« How scared of deflation were you, in 2008? | Main | The Bank of Canada vs the bond market »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

An excellent post, as usual. Thanks, Nick.

Do you think the cash-in-advance constraint of more modern models captures this desire to have inventories of money or do you think there is something fundamentally different between the old Keynesian/Monetarist story and the c-i-a constraint?

Primed: Thanks! (I really appreciate that, because I had an immediate "poster's regret"!)

Hmmm. I think that if we had a CIA constraint, plus a lot of uncertainty about the timing of expenditures and receipts of money (maybe I will need to fix my car and not get paid at all next period) we could get something similar. Or not all markets open every period. Or costly to get to a market unless you are passing by anyway. Something like that. Dunno.

Or, assume most goods except money are illiquid, and it's costly to buy and sell them quickly.

> Now suppose that all goods are services, like haircuts, and it is physically impossible to store unsold haircuts.

We don't even need a haircut economy; it works with apples, also. If inventories of apples rot after a period, then unsold apples go to waste.

> Because most shocks to inventory are purely transitory. Which is why we hold inventories. It takes time to realise the shocks are more permanent.

Thank you for saying this so plainly. I think too often we end up trying to hide that the story of money is really a story of time.

> Or, assume most goods except money are illiquid, and it's costly to buy and sell them quickly.

Or most goods require binding orders in advance, coupled with a no-default condition. This is even close to the truth for labour.

Majromax: "If inventories of apples rot after a period, then unsold apples go to waste."

Yep. But I had to duck that, because it adds an additional flow out. And I can never remember how unsold rotting apples are handles in national income accounting. Depreciation?

" I think too often we end up trying to hide that the story of money is really a story of time."

Yes! M=Md has no time dimension. We might be talking about a stock of jewelry. V=Vd has the dimensions 1/time.

"Or most goods require binding orders in advance, coupled with a no-default condition."

Yes. I was wondering how to build in the stock of unfilled orders. Sort of like a negative inventory? And I thought of customers sitting in the shop waiting for a haircut similarly. But couldn't make my mind up about how best to think about it, so left it out.

My old post about red money, with negative value, is like a negative inventory of money, like unfilled orders.

> And I can never remember how unsold rotting apples are handles in national income accounting. Depreciation?

I believe so, yes. Like those notices in paperbacks saying "if you purchase this book without a cover, it was reported unsold and destroyed, and the author received no payment."

> Yes. I was wondering how to build in the stock of unfilled orders. Sort of like a negative inventory? And I thought of customers sitting in the shop waiting for a haircut similarly. But couldn't make my mind up about how best to think about it, so left it out.

Physical goods are all sold by mail-order, and services are by appointment (paid in advance) only. I like this story slightly better than one of transaction/capacity costs (say the barber rents the shop by the month but is paid per haircut) since everything remains totally certain, just with a delay.

One interesting and potentially-related mathematical result is that some delay differential equations (differential equations that incorporate terms with delay) are unstable even if the non-delayed version would be stable.

It is interesting to consider the same dynamics in the labor market.

First, I notice that labor is seldom paid until an inventory of money has been built by the employer. In other words, there is a demand for labor that is satisfied (let's assume on a daily basis) without compensation for several consecutive days. During these unpaid days, a money based account is maintained, and FOLLOWING each day, the account is incremented.

At some point in time, money is paid to each worker. In my view, this payment is an actual transfer of property from employer to worker.

This description seems to fit your model. The employer is creating production in exchange for a draw down of monetary property. The worker has exchanged services for future income (measured in monetary terms).

If there is an explanation gap here, it seems to be that labor inventory is not part of the model. This is similar to the inventory of haircuts that we can only measure after the haircut is complete.

Small non-quibble: "It takes time to realise the shocks are more permanent". This is surely true for businesses, but less true for labour. When an employee is fired he realises *immediately* that his income will be less than spending in the next period and so immediately cuts spending. It's his suppliers who are unaware.

Squeeky: Maybe. But a worker does not know if a temporary layoff, or temporary reduction in hours, will become permanent. And if he loses his job, he might think he can find another job quickly, and eventually learn that he can't.

Majromax: it would be neat to do a version where all goods are ordered in advance, and the length of the queue varies.

Roger: I wouldn't call that a labour inventory. I would say that workers are selling their labour and offering trade credit by being paid at the end of the week, or month.

"Now suppose that all goods are services, like haircuts, and it is physically impossible to store unsold haircuts."

A barber sitting around his shop waiting for clients *is* idle inventory, no?

=Peter

A great post!

The whole macroeconomic reasoning is a logical mess. Things need to change.

Starting with inventory. By definition, inventory means supply is always greater than demand. The supply curve will never cross with the demand curve. There is no market equilibrium characterized by Supply=Demand. Yet the market works just fine!

In macro, the inventory is a small story. The real big thing is spare capacity or slack. The service industry like hair cutting always has some slack.

@JJ Wayne:

> Starting with inventory. By definition, inventory means supply is always greater than demand.

No it doesn't, because inventory is sometimes desired -- it's not on the market.

Look at the real-life example of candymakers: they accumulate inventory all year to buffer expected peak demand around Halloween. During the rest of the year, that inventory simply isn't on the market. They're unwilling to sell it at the price that they are currently charging.

That in no way prevents an equilibrium, market-clearing price where quantity actually offered for sale equals the quantity actually asked-for.

It's *undesired* inventory that pushes prices out of equilibrium, but if that inventory is finite and demand is in any way elastic then that departure from equilibrium will also be finite.

Hmm, this sounds awfully close to 'buffer stocks' and 'saving desires', that PKers like to refer to when talking about commodity prices and the buffer of monetary savings.

I'm not sure that being 'off' the IS or LM curves is any more kosher than having identities that are temporarily suspended? Why not forget them curves altogether? And the equilibrium bit. Unicorns, if you ask me.

But putting that aside, it would seem that the main point of disagreement is not in the diagnosis but in the subsequent therapy. Monetary vs. fiscal policy, transmission mechanisms, base vs. bank money and all that jazz. If I squint hard enough, I think I can make out a miniature step forward. But then again, maybe it was just me being a step behind all along.

Peter: "A barber sitting around his shop waiting for clients *is* idle inventory, no?"

No. Because he does not represent past production.

But it's an interesting question whether we should count him as producing anything. He is producing the *option* for anyone to walk in and get a haircut instantly. But then the unemployed barber, with no job at all, is also available to hire, though the lag would be longer.

Are the fire brigade, when sitting waiting for a fire, producing anything?

Oliver: "Hmm, this sounds awfully close to 'buffer stocks' and 'saving desires', that PKers like to refer to when talking about commodity prices and the buffer of monetary savings."

I'm sort of trying to maybe find some common ground with the PKs, with some of these posts.

> But it's an interesting question whether we should count him as producing anything. He is producing the *option* for anyone to walk in and get a haircut instantly.

I think that the answer here has to be 'no'. The barber is not in any way receiving compensation for their time by just having the store open. Saying that the barber is producing the option of having a haircut is double-counting the haircut itself.

> Are the fire brigade, when sitting waiting for a fire, producing anything?

Weirdly, I think the answer here is 'yes'. The difference is that people are willing to pay for the availability of fire protection, regardless of whether there is an active need upfront.

They're being paid on retainer, like an attorney or an on-call doctor. Since emergency barbering is not in so much demand, the open-but-empty barber shop is not producing.

@ Nick

Sorry, didn't mean to sound derisive. In my view you have found some common ground.

Re this:

Old Keynesians waved their hands and talked about inventories of goods. Old Monetarists waved their hands and talked about hot potatoes.

Your point is that these are related concepts?

Majro: we wouldn't want to double count. But we could see the waiting around as an input into producing a more valuable haircut. I am prepared to pay more for a haircut if I don't have to wait or make an appointment.

Oliver: yes, related. Symmetric.

Imho, a buffer stock does not imply a hot potato effect.

I am reminded of this ongoing debate between you and others:

http://uneasymoney.com/category/law-of-reflux/

> But we could see the waiting around as an input into producing a more valuable haircut.

Perhaps as a probabilistic one? From your haircut's perspective, it doesn't matter if the barber walked into the shop 30 seconds or 3 hours before you showed up.

But as an input, I find this model to be a bit cumbersome. It would instead seem easier to model haircut-demands as a random process, and barber availability determines whether or not the demand can be met with a product. This is frustrating in that it turns a formerly deterministic model into a non-determinstic one, but on the other hand it provides a direct reason that the barber may not immediately notice a drop in demand.

Major max: the Econ world is in deterministic. That is why your indeterminists model looks so much more realistic. I would challenge you to expand your in deterministic model to replace old Keynesian macro models.

I have taken a similar step. Please take a look.

Economics has immutable laws: physics laws of social science. Please check out
http://econpapers.repec.org/paper/pramprapa/47811.htm


Economics has a universal mathematical framework like Maxwell’s Equations for
electromagnetism: a fundamental equation of economics. Please check out
http://econpapers.repec.org/paper/pramprapa/50695.htm

Majromax: Regarding your candy maker example. To me, because the candy market is not the perfect competitive market, the market equilibrium concept does not apply in the candy market. In the real Candy market, the supply is almost always greater than demand. When is your last time seeing the candy stores sold all their inventory? No. It almost never happens in the real world. As long as you are willing to buy at the market clearing (retail) prices, the supply is almost unlimited. You can try to buy 1000 lb of candies at the retail prices in Walmart, no problem.

As long as there are inventories or slacks like idle hair cutters, the market cannot be treated as the perfect competitive markets. There is no market equilibrium either because the supply is always greater than demand. Paul Krugman is calling a temporary equilibrium over NYT site.

Much of this discussion is about terminology. I’m not an economist but I spent many years working in a professional services business. Here is how I think about it.

Assume a hairdressing business has one shop (a physical asset), three chairs (physical assets) and three hairdressers (human assets or staff). Assume that the shop opens eight hours per day and that it takes 30 minutes to cut someone’s hair. The business has a maximum capacity of 48 customers per day.

A key measure for this business (and most other service businesses) is asset utilization. If the business has an average of 36 customers per day then average utilization of chairs and hairdressers is 75% (= 36 / 48).

The business will hold an inventory of raw materials e.g. shampoo. However, it is in the business’s interest to minimize this inventory to reduce costs. Any efficient business will seek to do this.

As you say, the concept of an inventory of finished goods does not apply to a service business. An inventory of finished goods is made up of goods which have been produced in advance of a specific customer demanding those goods. This makes sense in a manufacturing business but not in a service business.

Even in manufacturing, inventories of finished goods represent sunk costs which could otherwise be put to better use e.g. as money earning interest in the bank. The goods in the inventories also require storage space which represents unnecessary cost. They can also be damaged or stolen which represents unnecessary risk. Hence, any efficient business will also seek to minimize its inventory of finished goods.

Some businesses manufacture goods only to order e.g. cars, furniture. They do not build up any inventory of finished goods. Other businesses manufacture to stock e.g. car fuel. We all expect our local service station to have fuel when we turn up to refuel our cars. However, even here, an efficient business will try to manufacture and deliver their product ‘just-in-time’ in order to minimize inventory costs.

The advent of sophisticated computer systems means that efficient manufacturing businesses detect changes in demand almost immediately and adjust production appropriately. As they expect their raw materials to be delivered ‘just-in-time’ for their manufacturing processes, they also immediately stop ordering raw materials. Hence, a wave of information flows rapidly through the economy from customer to supplier to supplier’s supplier etc. This leads rapidly to a reduction in asset utilization in the various businesses involved.

Hence, inventories are irrelevant in understanding recessions in service industries and not very important in understanding recessions in manufacturing industries. The key issue in both is asset utilization.

I don’t think that the inventory analogy between goods and money is a good one. It is in the interest of any efficient business to minimize its inventories of raw materials and finished goods. I can’t see any similar consideration about an inventory of money. Indeed, one of the main reasons for reducing the inventories of raw materials and finished goods is to increase the inventory of money or to allow the money to be put to more productive uses. In a recession, there may be no other more productive uses so the money might just accumulate.

Finally, if a recession represents low asset utilization then it is unlikely that businesses will seek to invest in new assets during a recession, irrespective of interest rates or the amount of money in circulation. It is for that reason that a non-economist like me suspects that it is only an increase in consumer demand for goods and services that will pull through business investment and, hence, increase the demand for money to fund that investment. It’s difficult to see why an increase in the supply of money alone pushes businesses to invest unless that money directly increases consumer demand.

Jamie: A good writing! I agree everything you said. You are describing how the real economy works.

Only when you try to abstract the reality into a economic model, the argument starts. Not just folks here. On NYT web site, Nobel prize winner Paul Krugman are having the same troubles. He is arguing a concept called temporary equilibrium.

My take is that all these confusions are all because economics is not really a science. Otherwise, a giant like Paul Krugman would not argue about these simplest concepts. These concepts should be settled long time ago.

Jamie: you are absolute right that assets utilization is the key. And business inventory is not very important. However, just before the recession starts, unsold business inventory usually surges. This is one of key reliable indicators of the recession is on its way. When the recession is about to end, the unsold business inventory tends to recover first. Basically, the businesses can tell that the inventory is too low, they regains some confidence about the demand and starts to produce again. The restocking inventory is usually a classic ending of a recession.

JJ: Read up on monopolistic competition. Start here.

Jamie: "The business will hold an inventory of raw materials e.g. shampoo. However, it is in the business’s interest to minimize this inventory to reduce costs. Any efficient business will seek to do this."

No it will not. Otherwise inventories would always be zero. Businesses maximise profits. There are both benefits and costs to reducing inventory.

" It is in the interest of any efficient business to minimize its inventories of raw materials and finished goods. I can’t see any similar consideration about an inventory of money."

No. Otherwise we would hold all our wealth in the form of money, and none in stocks, and bonds, and houses, etc. Again, there are benefits and costs to holding more money.

Business investors typically do not have an infinite time frame to recover profits. Sure a hair dresser could speculate on shampoo prices and hold a larger than needed inventory waiting for those prices to go up. But shampoo is semi-durable (will eventually decompose) and hair dressers don't live forever. In real life, JJ is correct, businesses tend to hold a minimum inventory because the known costs (storage, depreciation, etc.) usually outweigh the unknown benefits (where will prices be tomorrow).

Curiousity, given that businesses are profit maximizers, are they also loss minimizers? During a recession, what costs are lowered first?

Frank: "In real life, JJ is correct, businesses tend to hold a minimum inventory because the known costs (storage, depreciation, etc.) usually outweigh the unknown benefits..."

No.

Principle of economics #1: there are tradeoffs.

Principle of economics #2: Rational people (and firms) look at *marginal* costs and *marginal* benefits.

Stop now.

JJ Wayne,

Thanks. I think that economics is a science. However, it is a science of a complex evolving man-made system rather than a simple unchanging natural system like physics or astronomy. I think that the role model should be Charles Darwin rather than Isaac Newton. Two aspects in particular are important here: evolution and variety.

Evolution means that economic behaviour will change over time. For example, a car manufacturer in 2014 might make cars to order only so will not build up inventory. However, the ability to do that depends on sophisticated computer systems which did not exist in many previous recessions. Hence, there may not be inventory build-up at the start of a recession now whereas there would have been in the 1930s or the 1970s.

Variety means that different economic actors and different circumstances will result in different behaviours. For example, if a car manufacturer accepts orders only from its own dealer network then there will be no build up of inventory at the start of a recession. However, suppose a clothes manufacturer makes clothes for a retail chain and has a three month contract to make a certain amount of clothes. It will continue to make these clothes even if a recession begins during the contract period and the clothes retailer will then suffer an inventory build-up.

I agree that it is confidence in demand which causes businesses to start producing again at the end of a recession. The key driver of any business is its own demand forecast. This drives both investment and operational planning.

Frank Restly: “given that businesses are profit maximizers, are they also loss minimizers”

Businesses need to manage cash flow as well as seeking to maximize profits. In my terminology I would call this an example of risk management rather than loss minimization. Another example might be retaining reliable and skilled staff during a downturn even if that hits short term profits. Profit maximization without appropriate risk management leads to Enron, banking system collapses and similar failures.

Nick Rowe: “No. Otherwise we would hold all our wealth in the form of money, and none in stocks, and bonds, and houses, etc”

In practice, a hairdressing business, or any other small business, is unlikely to buy stocks or houses or take any sort of risk with its cash, particularly during a recession. It’s main uses of money will be a) management of cash flow, b) funds to insure against a prolonged recession, c) funds to expand the business after the recession e.g. to refurbish the shop or buy a new shop. I’d think it would be likely to keep its cash in a low risk liquid form to allow it to reinvest in the business at the most advantageous time.

Nick Rowe: “Principle of economics #1: there are tradeoffs.”

You’re right. There are trade-offs. However, inventory represents a cost to a business, so any efficient business will seek to reduce inventory within the constraints of the logistics of its business.

For example, I went to my local coffee shop the other day. I usually have a toasted sandwich with my coffee. However, there were no toasted sandwiches available. The reason was that the shop had not had its daily delivery of supplies i.e. there was no spare inventory at all and failure of a single delivery resulted in a stock-out. The coffee shop could have avoided this by, say, having a freezer and stocking it with product for use in emergencies. However, they have obviously decided that trade-off is not worth it. Hence, I didn’t get a toasted sandwich and the coffee shop didn’t get the revenue for that sandwich.

@Jamie:

> Hence, I didn’t get a toasted sandwich and the coffee shop didn’t get the revenue for that sandwich.

I fail to see how that's meaningful in this context?

We're implicitly assuming a riskless, demand-limited economy like that described in Nick's linked post re: monopolistic competition. Orders for bread and cheese are always fulfilled, on time.

This discussion focuses on what happens when you go into the coffee shop and *don't* order the toasted sandwich. It's the demand that exogenously falls, whether on a temporary basis (you forgot your wallet) or a permanent basis (you're going on a diet).

Jamie's post said: "Nick Rowe: “No. Otherwise we would hold all our wealth in the form of money, and none in stocks, and bonds, and houses, etc”

In practice, a hairdressing business, or any other small business, is unlikely to buy stocks or houses or take any sort of risk with its cash, particularly during a recession. It’s main uses of money will be a) management of cash flow, b) funds to insure against a prolonged recession, c) funds to expand the business after the recession e.g. to refurbish the shop or buy a new shop. I’d think it would be likely to keep its cash in a low risk liquid form to allow it to reinvest in the business at the most advantageous time."

I agree. That can apply to large businesses like Apple too although Apple may do something like "ladder" its gov't bond (low risk) purchases. They are basically looking for the highest yielding asset that will not go down in value or default.

Jamie said: "Businesses need to manage cash flow as well as seeking to maximize profits."

Exactly. I believe Nortel would be a good example.

http://www.veritascorp.com/_files/file.php?fileid=filePNzCPQMsVp&filename=file_Pro_Forma_Numbers_Masked_Nortel_s_Woes.Mar.26.02..pdf

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad