans from Japanese banks. And that was really fun! [Laughs.] You know, we purchased everything from goose farms to bowling centers, to golf courses, etc.

I’d been in Japan for about 10 years, and I wanted to get a flavor for how things were done in the West, because I didn’t want to specialize in Japan for the rest of my life and stay there, so I moved to Australia. I met my Australian wife in Tokyo, and we moved to Australia. And after a stint working in development for a CBD tower with Lendlease, I joined Moody’s and their ABS team, asset-backed security team, where I helped to rate everything from credit card receivables to auto loan receivables to commercial real estate to esoteric privatizations, etc.

RG: So what years were you at at Moody’s?

JF: About 2001 to 2002, I think. And during that time, I met the Fitch people in Sydney. And CMBS was taking off in Japan in a big way during that time, and they gave me an offer I couldn’t refuse to go back to Tokyo and help them out on their CMBS ratings. But understand that I already had been in Japan for a long time, and I wanted to get out, and so I got them to relocate us to Chicago, where the CMBS market was on steroids.

Part of my role was to make sure models were consistent across the globe, etc. But I started running deals, CRE CDOs — commercial real estate collateralized debt obligations. I developed that model, actually.

RG: Right. And these are some of the acronyms we got used to during the previous financial crisis: collateralized debt obligations, and the like.

JF: Yes.

I worked on those, but my mainstay was conduit CMBS — conduit commercial mortgage-backed securities. CMBS are groups of about 100 loans put together, diverse portfolios of commercial real estate loans, pooled together and sold off into the market.

And that was in 2003 or 2004 or so. And I saw a lot of things come across my desk that were just crazy. And I took that knowledge, put it in my pocket.

And in early 2006 I joined Dexia. It was Artesia Mortgage at the time, and they were basically the Belgian central bank’s U.S. commercial real estate lending arm. And I joined them to ramp up for a CRE CDO. And we were getting ready to launch that and the financial crash took hold in 2008.

RG: Had you seen it coming? When the bottom started falling out, did you connect it to all of those shaky loans that you had seen coming across your desk?

JF: Yes, I think it’s fair to say I was. In fact, many of my colleagues at Fitch Ratings would remind me that I was frothing at the mouth many times in New York, saying, “this is all gonna end in tears.” And the writing was on the wall in so many ways to those in the industry. And I expected the contagion to carry over to the CMBS market as well. So yeah, I think it’s fair to say that I did see it coming along with a lot of others in the industry.

RG: Mhmm.

JF: But going back to Artesia, or Dexia: I moved there, they relocated me to Belgium to help their ABS team work out that portfolio, and that’s really my first run-in with the U.S. market and its credit issues in underwriting in commercial real estate loans.

As an investor, I looked at what the servicers were doing. On one hand, you had the worst credit underwriting period in history, you could say with a straight face. But on the other hand, you had limited put back for representations and warranties. If a loan is misrepresented, the lender has to sell the loan back because it lied about the loan quality, basically.

RG: So you had a lot of lying, but nobody was —

JF: Calling them out.

RG: Nobody was calling them out on it. Right.

JF: Yeah. And so I would pick up the phone, and get on the phone to the stateside servicers and say, “What the hell’s going on? And they were very responsive. And the same thing was happening in Europe as well. In 2013 and 2014, Dexia was renationalized into the Belgian central bank because of their exposure to ABS. And they said “all foreigners go home” — which I was one. And they were going to send me to New York, but we were going to end up in Australia at some point. So I decided to go straight to Australia.

But I kept my eye on the U.S. market. And I saw that the CMBS borrowers were getting raked through the coals, when, for example, a default occurred on their underlying loan, the servicers would not call out the lender liabilities. Many times I would see 75 percent built buildings being sold as fully complete; I’d see environmental catastrophes on loans being sold undisclosed; halfway-occupied buildings being sold as fully occupied; you know, the issues go on and on. And all of those are misrepresentations.

And so my tactic that made me unique in the market was that I would take these issues and I would say, “Well, look, there’s also lender liability here. And servicer, you have an obligation to call these out. And you have to, in fact. It’s your obligation.” And they hated me for it. They weren’t used to anyone bringing liabilities to the table.

And I got a bunch of successes under my belt. I worked for the first six months at a private equity firm. And then me and two guys took off and carved off CRE loan advisors. And, you know, I was achieving discounted payoffs for borrowers of up to 70 percent of their loan. So, meaning that the borrower, if they had a $10 million loan, they would purchase a loan back for $3 million, and the $7 million would be forgiven.

So I was doing well. But then the servicers got wise to what I was doing and they kind of stiff-armed me and refused to work with me. That was as the market was turning.

The whole market was expecting this whole waterfall of maturities from badly underwritten loans in the 2005-2008 era. As they matured, the whole market was expecting these loans to default upon maturity from 2015 to 2018. Well, they were defaulting and servicers were foreclosing on assets and they were building up these assets on the quote-unquote trust balance sheet to many billions.

But then in 2016 and 2017, they started refinancing these loans en masse into new CMBS 2.0, and CMS 2.0 is basically any loan sold after 2013 under new, better structures — “new” and “better” being kind of tongue-in-cheek there.

And everyone blamed the good performance on, number one, new entrants into the loan market, new non-bank lenders, right? Also low interest rates and a positive economy. I’d remember back to the Fitch days and thinking well, no way — they’d get refinanced. So I dug deeper, and dug deeper, and I basically followed the premise that if the information was difficult to find, then you should go through hell and high water to get it because there’s something there.

Part of the solution that they came up with was that investors should have a one-stop shop to get all their information that they need in CMBS. So that solution was for Wells Fargo to generously operate a website portal that would make all this information available to investors and servicers and rating agencies and analysts. And that website is

So the big data feeds that come through there and feed into the models until the waterfall of the CMBS structures, which are very complex, those automatic feeds were taken from CTSLink, and then put into the models, which, by the way, are now totally black boxes. It used to be that you had to model out the waterfall, but now they generously do that for you so, as an investor, you don’t even have to worry about the waterfall because that’s all modeled out for you, of course, by the banks or by the loan sellers.

RG: Wells Fargo. What could possibly go wrong?

JF: [Laughs.] I know!

[Musical interlude.]

RG: Jon Schwarz, you’ve done a lot of reporting on Wells Fargo in the past. And both you and I have been talking to John Flynn a lot over the past year, sussing out what he found and fitting it into the historical context of the financial crisis. And what parallels did you find?

JS: Well, if people think back to what was going on in 2008, if you’re following the news then, it was this incredibly confusing swirl of acronyms and numbers and so on. But, in fact, what was happening was not complicated at all.

The thing about Wall Street is: It really only has about four basic scams, and it just continually runs them in different combinations and in different ways. And the particular scam of the housing bubble in the 2000s was counterfeiting.

So everybody knows regular counterfeiting is you just print something that looks like money, but it has no actual value. What Wall Street was doing in the 2000s was they were printing another form of paper wealth, bonds, that had some value but much, much less than they purportedly did on their face.

So to understand this, you have to think about regular mortgages in the past and how mortgages began to be issued more recently, including in the 2000s. So people think, “I want to buy a house, I go to a bank, they give me a mortgage — they loan me money — and then I have to pay that bank back over a period of 30 years.”

Now, in the past, if you think of like, “It’s a Wonderful Life” —

Jimmy Stewart [as George Bailey]: I’m in trouble, Mr. Potter. I’ve got to raise $8,000, immediately.

JS: — the bank there, when it was making loans, cared whether they could pay that loan back.

Lionel Barrymore [as Mr. Potter]: What kind of security would I have, George? You’ve got any stocks.

Jimmy Stewart: No sir.

LB: Bonds? Real estate? Collateral of any kind?

Jimmy Stewart: I have some life insurance.

LB: How much is your equity in it?

JS: If you loan someone money, you want them to be able to pay you back.

But what happened in the past couple of decades was the development of something called securitization with mortgage loans. And so what that means is one bank would issue the loan, and you’d get the money, you’d buy the house, but they would then take that loan and sell it off to someone else. So the incentive to make sure that the person could pay the loan back dropped significantly. [Laughs.]

RG: [Laughs.]

JS: And they would sell these loans off, usually to another bank, then banks would take these loans — say 50, 60, 70 loans — package them all together into something called a trust and then the trust would issue bonds based on the loans that were part of this pool.

Now, if in fact those mortgages had been correctly underwritten, if people could actually pay these loans back, then that’s a great deal. You buy one of these bonds, and then you get regular payments over a period of years.

But what happened in the 2000s was the development of something called “liar loans,” where banks no longer cared about whether a loan actually was correctly underwritten. And so they had a great incentive to just collect the fees for originating the loan, so they wanted to issue as many loans as they possibly could. So they would find lots and lots of people who actually could not afford to pay this money back. They would find someone who’s a bartender in Florida making $25,000 a year and the bank would claim, in the paperwork, that they were actually making $300,000 a year, so they could afford to buy a house that cost $1.5 million.

And so all of these bad loans would go into these pools, into these trusts, and then the bonds that were issued by these trusts were essentially counterfeit. And that is the thing that turned it into such a catastrophe in the 2000s, because you would have pension funds — that really do need regular income — buying these bonds. And when the whole thing collapsed, both sides were ruined: you would have the people who had taken the loans out to buy homes get thrown out of their homes and the pension funds would be not getting the payments that they had counted on. The only people who profited were the middlemen, the people who had done the packaging; Countrywide was notorious for this. They would have made hundreds of millions of dollars on all of these packaging fees.

And so that’s what happened then. What seems to be happening now is potentially something not in the residential mortgage-backed securities market, but in the commercial real estate-backed securities market. So that means loans that are based on mortgages for shopping malls, or hotels, or things like that.

RG: Right. So John Flynn, as you start digging deeper into the underlying documents, behind the CMBS loans, you start to find something that looked a little bit similar to what John Schwarz just described. When did you start to realize that you were onto something fairly huge?

JF: It was in late 2017, basically. I started out: “OK, I’m just gonna do 10 of these and I’ll see where it goes.” And all ten of them were inflated. I’m like, “What?” And so I kept going from there, basically.

RG: And before you go further on that I have been blown away by how hard it was for you to find what you found. You shared with us all of the different documents. And you walked us through how to confirm everything that you were saying. And even with all of those documents already in front of us, and the path, the roadmap, of how to get there, it took an enormous amount of work because of the changes of address, the changes in the names of these loans.

So explain to people what they would have to do if they want to do the same thing that you did?

JF: Well, the data feeds that I was referring to before, none of the comparable data — to compare two sets of numbers, you need two sets of numbers, right?

RG: Mhmm.

JF: So none of the comparable data is in the data feed.

RG: Which is wild, because wouldn’t investors want this comparable data?

JF: Well, yeah. An investor that is curious, certainly they would want that data. But again, they’re managing billions and billions. And do they have the incentive? And would they do anything about it?

I don’t think the average investor would rattle the cage of a bank who is doing this because, number one, they wouldn’t see any deal flow because the wealth-management arms are closely tied to the same large banks sponsoring the deals, so they don’t really have an incentive to do it. And plus, the lawyers for the banks got wise to the kind of tactics that they used for RMBS litigation, and they patched up or they kind of rigged the decisions in the courts, so that it’s a very difficult path to walk to actually bring a case to either put back a loan or to do otherwise — change of service or whatever.

RG: So there’s no point.

JF: Yeah. There’s no point.

RG: So again, back to the steps on how to do it, the building name and address changes is something that I was aware of earlier when I started representing borrowers. The first five minutes of the conversation would be to establish: “Are we talking about the same building?” The borrower would call it “building XYZ” and the bank would call it “building ABC.”

And sometimes the building location would change! I mean, it would actually change the physical location of the building, if it’s the same collateral. So even if an investor was curious, it would be very difficult for them to do it.

RG: Right. So these banks, when they’re refinancing these properties, they’re changing the address and they’re changing the name of the property. So any attempt to kind of compare the numbers that they were using to finance it the first time around, to compare that to the numbers they’re using to finance it this second time around, it’s not entirely impossible, but it would take somebody an awful lot of time. Am I right?

JF: Yes. And it raises a lot of doubt. You have to be certain when you’re doing these comparisons that “building A” is “building A.”

And there’s many legitimate reasons for changing a building name, right? And there are some legitimate reasons for changing an address. But not on the systematic scale that I see in the data.

RG: And so, as you would compare the numbers from one pile of paperwork to the next pile of paperwork, even though you’d find it was representing the same year, you kept finding the income inflated? Is that right?

JF: Yes. In short.

RG: And so what else could account for that, other than the obvious?

JF: Well, the main thing that people put forth is adjustment for capital expenditures, and for tenant improvements and leasing commissions, or TI/LC.

RG: Mhmm.

JF: And that takes place when an underwriter is estimating the income for a building, the net income for a building, there are expenses that have to do with the integral structure of the building. And those are capital expenses.

There’s also an operating expenses. So what one does when they underwrite a loan is they split out the operating expenses from the capital expenses, and make adjustments in the expenses.

And so that’s what they put forth as an explanation. But, in both cases, in the old and the new trusts, the underwriters were normalizing expenses. So that wasn’t an explanation. And they were following industry guidelines in doing so. And, in many cases, the different cash flows that were being reported were made simultaneously as the loan was being sold into a new trust. So the reporting in 2015 was done, maybe in December of 2015 and a different loan for the same collateral was being sold in January of 2016. So this reporting is only like months old, right?

RG: Right.

JF: So how does it change so drastically in that month? And it shouldn’t change at all, because you’re talking about the same time frame, same collateral, same circumstances.

RG: Right.

JF: So it shouldn’t change at all

RG: Right. And so you found that often these numbers were different. How often was it that the error was inflating the amount of income? Just like how John Schwarz talked about the bartender who said that he or she was making $300,000 a year. How often was the number inflated? And how often did you find it deflated, where they kind of stated the income lower than it was before?

JF: It was inflated over 90 percent, maybe over 95 percent of the time. And I never went back and measured it because this is very laborious work.

RG: So what’s the function of that? What’s the benefit to the lender?

JF: So the liar loans in RMBS, the credit driver for RMBS is the borrower’s credit, the borrower’s income. And the credit driver for CMBS is cash flow. Cash flow is king, right?

And so if you inflate the cash flow, you — of course — get a higher income that increases what they call the debt service coverage ratio, which is the NOI — the net operating income — and there’s thresholds under regulatory guidelines that loans have to meet to be sold into a pool.

And so the benefit is, one, that they get the loan sold, that is an otherwise unsalable loan; benefit number two is it can load on more debt, because with that NOI increase, it also increases the value. Number three, is they show a less volatile cash-flow profile of the underlying asset; less volatility is a higher credit quality, right?

RG: Right. So there’s an extraordinary amount of benefit to the lender to be able to inflate the cash flow that their borrower claimed to have.

So John Schwarz, where do you see this going from here?

JS: It’s tough to predict. This is a situation, a huge financial crisis, where you would expect any kind of fraud to become apparent. There’s a very famous Warren Buffett quote where he says, “When the tide goes out, you find out who’s been swimming naked.” And that is the problem that happened in 2008, is that when the economy collapsed, everything that had been done in the residential mortgage-backed security market became apparent.

It’s happening to some significant degree now. We have some graphs and charts in our article that people can consult where the companies, shopping malls, hotels are having a very, very hard time paying back the mortgages that they had taken out on the properties. But we’ll just have to wait and see if it’s as big a catastrophe as it was in 2008, or if it’s something different.

RG: And John Flynn, you know this industry well. Where do you think it’s headed?

The Federal Reserve is playing an interesting role, because it seems like there could be a significant amount of economic downturn and fraud combined that if the Fed soaks enough of it up it can just keep kind of limping along. As Keynes said, “We’re all dead in the long term.” So it could limp along in the medium term, or in the short term. What’s your read on how much trouble the industry is in?

JF: Well, look, we’re in a low-interest-rate environment right now. Right? So the hunt for yield is on. Any money out there needs to be placed to earn a yield — any kind of yield. And CMBS still offers a reasonable yield as compared to treasuries. And so the demand will still be there, and if there’s no accountability in this low-interest-rate environment, all they need to do is inflate numbers again.

So to get these troubled loans refinanced, people say, “Oh, how are they going to get refinanced?” Well they could just create value.

RG: Same way they just did.

JF: Yeah. Why wouldn’t you?

RG: Well, to that question: Why wouldn’t you?

Jon Schwarz, what is your read on what the Biden administration could do about this?

JS: Well, if indeed, this was fraud on the part of any of the players, generally speaking, that is illegal. If the Biden administration wants to actually enforce the law in a way that the Obama administration did not in 2009, then they have many, many tools to do so.

And, in fact, it is happening already that the SEC is getting involved in the kind of litigation and enforcement that you would expect to see in a situation like this. The FCC just filed a lawsuit against a credit rating agency for manipulating its CMBS ratings methodology. Another credit rating agency has already paid $2 million in fines to settle an SEC suit. So the government can put a stop to it; they can hold the people responsible. The issue is really one of the political will rather than some sort of abstract capacity. They can do it if they want.

JF: Let me interject there: everyone points at the credit rating agencies as the culprit, but they’re easy targets because they can’t fight back, right? They don’t suffer any damage or blowback from the credit rating agencies because they’re not the banks, they won’t suffer in deal flow, etc. But the real culprits are the ones doing the inflating.

RG: Would rating agencies have to be fundamentally differently structured to be able to catch what you found? My read is that this isn’t exactly them asleep at the switch, it’s that they don’t actually even have somebody stationed at this particular switch, because they kind of take lenders’ word to a certain degree and then look at things above that. They’re not doing the laborious work that you went through. Is that right?

JF: No, they had to take the lender on their word, right? The issuer pays model is inherently open to that, open to abuse Maybe take a page out of China’s book. You have to do some enforcement on it, basically. But again, it’s only their opinion. And they say that very loud and clear. It’s only their opinion.

RG: Right. Well, John Flynn, thank you so much for walking us through this.

JF: No, thank you for having me. Appreciate it.

RG: And John Schwarz, thank you for collaborating with me on this story. This was something to get our hands around, but I hope that people start to pay attention to it.

JS: Yeah, I hope people can take a look at this. It is the kind of thing that it takes a little while to understand but once you do understand it, it really is incredibly interesting.

RG: Thanks so much, guys.

JF: Thank you. Bye.

[End credits music.]

RG: That was John Flynn and Jon Schwarz. You can read the story that Jon Schwarz and I wrote, “The Bigger Short,” at And a big thank you to everybody who became a member of Deconstructed last week; the promotion did so well, we’re extending it through the end of the month. If you go to and make a donation of any amount, you’ll get a signed copy of my book, “We’ve Got People.”

Deconstructed is a production of First Look Media and The Intercept. Our producer is Zach Young. Laura Flynn is our supervising producer. The show was mixed by Bryan Pugh. Our theme music was composed by Bart Warshaw. Betsy Reed is The Intercept’s editor in chief.

And I’m Ryan Grim, D.C. bureau chief of The Intercept.

If you haven’t already, please subscribe to the show so you can hear it every week. If you’re subscribed already, please do leave us a rating or review — it helps people find the show. And if you want to give us feedback, email us at

Thanks so much, and see you next week.

Join The Conversation