Yves Smith packs a powerful insight into an unassuming sentence:

Liquidity is not a virtue in and of itself unless it produces a benefit to the real economy.

Liquidity is often said to be the great lubricant of financial markets. Let's go with that metaphor for a moment. Yeah, baby, liquidity. It's high performance motor oil that turns hard metal to smooth silk and keeps the engine of capitalism firing on all cylinders! Pop the hood and pour that stuff in. Rub it onto the gears and axles, so nothing ever squeals, pops, or (God forbid) grinds to a halt. Slather it all over the tires, so that no friction comes between our purring metal machine and the sweet American road.

Ummm, wait a minute... Putting lubricant on the tires might not be such a great idea after all. Friction is precisely what tires need to do their jobs. Throw a lot of oil on the tires and, well, something bad might happen.

Similarly, in financial markets, we want liquidity at some times and in some places. But there are times and places where we want, even need (gasp!) illiquidity!

Illiquidity. That word is so ugly. What might be another word for the same phenomenon? How about "commitment"? When a person invests in something that is not very liquid, they are committed. They are necessarily betting on its fundamental value. Liquid securities can be bought or sold as a trend or a trade or a play for a greater fool. But if the thing you are buying can only be sold with a big haircut, you'd better hope for a really gigantic fool if you have no confidence in its underlying value. (Clever managers did find ways around this problem, but let's put principal/agent issues aside for the moment.) When financial markets are too liquid, everything looks like cash. Superfluous distinctions — like the economic meaning of the assets bought or sold — fall by the wayside. Sure, investors always prefer liquidity to illiquidity. An option to buy or sell quickly and cheaply is preferable to an option to buy or sell slowly and with large transaction costs. But just because investors like something doesn't mean that it's good. Investors like rainbows and ice cream and free money from taxpayers. But the rest of us prefer that investors make serious, informed decisions about what is and isn't of value, and that they be paid for evaluating and actually bearing risk, rather than artfully shifting it (or whining when it cannot be shifted, because omigawd-there-is-no-liquidity!). Of course there is a balance here, commitment is one thing but a ball and chain is another, if assets become too hard to buy or sell, the costs of financing genuinely useful enterprises would increase until even good risks are not borne at all. It's not that liquidity is a bad thing. It's a good thing of which there can too much.

But how much? Another word that should be attached to any conversation about liquidity is "accuracy". There is, in some sense, a "right" level of liquidity, defined by the uncertainty surrounding the present value of an assets future payoffs. We laud markets for "price discovery", their ability to distill complex economic facts into simple prices that put a value to unknowable future events. But we need markets to communicate the uncertainty surrounding those valuations as well. The depth-weighted spreads of assets whose values are nearly certain should be much narrower than those of assets whose payoffs cannot be accurately predicted. When that is not the case, it represents a market failure. The recently wide spreads on complex structured credits are not the crisis — those spreads accurately reflect the uncertainty surrounding what the instruments are actually worth. Nobody knows, so spreads should be wide. The real crisis was two years ago, when "oceans of liquidity" meant that whatever the underlying value of a thing, you could sell it quickly for near what you bought it, so spreads grew artificially narrow. Confidence is good only when confidence is merited. We need not only accurate prices, but accurate confidence intervals, accurate spreads, accurate levels of liquidity rather than simply more, more, more.

Steve Randy Waldman — Monday May 5, 2008 at 5:15am permalink
E. Cartman (mail) (www):
How about another word for liquidity: "trust".

Trust is not in and of itself a virtue. Economic booms produce misallocations of trust between capital participants.

Liquidity strikes are statements of distrust from participants to a credit originator.

Fed repo barrages do nothing to rectify misallocations of trust; they only keep all participants in unsustainable suspense until a major new piece of negative marginal information sunders the dam once again.

Anything dependent upon an economy-scale bailout to survive is, in the long run, unsustainable. The aggregate of bailouts for the credit industry ($420bn-$1.1trn from the GSEs, $150bn in Congress-mandated stimulus, $250bn or so in outstanding repos from Fed facilities, $350-400bn in Fed Treasuries swapped for garbage MBS) is catastrophically unsustainable.

It is a matter of when, not if the next liquidity drought (collapse of intramarket trust) occurs.
5.5.2008 9:04am
RueTheDay:
You are asking some very good questions here. Why is it that financial markets, which probably resemble textbook models of perfect competition more than any other existing markets, require all of this "liquidity" in order to function efficiently.

In his book, Wall Street Capitalism: The Theory of the Bondholding Class", economist E. Ray Canterbery notes that secondary market transactions account for 95% of all financial market transactions. Thus only 5% of financial market transactions result in corporations raising financial capital. However, the price levels of asset markets do have profound effects on the real economy. Over time, a country's monetary policy becomes more and more focused on keeping asset markets operating in an orderly fashion and less and less on output growth, inflation, and unemployment. Sound familiar? The book was written 8 years ago. The problem is reconciling the need for the liquidity that is provided by the secondary market in order to have a functioning primary market with the potential damage caused by the inflation and deflation of asset price bubbles. The solution Canterbery offers is a Tobin-like tax on all financial trades, designed to minimize speculation while not interfering with productive investment.

James Tobin himself said the following on the Tobin Tax:

"The idea is very simple: at each exchange of a currency into another a small tax would be levied - let's say, 0.1% of the volume of the transaction. This dissuades speculators as many investors invest their money in foreign exchange on a very short-term basis. If this money is suddenly withdrawn, countries have to drastically increase interest rates for their currency to still be attractive. But high interest is often disastrous for a national economy...My tax would return some margin of manoeuvre to issuing banks in small countries and would be a measure of opposition to the dictate of the financial markets."

The original intent of the tax was focused on foreign currency market transactions, but are quite appropriate to financial markets in general.

One also can't help but notice the number of times Hyman Minsky's name seems to be coming up in the blogosphere and the financial press as of late. It's high time for his ideas to be taken seriously, both by the academic economic community and by the financial regulatory authorities. All of the empirical evidence from hundreds of years of market activities validates his model. The primary reason he hasn't received more attention from the academic community is that it's impossible to fit his (inherently dynamic) processes into the standard comparative static neoclassical general equilibrium models. Tough. His models are actually relevant for real world decision makers and regulators, the standard neoclassical models are largely nonsense with zero relevance for real world economies.

Any sort of reform that is undertaken needs to focus primarily on limiting the damage that financial market speculation can have on the real economy. "Building firewalls" as I believe you stated in a previous post. A generic Tobin-esque Tax can be a part of that. As I indicated in a post on a previous topic here, the other substantial parts should be re-instating Glass-Steagall to separate commercial banking from investment banking, and applying capital adequacy standards to significant non-bank financial market entities. The reforms must be structural in nature. Unfortunately, I don't see it happening, rather, what I see coming from Congress is a patchwork of nonsense - asking the GSEs to relax their standards and back more mortgages, tax breaks for homebuilders and buyers of foreclosed houses, taxpayer funding for credit counseling services, etc.
5.5.2008 9:13am
RueTheDay:
One more thing.

We need to revisit the concept of "too big to fail".

One commonly argued viewpoint is that the size of the positions held determines whether or not the entity is too big to fail. Forcing the entity to quickly unwind large positions in equities/bonds/commodities/etc would be disruptive to those markets. I disagree. Let the parties involved lose their shirts. The underlying markets will recover.

The introduction of leverage changes things; the introduction of bank-financed leverage rewrites the entire scenario. Banks occupy a unique position in not only the financial market, but the economy overall. I realize some consider it passe to harp on the distinctions between banks and non-banks, but the distinctions are very real. Only banks can create money ex nihilo, because only banks can create deposits. Likewise, the banking system, by providing the nation's primary payments system, links together all economic agents, both bank and non-bank. One failed bank can cause effects that quickly ripple throughout the entire banking system and quickly spill over into the non-financial sector, including firms that did not have any direct dealings with the failed bank.

There is a lot of buzz around "counterparty risk". However, if the agents involved are not banks, the extent of damage is inherently limited. A "pure" hedge fund failing or a "pure" investment bank failing will likely not cause much collateral damage beyond other firms that had direct dealings with them, and agents absolutely need to take this risk into account when determining which firms to deal with. Pure banks, on the other hand, being at the nexus of the payments sytem, and creating money as part of their operations, are the principal source of systemic risk.
5.5.2008 10:05am
E. Cartman (mail) (www):
Rue: taxing the foreign exchange market is exactly what one can expect from a foolhardy academic such as Bernanke, who can debate theoretical arcana with anyone, yet is utterly ignorant of the underlying reality. Foreign exchange markets will always have havens.

Bretton Woods II evolved in *spite* of governmental intervention, when Japan and Germany ran massive current account surpluses, and the US ran a massive twin deficit under Nixon.

The only genuine "firewall" is punishment of the most overextended speculators. Unfortunately, Bernanke's combination of slashing interest rates, dilution of AAA Treasuries for MBS garbage, bailing out BSC and thus also bailing out JPMC's $78 trillion credit default swaps book (which would have been blown up if JPMC had had to make good on all the CDS it wrote for BSC and other highly levered companies which also would have blown up in the event of a BSC/Lehman detonation), and spiralling GSE leverage, have done exactly the opposite.

In fact, the BSC bailout and subsequent "BS Bernanke printing press put" amounted to a nationalization of the CDS market. If the nerve centers of the financial system are backed up by Fed presses, overall systemic risk is massively reduced -- at the price of much higher future inflation.

This is not a liquidity issue, or even a solvency issue, considering how poor everyone's solvency information is. It's a trust issue. There has been way too much intramarket trust, especially within securitized debt. The Japanification of the US economy rewards aged property holders by institutionalizing higher prices. Anyone who already held (too much) property because of (too much) trust in the leverage system has been lavishly rewarded.

Liquidity is no more an unqualified network effect good than is trust. Everything in life has network effects and feedback loops. Saying the market is entitled to maximum liquidity because of feedback-loop scaremongering just obfuscating the real issue, which is rewriting market rules, especially CDS rules (why anyone trades CDS anymore is beyond me; then again so are a lot of things), and transferring wealth from hedge funds which generally calculated correctly, to institutions whose time horizons were shortened by leverage, and which couldn't deal with the consequences.

Call it the freebase free lunch free market...
5.5.2008 11:02am
byrneseyeview (mail) (www):
You seem to be looking at liquidity only from the perspective of someone who is long an asset. But I can think of plenty of folks who still value illiquidity -- KKR is probably thrilled that nobody can call them up and demand 100 cents on the dollar right this millisecond the way they could with a publicly-traded stock. Viewed from that perspective, it's obvious that there's a market in illiquidity, too: that what an investor wants is to borrow illiquidly, and buy something liquid, at minimal cost. Usually, this fails and people end up with a 'matched book' -- KKR uses illiquid debt to snap up illiquid assets, while a stat arb fund would use high-liquidity debt (e.g. margin) to buy extremely liquid stocks.

Instead of looking at total liquidity or average liquidity, I think you get a much clearer view by looking at bilateral liquidity -- and since all financial transactions are transactions between counterparties, that's the only way that really makes sense.
5.5.2008 12:44pm
Steve Randy Waldman (mail) (www):
byrneseyeview — i'm feeling slow here, but can you unpack your comment a bit? i'm just not getting it. i get that KKR exploits opportunities that wouldn't present themselves if their targets were more liquid — financial strength and a tolerance for illiquidity can be a comparative advantage — but i don't think that's what you're talking about here.

i did write the piece in terms of longs, but as i've written it, i could flip it to the short side as well. the liquidity i described was an "emergent" option to buy or sell quickly/cheaply with no one actually obligated to provide that liquidity. it's like the weather of the market (and indeed the weather can change!). you seem to be talking about the experience of someone short an explicit obligation to provide liquidity, e.g. someone short a put option. to the degree that one accepts an obligation to provide liquidity, having a "matched book", something liquid i can sell or put myself in order to fulfil seems like natural risk management.

liquidity is perhaps the least well defined term widely used in finance, and i think we're talking past a bit, as your "bilateral liquidity" represents committed arrangements my counterparties to convert to/from cash under contractually agreed terms, whereas i'm talking about offers and expectations regarding as yet unconsummated exchanges and statistical properties of groups of deals that do get consummated. but i am interested in understanding this: "what an investor wants is to borrow illiquidly, and buy something liquid, at minimal cost." can you explain?
5.6.2008 12:31am
Steve Randy Waldman (mail) (www):
RTD — the reference to the Tobin Tax is very a propos. It's a blunt instrument, but it is a way that degrees of commitment could be reguated into financial transactions, if we decide that's desirable. As you say, fixed transaction taxes could apply to any financial instrument (and to domestic securities markets much more easily than international currency markets). China does have its famous stamp tax, but China's markets are so Wild West East, I'm not sure I'm confident we can learn much from the experience. Setting a meaningful Tobin Tax is challenging... it'd have to scale with the volatility of the asset to be uniformly meaningful. There may be other approaches to achieving the same end, such as introducing some uncertainty into the timing of settlement, but it'd take some work to reduce vague ideas like that to workable reforms.

I think the question of banks vs nonbanks is more complicated than you suggest. Anyone can "create money ex nihilo", really. It's just that what banks create is taxpayer-insured legal tender, whereas other people's money can disappear with a broken promise. What we've learned, or chosen (perhaps foolishly), is that nonbank money is also effectively insured legal tender, if it is deemed (by whom?) that its disappearance would be disruptive.
5.6.2008 12:43am
Steve Randy Waldman (mail) (www):
E. Cartman — With regard to credit issues, cash-flow certainty and trust are nearly the same thing, so in credit markets I largely agree with your characterization. With equity, one can be very trustworthy and still fail to pay good returns, as projects are inherently uncertain and stockholders agree to bear residual risk. So, with equities (and to some degree with credit too), distrust=illiquidity for sure, but trustworthy does not necessarily map to liquidity.

I think you're right that our recent "oceans of liquidity" experience was basically a matter of too much trust, touted as a wonderful thing even when there was little evidence the trust was well founded. I also agree that it's a bit perverse that the people who took care to price trust properly have been screwed by acts of public policy that try to make scoundrels and fools trustworthy in retrospect, because it would be too unpleasant to let them welch. It might be possible to price likelihoods that a particular counterparty will default, but pricing the intersection of counterparty default and absence of political bailout is an order of magnitude more complicated, and just yucky. Participating in a CDS on anything remotely TBTF or "interlinked" is now largely a matter of gambling on government behavior, and that's a bad thing. I would have preferred that we simply permit things to fail, and save our extraordinary interventions for managing the post-default wind-downs. But for reasons good and ill, that's not the road that's been chosen for us.
5.6.2008 12:59am
Ancient Mariner (mail):
"Liquidity" needs to broken down into its various parts, and then either be reassembled, or liberated from the meanings being attached to it.
5.6.2008 12:12pm
JKH:
A Brief Essay

Liquidity is a multi-faceted risk characteristic. It can apply in various ways to financial instruments, markets, and institutions. Its definition is open-sourced. And it is difficult to disengage the aspect of liquidity risk from other instrument specific risks such as interest rate, foreign exchange, credit and equity risks. Liquidity risk is distributed across all of these categories, in both cash markets and derivative markets.

The interaction between commercial banking and investment banking has been a powerful force in the development of the world's liquidity profile. Before the days of the first interest rate swap or securitized mortgages, the assessment of a commercial bank's liquidity profile was straightforward enough. Short term assets and long term liabilities were both good. Marketable high quality short term assets were very good. Equity was exceptionally good. These attributes delayed the day of reckoning if a bank ran into problems. Liquidity in this context was judged mostly on the basis of the maturity profiles of a vast collection of non-liquid assets and funding. Investment banks were not completely different, but their assets were on average more liquid than commercial banks and they tended to run very short term liabilities. They could back into their commercial bank clearing agents for what was essentially an overdraft if they ran into problems.

Then along came derivatives and securitization. In large part, investment banks targeted the assets of commercial banks to produce a liquidity transformation of previously illiquid portfolios. At the same time, commercial banks started to buy up investment banks. Ironically, the child devoured the parent, in both instrument and cultural terms.

The iconic example of this process has been housing finance, using mortgage securitization and its structured variations. This was a central part of the liquidity rush of the past 20 years. It's hard to imagine a more extraordinary example of secular liquidity dysfunction.

The core liquidity risk for a commercial bank is a “run on the bank”. This is a function of perceived and/or actual solvency risk. As financial asset categories that had originated with commercial banks became more liquid, the distinction between institutional liquidity risk and instrument liquidity risk became less obvious. The generic liquidity idea morphed from institutional to instrument risk. The daily concern typically was not so much about institutional funding stability, but about the pricing of financial assets, and the ability of markets to process price changes continuously. Nevertheless, the “bank run” remains the preeminent institutional liquidity risk, investment banks included, as we've been reminded in the case of Bear Stearns.

In the assessment of instrument liquidity risk, circumstances usually conflate different risks and their pricing. It is difficult for example to identify the separate causes of deterioration in CDS spreads as between a reassessment of credit risk versus liquidity risk per se. There is a blur.

Liquidity risk reacts increasingly to closed system boundaries as market size increases. China is typically perceived to provide liquidity to the world in the form of a “savings glut”. This may be excess liquidity from one perspective, but from another one it only neutralizes the liquidity effect of mortgage equity extraction by the US consumer. The net impact of the yen carry trade is constrained by the size of Japan's net foreign asset position, as determined primarily by its ongoing current account surplus. This limits net outflows against Japan's own yen valued “outside” equity position. Closer to home, the Fed provides liquidity to the domestic banking system even in the most stressed and seized up circumstances, by injecting bank reserves that ultimately circulate within an essentially closed aggregate reserve system. The closed property of inside financial assets in general sets a limit on liquidity related pricing distress. Central banks can prime the pump that activates the stabilizing effects of these system boundaries – like ‘breaking' in a game of billiards.
5.6.2008 3:18pm
Mencius (mail) (www):
"Liquidity" is simply a misused word. The correct usage of the term is in markets in which transactions are difficult - such as real estate. My house is highly illiquid, because it would take months to sell, I would have to pay 6% to the real estate industry, the walls are covered with huge pawprints from the world's smelliest Labrador retriever, etc, etc.

This concept is important. It deserves its own word. It should not have to share it with anything completely different. It is almost the polar opposite of any financial asset, which can be traded instantly and for almost no cost - at the right price.

What everyone on this here blog means by "liquidity" is "artificial demand for present rights to future money of high maturity." This demand can be called "artificial" because its ultimate source is not savers who actually demand money in the distant future - eg, 2038 for the last payments in a 30-year mortgage.

Instead the demand for these securities comes from one of three sources: (a) savers who demand money in the near future, but whose claims are balanced by payment streams in the distant future (through the magic of maturity transformation, aka "fractional-reserve banking"); (b) Uncle Sam; or (c) their present holders, who do not want to sell their securities (eg, subprime MBS) at the current market price, because they find it advantageous to hold said assets on their balance sheet at the pre-panic price.

It is (c) for whom the bogus Empsonian ambiguity of the word "liquidity" is useful. It allows the present holders to claim that the drop in the price of these securities is merely a technical glitch, in which the market has "frozen up," and is behaving for some reason like a real-estate market in which it takes me months to find a buyer who also has a dog with big feet.

Ultimately, (b) is the answer behind all three doors. It is the dubious lenience of Uncle Sam that allows parties (a) to claim that payments in 2038 can "balance" obligations in 2008. And it is the infinite forgiveness of Uncle Sam that will ultimately rescue the market with a demand injection, allowing parties (c) to dispose of their assets at something like the official price.

Maturity transformation, the art of defining a short-term liability as equivalent to a long-term asset, is quite simply a shady practice. It is a form of market manipulation. It should be illegal. The fact that it has been practiced continually, with official sanction and indeed favor, since the late 17th century, tells us a lot of things. But it is not a rational justification.

Therefore, I'd go a bit further than Mr. Waldman and say that the right amount of "liquidity" in a financial system - if by "liquidity" we mean maturity-transformed demand for high-latency future payments, and we don't mean the ease of finding dog-loving homebuyers - is exactly the same as the right amount of heroin in your teenage daughter, the right amount of cowbell in the "Goldberg Variations," or the right number of Confederate dollars in your bank account: zero.

The problem of withdrawing the financial system completely from this pernicious drug, "liquidity," is a difficult and interesting one. To even consider solving it, however, you have to at least accept it as a goal. Unfortunately, I see no sign from anyone that matters that they are even close to reaching this realization. More cowbell!
5.7.2008 1:41am
Note: You don't have to register to comment! Just use the pull-down menu to comment as "Guest".

Post as: [Register] [Log In]

Account:
Password:
Remember info?