Saturday, October 24, 2015

Liquidity liquidity everywhere but not a drop to drink

One of Gustave Doré's illustrations of The Rime of the Ancient Mariner, plate 4

The minsicule bid ask spreads we see in financial markets today indicate that stocks and bonds have never been more liquid. At the same time, skeptics worry that the odds of a sudden evaporation of this liquidity has never been higher. This Jekyll and Hyde world of ultra liquidity coupled with heightened risk of liquidity famines is one of the core themes running through a great series of posts on market liquidity from Liberty Street, the NY Fed's blog. See here, here, and here.

To protect their portfolios, investors need to be able to look beyond the incredible amounts of potentially superficial liquidity coursing through markets and plan for future illiquidity crisis. For this sort of preparation to be possible, what investors really need is a market in long-dated liquidity-related financial products.

Central banks have historically been the chief providers of liquidity-related financial products, namely liquidity insurance, or the guaranteed use of central bank lending facilities in a crisis. The problem, as Stephen Cecchetti and Kermit Schoenholtz point out, is that we simply don't know if central banks are offering this financial product at the right price and in appropriate quantities. Cecchetti & Schoenholtz say that providing lending facility access: 
without limit and without penalty can lead to enormous moral hazard, causing overreliance on the central bank. If market participants are trained to ignore liquidity risk in good times, they will do little to make markets less fragile or to prepare themselves for unanticipated, but persistent episodes of market illiquidity.
The other problem is that central banks only provide liquidity insurance to banks. What about the rest of us? How can all investors, and not just bankers, benefit from properly priced liquidity insurance products?


A central bank is a monopolist without much business sense. It has no idea how to set the proper price for insurance products. Which is why I think a private market in liquidity options, or liquidity insurance, may be an ideal solution.

How might these liquidity options look?

An option to sell at the ask price

In financial markets there is a price at which participants are willing to buy and a price at which they are willing to sell. This is the famous bid-ask spread, or bid-offer spread.

Say the bid, or buying price, for Google shares is $650 and the ask price is $660, the width of the spread being $10.

A seller with time on their hands will join the queue of sellers already at $660 and wait for a buyer to step forward. If the seller is desperate for liquidity, they will sell at $650 to the first buyer in the bid queue, absorbing the $10 loss. When a liquidity crisis hits, this spread may widen out such that a desperate seller will only be able to get out at $640, or $600.

Liquidity risk can be thought of as thusly: We'd all love to rapidly sell at the offer, or ask price, but reality forces us to trek the distance across the spread and sell at the bid. In normal times, who cares; the spread is miniscule. But this trek-to-the-bid gets much costlier in a crisis when spreads widen out.

A liquidity option would be designed to allow investors to do the impossible: sell rapidly at the offer price. Using the Google example from above, an investor who buys a Google liquidity option would be allowed to exercise the option and immediately buy Google from the option seller at the current market offer price, say $660, and not the bid price, say $650. If buyers were to flee and liquidity evaporate such that the level of bids falls to $600, the owner of the option is protected since they can sell Google at the offer price of $660, and not the much lower bid price of $600.

Think about it this way. Whereas a regular put option provides downside price protection by allowing the owner to sell at a fixed price, a liquidity option allows them to sell at a fixed spread. Buying liquidity protection for one Google share would probably be much cheaper than getting downside price protection for that same share.

What should the price of a liquidity option be? Say that the difference between Google's bid and ask is typically $10. An insurance writer who is is required to purchase Google from the option owner at the offer price can typically only offload this risk by turning around and selling Google at a $10 loss. They will therefore require an initial insurance payment, or premium, of at least $10 to compensate.

When a liquidity crisis hits the current bid-ask spread will widen and insurers will ask for higher premiums on newly issued insurance. If current liquidity stays healthy but the odds of future liquidity crisis increase such that future Google bid-ask spreads are expected to be quite wide, then the liquidity insurance writer will require more compensation as well. The value of a liquidity option depends on both current and expected illiquidity. Conversely, if liquidity risks are expected to decline, buyers will ask for lower premiums since they don't expect the insurance to offer much protection over its contract life.

Those investors who have hedged against liquidity risk by buying liquidity options need never fear illiquidity again. If liquidity stays healthy their liquidity options will expire worthless but they'll have no problems exiting their positions. If liquidity deteriorates they can no longer exit their positions directly by selling on the market but can just as easily get liquid by exercising their options.

In addition to Google, a well designed liquidity market would have liquidity options on all major equities, ETFs, and widely traded fixed income products. Full democratization of liquidity insurance would be achieved by having these options trade on public markets. Information about the price of liquidity would become widely available so that investors could "internalize liquidity risk", as Cecchetti & Schoenholtz put it. If they didn't like the risks they found themselves facing, investors could use these products to reorient themselves. Take a family with a mortgage that was too afraid to buy Google because of the potential for an outbreak of illiquidity at the same time that a mortgage payment comes due. The can now own shares and hedge away their liquidity risk by purchasing a liquidity option. Folks like Warren Buffett, a conservative investor with a strong balance sheet capable of withstanding liquidity crisis, would be able to earn extra income by writing liquidity options and collecting premia.

In sum, with a well designed liquidity options market, the risks of illiquidity are distributed to those who want to bear them and away from those who don't. Markets will probably be much less fragile. As for central banks, with the market providing both liquidity insurance and liquidity pricing, central bankers can focus much more on what they should be doing; monetary policy.


  1. Really interesting proposal.

    Would it be gameable/riggable?? Could I buy a liquidity option, then make a silly high buy offer? (I don't think so, because the liquidity provider could then just turn around and sell them me back, but I'm not sure.)

    1. Rigging the bid ask spread could be a problem. Someone who has written a bunch of liquidity options may try to artificially narrow the bid ask spread by submitting a stream of bids and offers so that the options they've written lose value, allowing them to buy back the insurance at a cheaper price.

      However, they run the risk of being filled on the orders they are using to manipulate the spread, so I'm not sure how aggressive they can be on gaming the market.

    2. I do not know. Streaming inside bid-offer should be unprofitable business. That is IMO market also, as market moves a lot there is a probability one will be left with a short stick. And even if-then the price of the option should be reflect the riggability, rendering it worthless strategy.

      But I'm not sure how you measure bid-offer at the time of the execution? With exchange traded stuff it might work, until illiquidity. Also most markets are working over-the-counter (OTC).

  2. JP,
    Liquidity is a systemic (network) dynamic. Can a "part" within the network offer insurance against it?

    In other words, the put writer would have two choices. One, hold cash reserves equal to the potential drawdown caused by put exercises during a sudden liquidity event. Two, maintain lines of credit so that they could cover such an event. I would argue that the first option is probably not a viable business model, and that the second does not protect liquidity buyers against the bankruptcy of the put writer caused by lines of credit being withdrawn.

    Something similar to this happened with AIG in '08. As put writers on subprime risk spreads that ballooned in part due to illiquidity, they experienced a liquidity crisis.

    1. Good points.

      I'm talking about standardized exchange-traded products protected by margin held at a clearinghouse, different from the OTC stuff AIG was writing. Buyers of liquidity insurance would know that they have a good counter-party.

  3. It would be interesting to actually look at data on this. I expect it would parallel volatility.

  4. In my illiquid experience, there is no such thing as the ask. The ask price is typically the pipe dream of asset holders in a market for lemons. In reality, there is only the bid. The bid is the market of millions, the ask is just a handful of guys. Ironically, you are trying to buy at the illiquid price to solve the liquidity problem.