Nonbank lending

We provide novel systematic evidence on the extent and terms of direct lending by nonbank financial institutions, and explore whether banks are still special in lending to informationally opaque firms. Analyzing hand-collected data for a random sample of publicly-traded middle-market firms during the 2010-2015 period, we show that nonbank lending is widespread, with 32% of all loans being extended by nonbanks. Nonbank borrowers are less profitable, more levered, and more volatile than bank borrowers. Firms with a small negative EBITDA are 34% more likely to borrow from a nonbank than firms with a small positive EBITDA. While nonbank lenders are less likely to monitor by including financial covenants, they are more likely to align incentives through the use of warrants. Controlling for firm and loan characteristics, nonbank loans carry 190 basis points higher interest rates. Overall, our results provide evidence of market segmentation in the commercial loan market, where bank and nonbank lenders utilize different lending techniques and cater to different types of borrowers.

That is from a new NBER working paper by Sergey Chernenko, Isil Erel, and Robert Prilmeier.


"our results provide evidence of market segmentation in the commercial loan market, bank and nonbank lenders utilize different lending techniques and cater to different types of borrowers."

Not exactly a surprising conclusion.

The successful fintech companies follow the same formula. Find a risk equalized pool of borrowers, offer them a S/L trading pit that is fair traded.

The one who started this was a company devoted to lending to MBA students who formed a very well risk equalized group. Ultimately the lenders and borrowers merge into a single fair traded liquidity pit, execute the abstract tree correctly, and generate a very accurate allocation of liquidity resulting in an accurate velocity equation, and pricing works.

This model is changing slightly, because of Coase. He says let transaction costs go to zero, the liquidity trades become automated and optimum at all times. The automated bank can sell ads and investment opportunities.

Clues in finance are entering brains at a rapid rate, currently. The clues are leading our central bank to a better designed Nixon Shock.

Plus, non-banks don't have FDIC, Fed, OCC, state bank examiners crawling all over them and forcing them to lend to low-to-moderate income borrowers, maintain minimum capital levels, write off defaulted loans, . . .

Are you saying maintaining minimum capital levels and writing off defaults are bad ideas?

By capital your mean equity holdings?

No, they do not help. Holding equity, yours or anothers, generally is of little use in a crash and not part of the S/L business in normal times.
But write off of bad loans is a good idea.

What banks really need to do, to be safe, is to make sure the stream of borrowers does not get out of line from the stream of depositors. They want neither a run on borrowing nor a run on depositing. They just do not want runs. Good shadow banks work that problem and succeed, regulated banks mostly foul up with bad regulations.

Regulated banks (thousands since 1933) mostly foul up by advancing too many loans that don't repay.

I'm saying nonbank lenders don't have bank examiners crawling all over them. They don't have access to FDIC-insured deposits, either.

The following are found in federal regulatory literature/manuals.

The "first line of defense" against asset risks/losses/insolvency are interest and other operating revenues.

The Purposes of Bank Capital

Absorbs losses (cash flows from revenues is the first absorber of losses); promotes public confidence, restricts excessive asset growth (a check on management risk-taking), and provides protection to depositors and the FDIC insurance funds.

Absorbs Losses

Capital allows institutions to continue operating as going concerns during periods when operating losses or other adverse financial results are being experienced.

Promotes Public Confidence

Capital provides a measure of assurance to the public that an institution will continue to provide financial services even when losses have been incurred, thereby helping to maintain confidence in the banking system and minimize liquidity concerns.

Restricts Excessive Asset Growth

Capital supports prudent growth and restrains unjustified expansion of assets by requiring that asset growth be funded by a commensurate amount of additional capital.

Provides Protection To Depositors And The FDIC Insurance Funds

Placing owners at significant risk of loss helps to minimize the potential "moral hazard"/risk-taking, and promotes safe and sound banking practices.

Surprise!! who knew?

This is a consequence of the banking reforms post 2008, including but not limited to Dodd Frank, which have disincentivized banks from making illiquid business loans to be held to maturity and thus have shifted traditional business lending, outside of the mega loans, to nonbanks - business development companies being one of the names for this kind of lender.

My question is a very basic financial one: I don't see how the use of warrants (a type of call option) aligns incentives. With the warrants, the lender has an interest in seeing the borrower's net worth rise, but wouldn't they want that anyway or at least to have a healthy borrower who'll be able to pay back the loan? Moreover it's the borrower's behavior that matters here, how do the warrants encourage them to behave better?

It seems to me that the warrants are more of a way for the borrower to pay lower interest rates -- by giving the lender an option to buy part of the firm. But I don't see how that aligns interests more than a straight loan would.

I don't see it either. It is like the requirement for banks to have available 'capital' (shares on the market). When hit fits the shan then equities fail too.

Granting access to equity through the bond somewhat alleviates risk to the creditor. An extreme example is where the firm borrows funds which are then disbursed as dividends. The firm goes bankrupt; the shareholders enjoy limited liability; and the creditor is left with nothing of value. If the creditor can at least convert to equity, they can recoup some losses during this scenario.

Obviously, you can have covenants that expressly forbid this tactic by the firm, but there's many other tactics that might not be ruled out. So instead of sinking more resources into drafting an air-tight bond covenant, a lender can use the cheaper option of tying the bond with a warrant.

Thanks, makes sense. I would've thought that the strike price would be too high to make that worthwhile for the lender, but obviously I'm still missing something here.

Central banking is a mess of errors.
Shadow banking is 200 lines of code in the Coinbase matching algorithm, very successful. Make the market by setting price such that their is no run for either buyers or sellers.

It is that simple for Fintech. One could download the code and hit go, become a viable shadow bank. The key is to make all lenders and borrowers be from the same risk equalized group.

Central banking if foul, badly designed, mostly about keeping government funded. Shadow banking has no requirements like that and works fine with a few hundred lines of code.

Here is what Fintech does.

If I down load the code, and put up a hundred grand for market making risk then I can run 20 million dollar S/L system, on line. The key is to select, clients from group who understands risk up to a quarter point of their loans or deposits on the trade board. If my client base collapses, I lose a hundred grand, not 20 million.

What I am doing is shifting about 6% of the 20 million in risk to the borrowers and depositors, with no guarantee except the guarantee that everyone is risk equalized,. I need some expenses in helping to track down the occasional scofflaw.

That is it, that is pure S/L banking without worrying about government finances. And it is happening daily, new start ups finding an S/L niche.

I can issue coins via S/L banking. There is the difference between central banking and standard currency banking. I can issue currency using my 300 lines of S/L code, I just set the market making risk to a slight, irregular loss and inform the user base about the losses ex ante so they can bet them. So to go from a standard S/L to an optimum currency banker, I have to change one variable, a number in the standard code. My real problem with currency banking is I need a ledger or a SecureID to run a currency issuer, I do not have Swift to do it for me.

Even if nonbank lenders are more volatile, people still choose to utilize them. I think it has more to do with people not trusting central banks then looking into the actual regulations, or lack there of, when it comes to central banks and nonbanks. Maybe if more information was easier to obtain or more centralized then people would understand the differences and the risk they are taking.

Comments for this post are closed