Buyer's liquidity vs seller's liquidity

Liquid goods are easy to buy and sell. Illiquid goods are hard to buy and sell.

I think we need to break up "buy and sell" into its two component parts.

In a buyer's market, goods are easy to buy and hard to sell. The same good is liquid from the buyer's perspective, and illiquid from the seller's perspective.

In a seller's market, goods are hard to buy and easy to sell. The same good is liquid from the seller's perspective and illiquid from the buyer's perspective.

I think it is true that buyer's liquidity and seller's liquidity vary inversely over the trade cycle. In a recession, buyer's liquidity rises and seller's liquidity falls. Recessions tend to be a buyer's market for normally illiquid goods. In a boom, seller's liquidity rises and buyer's liquidity falls. Booms tend to be a seller's market for normally illiquid goods.

If I am right about that stylised fact, it seems unwise to ignore it, when we build theories of the trade cycle. [Update: It's a large part of what we are trying to explain.]

When "keynesians" (I am using that term very loosely) talk about sticky prices, and "excess supply" of goods (like labour) in recessions, and "excess demand" in booms, they are trying to say something about that stylised fact. It's a very crude and theory-laden way of talking about that stylised fact, but that's what they are trying to say. A market in excess supply is a buyer's market, because buyers can buy the quantity they want to buy but seller's can't all sell the quantity they want to sell. A market in excess demand is a seller's market, because sellers can sell the quantity they want to sell but buyers can't all buy the quantity they want to buy. "Can" and "can't" is a crude binary way of distinguishing between "easier" and "harder". [Update: But it's better than nothing.]

We can argue about how precisely we define and measure "easy" and "hard". But there is some sort of trade-off between: how quickly you make the trade; how much effort and expense you incur to make the trade; and how good a price and quality you get. If you can trade a good instantly, with zero effort or transactions cost, and you can't get a better price or quality by waiting longer or with higher effort or higher transactions cost, then it is very liquid. And if the trade-off worsens, it becomes less liquid. Talking about that trade-off worsening or improving for buyers or sellers is a better way of talking about "excess supply" or "excess demand".

In a recession, the trade-off worsens for sellers and improves for buyers. In a boom, the trade-off improves for sellers and worsens for buyers.

If we lived in a barter economy, the distinction between buyers and sellers would make no sense. Nor would the distinction between buyer's liquidity and seller's liquidity. We would have to talk about the liquidity of each possible pair of goods being traded. It might be easy to trade apples and bananas, hard to trade apples and carrots, and easy to trade bananas and carrots. But we live in a monetary exchange economy, where the buyer of goods is a seller of money and the seller of goods is a buyer of money. But money is the most liquid of all goods (which is why it is used as money) so we ignore its liquidity.

You have to be a "monetarist" (I am using that term very loosely too) to even talk about these things. None of this would make sense in a non-monetary economy.

I am trying to persuade Steve Williamson to come on over to the Dark Side. Because this sort of stuff ought to be right up his street, and isn't really all that dark from his perspective, if we say it in the right way. I will probably fail, but there's no harm in trying.

Comments

[louis]

Two thoughts as a financial markets guy, hope they are pertinent:
1) Flexible prices would seem to prevent buyers' and sellers' markets, in that prices should adjust to correct imbalanced supply and demand. Yet in thinly traded equities, you definitely see markets where it is easier to either buy or sell quantity. This in a place you wouldn't expect stickiness. Part of this may be manuevering by the parties involved: if I have a large quantity to buy, I may get best execution by standing firm at a price and letting sellers come to me over time, rather than marching up price to the level sellers will fill my bid. This may include hiding the true size of your ticket. In the real economy, how much do sticky prices and non-clearing markets represent this type of gamesmanship?
2) Presumably you are talking about the markets for newly produced goods and labor becoming buyers mareas in a recession. In some recessions, but I don't think all, capital markets also become buyers markets in that liquidity is one-sided. Makes sense in that money is involved in both kinds of market. Still, what would you say if a recession saw goods markets glutted but financial markets functioning well?

[Nick Rowe]

louis: yes. Pertinent and interesting.

Financial assets are usually more liquid. (But IOUs from individuals or small firms may be very illiquid). So we would normally see less of a buyer's market for financial assets than for illiquid goods like labour, but I think we would still see a tendency in the same direction.

[Procyon Mukherjee]

Hi Nick,

Some of the commodity markets where price is well below the median of the cost curve for the industry and the interest rates are a means of making an allowance for the industry to continue to add to the glut, the liquidity of the entire industry cannot be sustained by financial products as a fillip; it is finally the end consumer demand that must match supply which makes the price equate the marginal cost of output.

Isn't it?

Procyon Mukherjee

[Nick Rowe]

Procyon: Suppose there's a big harvest due to good weather, and the price of wheat falls very low. So low that farmers regret planting it. It's still very easy both to buy and sell wheat.

[Kevin Donoghue]

Yes, you're using the term "monetarist" very loosely indeed. I suspect that a fruit-and-nuts economy would turn out to be a monetary economy, with the nuts being a store of value. (There has to be some scope there for distinguishing the views of traditional nutters from those of Market Nutters and Neo-Fisherian Nutters, which might be fun.) If you want to persuade Steve Williamson, you may need to rewrite your ideas using more jargon: essential sequence economies in Radner equilibrium, or some such gibberish. As this stands it isn't sciencey enough for him.

[Market Fiscalist]

I'm just thinking aloud here so this may not make sense or be relevant...

An other thing that is sort of involved in every transaction is leisure. If I don't buy so many banana I can work fewer hours. If am an apple producer this means I will produce and sell less apples. If I am not sure how many apples I can sell at what price because of market uncertainty I may choose more (guaranteed) leisure over (uncertain) bananas obtained by the hope of swapping apples for money, and then money for bananas.

Even if all prices adjusted immediately and eliminated monetary recessions you could still have a trade cycle driven by the "price" of leisure being correlated to uncertainty (and uncertainty being an exogenous variable).

[Avon Barksdale]

Liquidity is about information sensitivity, not recession vs no recession. In an illiquid market, if I see a seller, I assume that he knows something I don't, so I demand a steep discount to reflect my lack of information. If I see a buyer, who bids, I assume that he knows something I don't, so I demand a premium at the sale. That's why illiquid markets see large bid/ask spreads. If an asset is hard to sell quickly without price impact, it's also hard to buy quickly without price impact. Liquidity means that the asset is information insensitive. Index funds stay liquid even during recessions. Buying index funds in 2009 was easy to do, very liquid, and the strategy paid off handsomely for the willingness to hold price risk when no one else wanted it. Price risk, the price reset during a recession, is not liquidity risk (as LTCM found out the hard way).

[Min]

In terms of the metaphor, liquidity is about flow, so it would apply to trades, not buyers and sellers. If you had some sort of measures, liquidity(buyer), and liquidity(seller), the liquidity of trades should be the minimum of the two. For instance, if you had a lot of bids and few asks, if the asks are filled, the trading is over for the time being, and vice versa. As you point out, when you buy or sell something for money, money is the more liquid good in the trade, so we ignore the liquidity of money. That follows from the determination of liquidity of the trade by the minimum liquidity in the trade.

But this does not seem to be the kind of thing that you are driving at, as overall liquidity is different in booms and recessions. So it seems like you are breaking the metaphor. I can't say whether that is good or bad. You see some value in doing so. And you indicate that talking about supply and demand doesn't get at what you have in mind. As a good bridge player, I have an easy time getting a partner at a tournament. Doesn't that follow from my being in demand as a bridge partner? (As Victor Mollo once quipped, the supply of bad bridge partners exceeds the demand. ;))

For us to see what you are driving at, perhaps you could say more about why you find talking about supply and demand unsatisfactory. :)

[kiers]

buyers liquidity w sellers relative illiquidity? how does that work? who are the buyers buying from then?

[Min]

BTW, one way, perhaps, not to break the metaphor is to talk about pressure, buyer's pressure and seller's pressure. I don't know if that fits, but there are terms relating to different aspects of flow that might be enlisted. Seller's pressure vs. buyer's capacity? In terms of electrical flow, terms like voltage, resistance, capacitance?

[Frank Restly]

Nick,

It seems you are contradicting yourself a bit here.

"I think it is true that buyer's liquidity and seller's liquidity vary inversely over the trade cycle."

If that is true then the liquidity of money held by the buyer must change over a trade cycle as well.

"But money is the most liquid of all goods (which is why it is used as money) so we ignore its liquidity."

You can't just ignore the liquidity of money if it's liquidity changes over a trade cycle.

I think the distinction needs to be made between the liquidity of an asset - how easily can this good be sold / traded for other goods, versus the liquidity preference of buyers and sellers.

What particular properties make one good more liquid than others (irrespective of the preferences of buyers and sellers)?
1. Divisibility - Air planes are illiquid because half an airplane is a hunk of scrap metal, Apples are liquid because half an apple is still an edible product
2. Portability - Paper money is more liquid than coin money because it physically weighs less and is easier to transport
3. Uniform measure of value - Apples are liquid because they can be measured by the pound, Paintings are illiquid because the value of a painting is subjective
4. Coercion - Legal or otherwise
5. Durability - Paper money is more liquid than apples because apples rot over a short time frame, paper money lasts for years

If we start there with our guidelines for assessing the liquidity of a particular asset, we can then establish the relative liquidity of two or more assets (money being more liquid than most other things).

We can then set up liquidity as one measure of the value of a particular asset. Other measures of value could be utilitarian value, ephemeral value, etc.

Then we could model an economy based upon supply and demand for the various attributes of goods rather than the goods themselves - supply / demand for liquidity value versus supply / demand for utilitarian value.

[louis]

Min - liquidity is not necessarily about flow.
In a metaphoric sense, think of liquid as opposed to solid or frozen. Calling an asset liquid means you can easily move part or all of it. There is a robust bid close to the current market price. If you change your personal demand for the asset, and become a seller at the current price, the resulting shift in the overall supply curve would not change the market price much; either because the demand curve is relatively flat (everyone agrees on the appropriate price) or because the overall stock of the asset in question is way bigger than the amount you hold.
Flows are indicative of liquidity but not necessary. If you want to swap Yen for Swiss Francs, there may be little flow on that particular currency pair. But you can do it easily because both currencies have deep markets against the US dollar.
The more bespoke the market, the more likely you are to represent a big chunk of supply or demand, and the less likely there is to be broad agreement on the appropriate price of the asset. So your decision to buy or sell is likely to move the market clearing price.
How could sellers' liquidity dry up but buyers' liquidity improve? Say a bunch of big asset holders, like banks, were required to raise cash for regulatory or other reasons. Their views on individual assets wouldn't have changed, but they would become willing sellers of many assets in their portfolios until they satisfied the need to raise cash. They also would cease buying assets unless they had really compelling options. So if you went out to buy assets in such a market, even in size, it would be easier to do so. If you tried to sell, it would be harder. Because the banks can choose which assets to sell, the assets they choose to actually sell will depend on where buyers show up, and so buyers will have an easier time showing up in any particular market.

[SvN]

Nick; Am I right in thinking that this line of argument is very much at odds with recent thinking and research on credit and collateral? For example, Brunnermeier's analysis of the 2007-08 Financial crisis puts emphasis on funding liquidity, which depends critically on the value of collateral. Gorton's recent work on financial crises stresses the financial system's need for information-insensitive stores of value (of which "money" is the paragon) such as "AAA"-rated debt. I thought that a key feature of collateral it that is has liquidity from both the buyers and sellers point of view (or, in this context, perhaps I should say the borrower's and the lender's.)

[PelinoC]

On its surface, this post seems to be a tautology: define liquidity as "ease of transacting", then define markets as being more or less liquid from the perspective of a particular market actor (buyer or seller), based on their perceived ease of transacting. So?

Suddenly it gets interesting, however, with the move from talking about a particular market actor or class of actor in a particular market, to "recession" and "boom". These are descriptions of the states of being of entire economies (collections of multiple markets including primary, secondary and final goods/services). In an economy, most actors are simultaneously buyers and sellers, albeit across multiple markets (there are a few actors, such as retired people, who are basically only buyers, and another few, such as corporations that own a single natural resource asset like a mining claim, that are only sellers).

However, I'm still not sure what is being added? If a "recession" is defined as: the state an economy is in when in a majority/most of its markets sellers find themselves facing low liquidity. Again, so? Does this point to a possible solution to recessions (improve ease of transacting for sellers?) that has not been considered before? Or does it lead to a novel cautionary tale about "booms" (if things look like they are getting out of hand, create some transaction costs?)? Is there some other insight being striven for?

I am confused about the objective...

[Nick Rowe]

SvN and Avon: Take labour as an example. Labour is a very illiquid good, because each worker (and each job) is different. It gets harder to sell labour and yet easier to buy labour in a recession. That fits the anecdotal evidence, and also fits the rise in the unemployment rate and decline in the vacancy rate. I would say the same is usually true of many produced goods. It's easier to buy a haircut, and harder to sell a haircut. Or a car. Etc.

Now, does this same pattern also fit financial assets? What about the less liquid financial assets? That's where I'm less sure.

But "liquidity" is a concept that applies to all goods and services, not just financial assets.

[Nick Rowe]

Pelino: "If a "recession" is defined as: the state an economy is in when in a majority/most of its markets sellers find themselves facing low liquidity."

It's not obvious how we should *define* a recession. Is it defined as a fall in GDP? Why not define it as a fall in employment? Or a rise in unemployment?

What we usually do is note a pattern of co-movements in the data. Lots of data series go up and down together (with some being positively correlated and some negatively correlated). The business cycle is called a "conjuncture" in many languages, which is a good name precisely because it refers to that pattern of co-movements.

What I am doing is pointing to one part of that pattern that should not be ignored. We could say it is "part of" the definition of a business cycle, or we could say it is just an empirical fact about the business cycle.

[Min]

Thanks, louis. :)

louis: "Min - liquidity is not necessarily about flow.

"In a metaphoric sense, think of liquid as opposed to solid or frozen. Calling an asset liquid means you can easily move part or all of it."

Sounds like flow to me. :)

louis: "How could sellers' liquidity dry up but buyers' liquidity improve? Say a bunch of big asset holders, like banks, were required to raise cash for regulatory or other reasons. Their views on individual assets wouldn't have changed, but they would become willing sellers of many assets in their portfolios until they satisfied the need to raise cash. They also would cease buying assets unless they had really compelling options. So if you went out to buy assets in such a market, even in size, it would be easier to do so. If you tried to sell, it would be harder. Because the banks can choose which assets to sell, the assets they choose to actually sell will depend on where buyers show up, and so buyers will have an easier time showing up in any particular market."

That illustrates why I am curious as to why Nick does not think that the language of supply and demand is not a good way to talk about what he means. Perhaps he is talking about something that is subtly different.

[louis]

Min: "That illustrates why I am curious as to why Nick does not think that the language of supply and demand is not a good way to talk about what he means. Perhaps he is talking about something that is subtly different."

Yeah, I had the same thought as I was drafting that comment. I think the key is the stickiness of wages and other prices and the non-commodified features of the labor market. You don't see every worker leave their job once a month and get drafted by a new company at a new market-clearing wage. Nick had some posts talking about search models and shifting tradeoffs in finding a job, so that might be in his mind. When prices are above equilibrium prices, rationing, queueing, lotteries and the like end up determining allocation of resources. I think this post reconceptualizes that sort of stuff as assymmetric liquidity.

[louis]

^ when prices are *below* equilibrium prices. When they are above, I guess you have inventory builds, waste and unemployment. But also some form of queueing and jockeying for position.

[Frank Restly]

Nick,

"It gets harder to sell labour and yet easier to buy labour in a recession...Now, does this same pattern also fit financial assets? What about the less liquid financial assets? That's where I'm less sure."

Labour suffers liquidity problems that modern financial assets typically don't have:
1. Labor for the most part isn't very portable - There are significant costs for a person moving from say New York to San Diego to take a new job. Many financial instruments are held electronically and so transferring ownership of financial assets is done by typing on a computer.
2. Labor isn't divisible to the nth degree. Once you get down to a single worker, the next step is reduced hours for the single worker which can negatively affect productivity. A company that splits it's shares is virtually unlimited in the ratio of new shares to old. Same thing with currency.
3. Labor has a fixed lifetime, shares of a company can exist for centuries

That being said, your question was do financial instruments (especially the less liquid ones) become harder to sell and easier to buy during a recession. It depends on what you mean by recession (collapse in nominal or real GDP) and what you mean by harder to buy / easier to sell.

Suppose nominal GDP growth collapsed, but real GDP grew - indicative of a large growth in productivity / competition. Is this a recession? It will be harder to obtain the same price for a good than was received at a prior time, but the goods themselves are no more difficult to sell (assuming that money is infinitely divisible), in fact they may be easier to sell if that productivity increase reduced transportation, storage, or other costs. Now suppose nominal GDP growth grew, but real GDP collapsed. Is this a recession? It will be easier to obtain a better price for a good than was received at a prior time, but the goods themselves may be more difficult to sell (assuming a fixed supply of money).

Carrying it through to financial instruments, if stock prices collapse does that necessarily mean companies will find it more difficult to sell new issues? Again, the distinction must be made between a company able / unable to sell shares at the price it wants to receive and a company being able / unable to sell the quantity of shares it wants to sell. A company should have no problem selling additional shares if it will accept whatever price the market offers (assuming that money is infinitely divisible). A company may have a problem selling additional shares if it's asking price is too high for each additional new share (assuming a fixed supply of money).

And finally, is there a positive / negative / no correlation between the ease with which financial instruments are sold and the ease with which goods / labor is sold? I don't have a good answer for that one. If we define easy / difficult with regard to the quantity of goods / instruments being brought to market rather than the price (quantity of money) for each good / instrument, then I would say that financial instruments are easier to sell (recession or no recession) than labor / goods.

[Avon Barksdale]

Nick, this all makes little sense. You are talking about a market clearing mechanism as it finds a new equilibrium. In a recession, it is both difficult to find a job AND an employee. Unemployment increases which means people will not accept a lower wage offer. It's no picnic for the employer either.

An illiquid market means information is crucial and hence an option value in waiting - that's the origin of the discount. Search-match models in labour economics contain this option value. In the limit that illiquidity turns into irreversibility, information reigns supreme - the value of the option can easily exceed the intrinsic value of the investment. I suggest "Investment Under Uncertainty" by Dixit and Pindyck. Governments "misunderstand" the socially optimal value of delaying investment when faced with uncertainty and irreversibility. But, it's all about redistribution to the base that keeps you in office - economics be damned.

[Nick Rowe]

Avon: remember that trade-off curve I drew a couple of posts back, trying to make sense of "involuntary unemployment"? That was for a worker looking for a job. We could draw a similar sort of curve for a firm looking for a worker.

Which way do those two curves shift when there's a recession? I think the one facing the worker shifts right (worsens) and the one facing the firm shifts left (improves).

But in the output market it's the opposite, because the firm is now the seller, and faces a worsening trade-off.

[Avon Barksdale]

Nick,

No, the trade off curve you drew misses the option value in waiting. That's why we see unemployment and it's socially optimal and voluntary. Yes, a person can take a lower paying/worse job, but they choose to wait for more information. Choices (investment) under uncertainty with irreversibly/illiquidity imply an option. The employer sees a bunch of people waiting who refuse his offer.

[rsj]

This is a perfectly valid stylized fact, but I think the goal of convincing Steve Williamson is hopeless.

I think the big divide is between those who believe in the General Equilibrium fairies and those who don't. I.e. suppose you try to model some explicit mechanism -- search theory, explicit modeling of monetary transactions, etc -- it doesn't matter if, after setting up your equations, you just solve for the price vector and assume that the economy is constrained to move along that optimum, pushed along by Angels. Otherwise all these stylized facts will be treated as a constraint in the optimum solving fairy-world. Maybe things are not exactly Panglossian but are second-best panglossian. Or possible third-best panglossian -- It's still Panglossian with no micro-foundations until you explicitly have a price setting mechanism in your model.

Meanwhile, I went to my local convenience store who just raised their Sport Ritter prices to $4, and I complained that this was too expensive. You can get Sport Ritter for only $3 just across the street. He would make more money if he lowered the price. The store owner mumbled something about these being straight from Germany and so costing more. I told him that this was B.S. because the euro dropped. He got upset with me and said "well, oil is half the price now, so why don't plane tickets cost half as much?!". I walked away disgusted with this inefficient small-time Palestinian businessman, and took my money over to the big chain store that sold me goods at fair prices. Screw him and his additional dollar mark-up just because he thinks he's surrounded by high earning people who are too stupid to compare chocolate bar prices.

We need a model of the economy in which people stumble through making lots of mistakes, each employing different suboptimal price setting strategies. As soon as you agree to the General Equilibrium methodology you've already lost the game. You cannot then argue about stylized facts because the argument wont make any sense.

[Roger Gomis]

Hi Nick,

I'm still thinking on this issue that keeps being raised in your blog - I have not an opinion, yet. Nevertheless one might model what you are saying similar to this paper http://www.econ.umn.edu/~vr0j/papers/brs_Sept_27_2012.pdf. They only have cost of search in the "demand side", and no liquidity considerations - and their model is very RBC in spirit. But, I think, the framework could be easily adapted to what you are saying. I'm unsure about getting insights from doing that. What do you think?

[Nick Rowe]

Roger: interesting paper. I had a very quick skim. Here is a very simplified version of what I *think* is going on:

It's a haircut economy. Workers sit idle waiting for a customer. But customers have to search to get a haircut. The stronger their desire for a haircut, the more are willing to search, and the more haircuts get sold. So it looks like productivity rises in a boom, even though technology stays the same. This is the standard keynesian "retained workers" explanation for why measured productivity falls in a recession. Firms don't lay off idle workers. But I am unpersuaded that recessions are caused by people being less willing than normal to search for a store that sells what they want.

Avon: and what happens to that option value in a recession?

[Roger Gomis]

Hi Nick,

"But I am unpersuaded that recessions are caused by people being less willing than normal to search for a store that sells what they want." Indeed, this is what I did not like about the paper. In fact I'm convinced that the model works better than RBC just because these "preference" shocks combined with labour hoarding can capture Monetary/Keynesian shocks (also with labour hoarding) whereas standard RBC can't. I'm working on the issue from the effects of technology shocks controversy perspective. I was thinking that the model could easily include search and matching (instead of only search) and introducing money to formalize what you say here: "Talking about that trade-off worsening or improving for buyers or sellers is a better way of talking about "excess supply" or "excess demand" ". But I'm unsure about the insights it can bring regarding the procyclicality debate.

[JP Koning]

Interesting post, Nick. I wonder how well a stock market crash like the 2010 Flash Crash can serve as a microcosm of a recession. As markets plummeted, spreads widened dramatically. It simultaneously got harder to buy/sell stocks and to buy/sell money, since in either case the trader had to trek across a much wider bid ask spread than before, and had to transact against a much lower "wall" of of bids and offers. Buyer's liquidity and seller's liquidity both fell.

[Nick Rowe]

Roger: "labour hoarding". That's the word I was looking for!

You could build a very similar model with shocks to money demand or money supply. But only if you managed to get sticky prices in some way. Or you could introduce money, keep prices flexible, and have the same sort of RBC model underneath, driven by preference shocks. What you suggest might be worth doing.

[Nick Rowe]

Thanks JP! I wondered when you would show up, since I think this is up your street.

I *think* we see microcosms of recessions in the market for real assets like houses. A fall in house sales tends to be associated with a buyer's market, and a rise in house sales tends to be associated with a seller's market. Buyer's liquidity is negatively correlated with sales, and seller's liquidity positively correlated with sales.

But I'm much less sure about financial assets.

[Roger Gomis]

Hi Nick,

Thanks! I'll give it a try.

[Avon Barksdale]

"What happens to that option value in a recession?"

It goes up, hence more waiting.

[Nick Rowe]

Avon: right. Which means that labour has become less liquid to the seller of labour. The trade-off has shifted right, like in the diagram in my other post.

Now what happens to the option value of waiting to hire for the buyer of labour? I think it goes down. You get hundreds of applicants quickly, so there's little point in waiting for a better applicant. Labour has become more liquid to the buyer of labour.

[louis]

Nick: "I *think* we see microcosms of recessions in the market for real assets like houses. A fall in house sales tends to be associated with a buyer's market, and a rise in house sales tends to be associated with a seller's market."

How much of this is just behavioral effects? Sellers see falls in prices and are slower to revise down their mental reserve price for the house (or are reluctant to recognize a loss), so bid-asks widen. Anecdotally, you hear about that going on in the market for oil reserves today.
In financial markets, you see momentum effects, where a rising price makes sellers more reticent and buyers more anxious (they may be afraid of "missing" the move in stock).
In the labor market, you may see this effect in laid-off employees being unwilling to accept a job with a lower wage than they received at their previous job.

[Nick Rowe]

louis: maybe it's all "behavioural effects"!

But is it irrational, or imperfect information, or costs of changing prices, or difficulty in solving for the Nash equilibrium, or what? We don't know. That's the underlying puzzle of the Phillips Curve.

[JP Koning]

Thought this was apropos:

"The most striking characteristic of depression is not overproduction of some things and underproduction of others, but rather, a general “buyers’ market,” in which sellers have special trouble finding people willing to pay more for goods and labor. Even a slight depression shows itself in the price and output statistics of a wide range of consumer-goods and investment-goods industries. Clearly some very general imbalance must exist, involving the one thing–money–traded on all markets. In inflation, an opposite kind of monetary imbalance is even more obvious."

-Leland Yeager

[Nick Rowe]

JP: Good find!

[Avon Barksdale]

Nick,

No, the option value goes up all the way around during a recession. The employer does not get hundreds of applications because he makes no offers either. He waits, just like the potential employees for more information before making an offer. Everyone delays during a recession.

This discussion is a very silly notion of liquidity. The market is two sided, you can't have a buyer without a seller. A liquid market for one side means a liquid market for the other. If I can sell an asset without price impact it must be true, by an identity no less, that I can buy without price impact.

[LAL]

im not sure why we have to talk about only one kind of moneyness...we can just form a probability of acceptance, and use that as a measure of liquidity for a basket of potentially liquid assets and goods. then we don't have to ignore the fact that plenty of places use multiple currencies (like US border regions) , and have multiple simultaneous price systems like in countries where credit cards are common...

[Nick Rowe]

Avon: you work in finance, right? Ever hear finance guys talk about a "buyer's market"? What do you think they mean by that? Isn't it the opposite of a "seller's market"?

LAL: almost anything could be acceptable, at the right price. But I think I would usually get a much worse deal if I tried to buy a canoe and use my car for payment. I would first sell my car for dollars, then use those dollars to buy a canoe. Which is what people nearly always do.

[louis]

Avon: " A liquid market for one side means a liquid market for the other. If I can sell an asset without price impact it must be true, by an identity no less, that I can buy without price impact."

Certainly false. The bid side can very well be deeper than the ask. There is no reason they need be symmetrical.
When the SNB put a 1.2 floor on EURCHF, you could sell as many Euros as you wanted for 1.2 CHF, because the SNB had an unlimited desire to buy EUR at that price. If you wanted to sell CHF in size instead, there was no guarantee that you wouldn't move the exchange rate.

[Nick Rowe]

louis: good point. Wish I had said that.

[Avon Barksdale]

louis and Nick,

What, are you saying the CHF was illiquid? Given me a break. Besides, government intervention is not a market outcome. You guys are all strange. When economists debate identities, you know you're in trouble.

[Nick Rowe]

Avon: take any financial asset, like a share. The bid side can be deeper than the ask side, or vice versa.

[louis]

Avon: If you define liquidity as the ability to exit your position without moving market price much, yes. Talk about the *position* you hold being liquid or not, as opposed to the market. At times, it makes a difference whether your position is long or short. You don't need government intervention per se, just some sort of rigidity or asymmetry in the market participants' behavior.
If you look at the distribution of 1 day performance for IPOs, you can see this asymmetry. In the US, underwriters are traditionally given an option to buy more stock from the issuer at deal price within 30 days of the offering. So on a 1 mm share offering, they can choose after the fact to upsize the deal to 1.15 mm shares. This allows the underwriter to oversell the offering - allocating 1.15 mm shares in aggregate to the buyers of the IPO. The underwriter is thus short 150k shares at the outset, but with an option to buy 150k shares at deal price within a month. The incentive created here is for the underwriter to "defend" or "stabilize" the deal, by buying up to 150k shares at deal price or below, but not a penny above. This provides great liquidity for a seller of IPO shares, as there is a large natural buyer at or around deal price. But if you want to buy the shares, you must get someone else to sell to you... the liquidity is not there in the same way.
You can come up with loads more examples like this. Markets are interesting things, real supply and demand curves are not perpendicular straight lines.

[Avon Barksdale]

Louie,

You are talking about market segmentation, not liquidity.

[louis]

From today's WSJ:
Apartments in Copenhagen that used to be on the market for months now sell in days or even hours, realtors say, while prospective buyers line up early at open houses and squeeze inside along their rivals.
“People will often put in an offer even before they see the apartment,” said Christopher Christiansen, a property broker at Danish real estate agency Home A/S. “Sometimes they will even sign before they see it.”

Tell me that liquidity is not greater for a seller than a buyer in such a market.

[Oliver]

Liquidity: the ease with which I can sell an asset. That only works for sellers because buyers already have money. The definition of liquidity is built into the dichotomy of money / non-money.

Or

Liquidity: the volume of trade within an asset market relative to another period. An observation that is symmetric for buyers and sellers. Are more people signing new leases in Copenhagen than 5 years ago? Do movements in trade volumes correspond with economic cycles? Can asset markets freeze up / become less liquid during a boom? Does high trade volume necessarily drive up prices? Isn't a fire sale market a high volume market? Which would mean it is very liquid (from the point of view of both the seller and the buyer). It's just that sellers feel compelled to record losses.

You (Nick) seem to be defining liquidity as the ease at which an agent can find a profitable trade. But I think that's missing the macroeconomics. A bust may be a buyer's market for the buyer who still hase money. But there will be less buyers with money during a bust. A buyer's market by your definition is actually one that is short of buyers.

[Nick Rowe]

Oliver: "A buyer's market by your definition is actually one that is short of buyers."

A buyer's market is where it is hard for sellers to find buyers, but easy for buyers to find sellers. Sure, it's short of buyers.

A sellers's market is where it is hard for buyers to find sellers, and easy for sellers to find buyers. Sure, it's short of sellers.

[louis]

Oliver -
The concept of a "fire sale" is not that the sellers are recording losses that reflect a change in the underlying fundamentals. It is that in the trade-off between speed of closing the sale and finding the best price, the seller is forced to put far more weight on the first consideration than usual. From a buyers' perspective, a "fire sale price" connotes a bargain relative to fundamentals, not just a sharp decline from where the asset was historically priced.
If the asset that the seller needs to sell is liquid - i.e. there is a robust pool of potential buyers to market the product into, who can easily come up with the money to transact in short notice - then he is less likely to have to accept a terrible price if he is forced to liquidate at short notice.
In a market where liquidity (in the sense of access to money) is in short supply, sellers' liquidity (in the sense of ability to sell an asset without a sharp tradeoff between time on the market and price) is likely to be diminished. Or at least that's what I think Nick is driving at. I have a harder time wrapping my head around how this works in the market for goods or labor vs assets, whereas I think Nick has the opposite perspective.

[rsj]

Nick,

I am not missing the point. Currency is an irrelevant detail in a modern economy with a modern financial system. Just assume everyone exchanges with bonds -- that's kind of the Woodford assumption and I think he is right about this. The Old Keynesian view is that when I purchase a financial asset instead of a newly produced good, then I am decreasing someone else's income. I.e., income adjusts so that savings equals investment, rather than interest rates. The New Keynsian view acknowledges this as a possibility at the ZLB. In neither model is currency demand particularly important as distinct from bond demand.

Moreover, no one is going to want to hold more currency as a result of a capital income tax, because such a tax is only paid on positive returns. A capital income tax can never drive interest to be negative (that would be a tax on wealth or on returned capital). All it can do is lower the real interest rate earned to something above zero (or actually zero if it is a tax rate of 100%), because we live in a world in which the marginal return on capital is not equal to the actual return on capital for everyone in the economy.

[rsj]

Here, let me plug "Keynesian Macro without an LM curve"

http://eml.berkeley.edu/~dromer/papers/JEP_Spring00.pdf

[Nick Rowe]

rsj: wrong post! But if you want, copy and paste to the other post, and I will unpublish these.

[rsj]

Oh, I am sorry!! Yes, please delete these.

[Oliver]

Nick

OK. Just saying, as a macroeconomist / if referring to the economy as a whole, I think one shouldn't call it a buyer's market.

Louis

You're right. I was referring to the case when the economy as a whole goes into fire sale mode. A crash, I guess would be the more accurate term. Not sure whether fundamentals are a useful measure in that case. Long term perspectives, maybe?

In any case, I think there's a micro - macro thing going on here, at least if one frames it in a boom / bust context (as Nick explicitly did) as opposed to just markets for one asset or another. During a bust, for the shrinking group of individuals who can still afford to buy, their position as buyers improves. Yay for Warren Buffet!

[louis]

Oliver -
1) I'd define "fundamental price" in any market for goods or services as solving for the market clearing price of the supply and demand functions for that good.
2) We definitely saw conditions in the financial markets in early 2009, where the small pool of people who had access to liquidity were able to capture real bargains. Other people recognized these bargains at the time but were liquidity-constrained.
3) Nick is always saying that recessions only make sense in the context of a monetary exchange economy, where the medium of exchange is undersupplied. Clearly that is what happened in the financial markets in 2008-2009. I think the interesting observation here is that not only do you see a worsening trade-off in ability to sell an asset quickly or capture its market price when money is undersupplied, but you see the same thing in your ability to sell labor. They may be two sides of the same coin.

[Nick Rowe]

louis: " I think the interesting observation here is that not only do you see a worsening trade-off in ability to sell an asset quickly or capture its market price when money is undersupplied, but you see the same thing in your ability to sell labor."

Yep. And newly-produced goods too.

"They may be two sides of the same coin."

Yep. And we should understand that metaphor almost literally!

[Oliver]

Louis
1) So the fundamental value of my company stock is whatever it is? That sounds fundamentally tautological and very unheplful, especially during interesting times.
2) I'd say the economy as a whole was risk averse. Agents, especially banks, engaged in defensive behaviour with the aim of protecting and repairing their respective balance sheets. And individual constraints and risk aversion meant that some investment opportunities were left underpriced vs. some fuzzy fundamental value and thus represented an opportunity to those who were both smart and credit worthy or liquid. And sure, the paradox of thrift made sure the labour market was affected in a second order effect.
3) Surely the argument is that only in a monetary economy can the monetary authorities do something against a recession? I'm sure barter can freeze up for many reasons, no? Why is that not a recession? In which case I'd be enclined to agree to a certain degree. But, tying in with point 2, I'd put investment, and in particular the bank / client nexus, at the beginning of the chain of events. All else, including the labour market and goods markets, follows. Monetary authorities only show up on the sidelines in that they can influence both banks' and clients' behaviour with various tools at hand. Fiscal authorities show up as separate investors endowed with extra credit worthiness and a national / macro mandate. Importantly, a supply of money / medium of exchang does not figure as a causal factor in this. Thus, it makes no sense to claim that it is undersupplied. It is a residue. But I doubt I can convince a monetarist of all that, seing as many much smarter and more educated people than myself have failed.