« Hardball, lowball and the allocation of Nortel's assets | Main | Sterilised, unsterilised, beggar-thy-neighbour, forex market interventions. »

Comments

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

Lenders have responded to ultra low rates by increasing leverage.

5) The supply of broad money increased because of the relaxation of reserve requirements. Therefore supply expanded--the shift in supply schedule lowered rates.

6) Rates fell because of increased monetary pumping, therefore supply expanded--the shift in the supply schedule lowered rates.

That model is way too simple minded.

There is not one interest rate, there are a plethora of them, depending on the instrument of debt in question. In particular, for lenders into the consumer debt market, the interest rate is exceedingly sensitive to the credit worthiness of customers, which is why loan sharkers can compete.

Consumer credit is thus limited by the demand of credit worthy consumers, who are finite. Consumer rates are always(?) higher than rates on instruments like T-Bills. So banks and others will always lend to the limit to credit worthy customers. If they have more money than that (or probably more accurately, the amount of money they have greater than that) will flow into T-Bills and similar instruments. Call that reserve debt. So, when interest rates fall, the credit worthy demand more, and money flows out of reserve debt into consumer debt.

This looks an awful lot like the simple minded explanation you complained about.

The total amount borrowed and lent is an accounting identity but the amount that is in each market segment is not.

Simple, perhaps, but not simple-minded: introducing different asset types wouldn't affect the conclusion. It just makes the analysis even harder to follow.

rp: "Lenders have responded to ultra low rates by increasing leverage."

If I use leverage, I borrow instrument A, and lend instrument B, then there must be someone else who is willing to lend instrument A and borrow instrument B. Why should low interest rates make me want to go one way and the other guy want to go the other way?

Jon: Suppose we have an increase in base money (currency) through (say) an open market purchase of bonds by the central bank. Bonds in public hands falls, and cash in public hands rises by the same amount. If you count cash as "debt", then total debt stays the same. If you don't count cash as debt, then debt falls.

Broader money is equivalent to leveraged lending. Same response as to rp.

Jim: There are lots of different types of apple. But it makes no difference to the analysis. For each type of apple, quantity bought = quantity sold. If we want to explain why quantity of apples traded increased, we must explain why BOTH quantity demanded AND supplied increased. Same with different types of debt.

Stephen is absolutely right. We need to make sure we understand the simple stuff first, before we get into the hi-falutin stuff. Right at the beginning of Mankiw's intro text, before he even talks about supply and demand, he has a chapter explaining why people trade (comparative advantage). That's all about differences between people motivating trade. Differences between people explain debt. We have to get the fundamentals right, before we get lost in the fog.

Yes it's simple. Simple is deep. Simple is good.

Fractional reserve banking system and the moral hazards introduced by the government -- enough said.

Governments around the world are desperately propping up the worldwide credit bubble.


But most Ph.D.'d economists are demonstrably incompetent when it comes to the logic of choice between longer time taking production processes which promise greater value and output and shorter time taking production processes that produce less value and output. You see economists blundering here all the time, avoiding the topic, and just not getting it. Take Keynes, for example.

The key fact is that artificially expanded credit at artificially lowered interest rates changes the demand between longer production processes and shorter production processes.

Does even one economists in a thousand deal with this? Does even one economist in a thousand even think of it?

Jon: Suppose we have an increase in base money (currency) through (say) an open market purchase of bonds by the central bank. Bonds in public hands falls, and cash in public hands rises by the same amount. If you count cash as "debt", then total debt stays the same. If you don't count cash as debt, then debt falls.

I don't count cash as debt, but the creation of cash moves the AS curve. Monetization by the CB shifts the supple curve RIGHT (it creates money) which allows us to travel up demand schedule and satisify more loans at with a lower equilibrium rate.

So I make two specific objections to your challenge:
1) Switching to publicly held debt changes the terms of debt
2) You ignore that monetization moves the equilibrium rate. For very reasons you give in your post, that happens because AS increases. This is so fundamental, I find your resistance; it is not 1-1--otherwise how do CB move the interest-rate!

mh: "Fractional reserve banking system and the moral hazards introduced by the government -- enough said." Not enough to count as an explanation, I'm afraid.

Greg: I face exactly that "..logic of choice between longer time taking production processes which promise greater value and output and shorter time taking production processes that produce less value and output..." right now (there's a bottle of wine aging nicely in my cupboard). But if everyone is identical, and has a bottle of wine just like I do, there's no debt.

Jon:

Clarification first: by "AS" you don't mean "aggregate supply of goods and services" do you? (I don't think you do, because monetary policy shifts the AD curve.) Do you mean "supply of loanable funds"? Or was it a typo?

Well, everyone is different, and then entry points for cheap credit and expanded leverage makes people even more different and increasingly different --in an unsustainable fashion, esp. when new profits are coming where they can't be sustained.

Say you got an adjustible rate 2nd mortgage at well below the natural rate, allowing you to suck down your wine, buy three more bottles, using one as collateral to borrow a six pack you hope to
pay for with profits from aged wine.

Suddenly money and credit gets tight, because lots of folks were attempting to expand consumption and long period investment at the same time, but money had come due. You have to sell your wine to cover the mortgage, now not later. You can't pay for the beer and lose your collateral bottle. You go
bust. What looked like doable and profitable consumption and investment wasn't.

"So too many apples got bought and sold. Rather than eliminating a tax, financial markets went further and created what looked like a subsidy."

Would this help explain the following:
1) The idea that Subsidized GSE Loans led to private lenders acting as if they were subsidized.
2) The idea that lending standards decreased or held steady as the price of houses rose.
How can you explain these events without an implied subsidy?

The number of lenders do not equal the number of borrowers, because in a fractional reserve system, the banks create money. They will lend out as much as they feel like.....

The CDIC (FDIC in the US) insures savings so savers don't mind receiving a low interest rate, and more importantly they don't care how the banks use their capital. The banks of course make piles of money by lending out these funds many times over at higher rates.

http://americacanada.blogspot.com/2009/07/cmhc-and-our-government.html
"The only growth has been in the securitization of Canadian mortgages [from 2007 to 2009]. In Canada this scheme has worked very well despite the credit crisis since the government of Canada insures 100% of any losses (not just the 20% downpayment). This means that the securities are as secure as government bonds, yet pay a higher premium (currently 3.1%). "
"While many banks were flogging that it was a great time to buy a home, not one of them increased their mortgage holdings. Between the beginning of 2007 and 2009 Canadian Banks increased their total mortgage credit oustanding listed on their books by only 0.01% (see CMHC chart below). One has to question if real estate was such a great investment, why didn't they want to touch it?"

Can we say moral hazards? Profits to the banks, and losses to the taxpayers.

The combination of fractional reserves and moral hazards are more than enough to explain why our society is in so much debt.

Not to mention, for a fiat currency system to remain stable, credit (or debt) must be increasing at an exponential rate so that all of the outstanding debt can be serviced. At some point, debt must be deflated or there will be a currency crisis.

Greg: "..and then entry points for cheap credit and expanded leverage makes people even more different and increasingly different." Why? Why do lower interest rates (caused by what?) increase those differences, or accentuate the effects of those differences and increase total debt?

"Say you got an adjustible rate 2nd mortgage at well below the natural rate, allowing you to suck down your wine, buy three more bottles, using one as collateral to borrow a six pack you hope to
pay for with profits from aged wine." Sure, I want to borrow more when interest rates fall, but who wants to lend more?

Don: I'm not sure what a GSE is. (It still stands for General School Exam in my poor British brain). GSE "O" levels ("Owls" for Harry Potter fans) were what scared me at age 15.

If the government subsidises apple sales, I will sell apples to myself, just to capture the subsidy. The key is that the price to the seller can be higher than the price to the buyer when there's a subsidy to a sale. But I am using "subsidy" as a metaphor to help us understand why the volume of loans might increase why rating agencies (for example) say that a mortgage is AAA rated when it's really very risky.

mh: still way too complicated. You guys must learn to simplify.

"The number of lenders do not equal the number of borrowers, because in a fractional reserve system, the banks create money. They will lend out as much as they feel like..... " For every dollar borrowed a dollar is lent. This is just two ways of describing the same transaction. I write "IOU" on a bit of paper. I sell it to you for $5. You buy it from me for $5. I just borrowed $5 from you. Fractional (currency) reserves just has to do with whether or not the liabilities function as media of exchange. That's a totally different question.

"The combination of fractional reserves and moral hazards are more than enough to explain why our society is in so much debt." Nope. Fractional reserve banking and moral hazard are nothing new. Why would they increase both the demand for and the supply of loans? (Unless it is like my subsidy argument?).

"Not to mention, for a fiat currency system to remain stable, credit (or debt) must be increasing at an exponential rate so that all of the outstanding debt can be serviced. At some point, debt must be deflated or there will be a currency crisis." Nope. That's not correct at all. It is easy to have a model with constant debt and money.

From mh


The number of lenders do not equal the number of borrowers, because in a fractional reserve system, the banks create money. They will lend out as much as they feel like.....

The CDIC (FDIC in the US) insures savings so savers don't mind receiving a low interest rate, and more importantly they don't care how the banks use their capital. The banks of course make piles of money by lending out these funds many times over at higher rates.

You can't leave out institutions and expect to get reality.

The apples analogy fails because apple types are not fungible. I cannot create apple type a by selling apple type b. I can create loan type a by selling loan instrument b, especially if I'm a bank.

In a reasonably functioning economy, consumer credit supply (for credit worthy customers) appears infinite. Given the relatively high interest rates and the relatively low default rates (note: no limited liability, bankruptcy is painful) the limit to consumer credit is how much consumers are willing to take on. Consumers not under severe stress will limit their borrowing to what they can support or less. Explanation 4 is pointed in that direction.

If consumer credit supply starts to be less than consumer credit demand it's probably a clue that the economy is screwed in some way, at a guess, a way station on the road to hyper inflation.


>>>" Fractional reserve banking and moral hazard are nothing new. Why would they increase both the demand for and the supply of loans? (Unless it is like my subsidy argument?)."

I guess you can call it a subsidy, but such a subsidy is unstable and will explode in the government's face when loans go bad.

>>>"That's not correct at all. It is easy to have a model with constant debt and money."

Then can you explain how debt can serviced without growing the money supply? In a simple closed system with 1 bank and finite borrowers, where does the money come from to cover the total interest of the loans if every single borrows decide to pay back their debt. My guess is that you need new borrowers to take on debt to create new money. Wouldn't that just be one big ponzi scheme?

>>>"In a reasonably functioning economy, consumer credit supply (for credit worthy customers) appears infinite. Given the relatively high interest rates and the relatively low default rates (note: no limited liability, bankruptcy is painful) the limit to consumer credit is how much consumers are willing to take on. Consumers not under severe stress will limit their borrowing to what they can support or less. Explanation 4 is pointed in that direction."

Inherently borrowers are not financially savvy so they will borrow until they reach the point of no return. kinda like a frog in water, and you slowly crank up the temperature. by the time they reach their limit, they will probably default. Or they can simply be surprised by the loss of income due to unemployment.

Debt accumulation due to asymmetric incentives (like what is happening today) will inevitable cause a credit bubble (if not already); the bubble will inevitable deflate either via defaults or aversion to risk.


1. Perhaps not people becoming different but different people, some who didn't have the money to lend before now having it and wanting to make more money by selling to those borrowers, and some converting their present apples to apples in the future.
2. Really both better and "better", especially when driving the price of apples up to lend more for their purchase.
3. A Canadian thing? Though since these people, the lenders, were largely foreign central banks, one could call it forced.
4. If everybody wanted to lend more, who are the borrowers? The least among them?

Primarily 1 and 2 I would say.

Do you mean "supply of loanable funds"? Or was it a typo?

I meant the supply of loanable funds.

Well, rising inequality drives your first point about greater differences between borrowers and lenders. No surprise that inequality rose along with debt levels before the great depression and also before this latest blow-up.

But as I've argued many many many times here before, similar to some of what Jim and mh said above, only an increase in the willingness to borrow (demand) is required, there is no need for an increase in the willingness to lend (supply).

Demand for loans creates its own supply (deposits). I go to the bank to borrow $500,000. The bank deposits $500,000 in my account. Where did the supply for my loan come from? It came from my deposit which was created by the loan!

Note that this is not how things work with apples or any other item. Apples don't simply get created out of thin air at zero cost *because* I want to eat an apple. If they did, the supply curve would be pretty flat and quantity would be determined by demand alone.

Theoretically, in the absence of any effective reserve requirements, the only real limit on the amount of debt is either the amount of willing, creditworthy borrowers there are or the willingness of the banks themselves to keep increasing their leverage (re: your earlier comment, note that banks create their own leverage, they don't have to borrow from anyone to do it - they're banks, that's the whole point!) - and a government guarantee that banks wont be allowed to fail can certainly mitigate the leverage concerns of the banks. Once these run out then the credit cycle turns, interest rates rise and either you get a bunch of defaults to bring the debt level back down, inflation to bring the debt back down, or, in the present case, the government guarantees all the debt and floods the banks with cash at even lower interest rates in an attempt to keep the party going a little longer.

So debt has gone up (I assume we're talking household debt here) primarily because demand for debt has gone up. My opinion is that the increase in demand is mostly down to cultural factors - we have a consume now because you're worth it culture and it's been a long time since the last big credit blow-up. I remember my mom mentioning that when they got their mortgage the idea of borrowing to buy a house rather than paying cash was still a new concept and considered somewhat risky/risque - that was one generation ago.

Also, low inflation means that old debt isn't being inflated away very fast, and it also means that people who only look at the monthly payment costs are being lured into doing more borrowing (I was surprised when I went to buy a car how difficult it was to get the car dealer to actually talk in terms of the price of the car rather than the size of the monthly payment).

Support for #1 might be found in the increasing income inequality since the 1980s, as income shares have become more and more concentated at the top of the income ladder.

During the housing bubble, people borrowed heavily not only to buy houses but also to compensate for the weakest job and income growth of any expansion since the end of World War II.

On the supply side, as incomes polarized, America's rich found their portfolios bulging with cash in need of a profitable investment outlet, and one of the outlets they found was lending to those below them on the income ladder.

http://economistsview.typepad.com/economistsview/2008/10/does-wealth-con.html

Great to see the comments rolling in. It was getting a bit quiet here. "Debt" always brings out the debate.

My brain can't yet focus on the money-debt-loanable funds-fractional reserve nexus. I need to say something right, and also say it simply. And it's an important enough topic to maybe deserve a post in its own right, rather than buried in the comments.

But on the meme: "increased income/wealth inequality caused the increase in debt". I have heard this argument a lot, and it too makes little sense to me. Or rather, it could only make sense if people made the distinction between permanent and transitory income. According to the permanent income hypothesis, an increase in permanent income causes an equivalent (proportional) rise in consumption. A change in the distribution of permanent income should have no effect on debt.

But a change in the distribution of transitory income would certainly affect borrowing and lending. If transitory income became more unequal (e.g. there were bigger transitory shocks to individuals' incomes, those with positive transitory income would save, and lend to those with negative transitory income, who dissave.

Asset values are rising faster than incomes. Younger people are taking on more debt to buy homes. Depression-era cohorts are being replaced by people more comfortable with debt.

There is also a general rise in credit card balances, much of which is people paying with credit for Air Miles etc. and the balance is cleared every month. Is this credit or just a change in purchase behaviour? 10 years ago I paid for everything with debit, now I pay for everything with credit and save 1% on every purchase but have a large CC balance every month. Am I one of those crazy people 'racking up expensive debt'? No.

Two good papers from the BoC. This one measures debt/asset patterns among age groups:
http://www.bankofcanada.ca/en/res/dp/2009/dp09-7.pdf

This one tracks total debt service ratios (and not just interest costs). http://www.bankofcanada.ca/en/res/wp/2008/wp08-46.pdf
Good fodder against those saying that people are getting bogged down in debt. The year 2007 was very similar to 1999 and the risky portion of the DSR curve is very small.

"The government-sponsored enterprises (GSEs) are a group of financial services corporations created by the United States Congress. Their function is to enhance the flow of credit to targeted sectors of the economy and to make those segments of the capital market more efficient and transparent. The desired effect of the GSEs is to enhance the availability and reduce the cost of credit to the targeted borrowing sectors: agriculture, home finance and education. Congress created the first GSE in 1916 with the creation of the Farm Credit System; it initiated GSEs in the home finance segment of the economy with the creation of the Federal Home Loan Banks in 1932; and it targeted education when it chartered Sallie Mae in 1972 (although Congress allowed Sallie Mae to relinquish its government sponsorship and become a fully private institution via legislation in 1995). The residential mortgage borrowing segment is by far the largest of the borrowing segments in which the GSEs operate. GSEs hold or pool approximately $5 trillion worth of mortgages"

When you say this:

"the volume of loans might increase why rating agencies (for example) say that a mortgage is AAA rated when it's really very risky"

That's what I'm saying as well. In the US, there has been a big debate about whether GSEs led to private lenders lowering their lending standards. I have argued that the private lenders acted AS IF they had the subsidy that GSEs actually did, and that meant that, if true, they were fooling themselves about the risk in lowering lending standards.

My other point is the same: Angelo Mozilo, the head of Countrywide, the largest home lender in the US, I believe, noted that the demand for loans rose as interest rates went up. There is a big argument here as to why this happened as well, but I would argue that the lenders again took excessive risk while downplaying it. Again, they acted as if the govt had an implicit guarantee, a subsidy, to help them if these loans went sour.

There is also the real question of whether or not banks took excessive risk because they believed that the govt had implicitly guaranteed them, agreed to subsidize them if necessary. I believe so.

It is possible that I'm simply taking your metaphor literally. If so, I still plan on using it.

Take care,

Don

Isn't there the distinct possibility that there was an excess of people willing to lend, but unable to find borrowers - a problem which the lower interest rates solved, thus making the original explanation correct?

In our economy, it always seems like there's loads of people and institutions constantly trying to get us to borrow their money. While lower interest rates may reduce the interest in lending, it would require a whole lot of reduced interest to actually make the two sides balanced.

Hi, nice post, and very thought-provoking.

You say "But on the meme: "increased income/wealth inequality caused the increase in debt".
But a change in the distribution of transitory income would certainly affect borrowing and lending. If transitory income became more unequal (e.g. there were bigger transitory shocks to individuals' incomes, those with positive transitory income would save, and lend to those with negative transitory income, who dissave.".

A paper (of mine, actually) published in JMCB last year makes exactly this point. The paper is available here
http://www2.bc.edu/~iacoviel/research_files/JMCB_2008.pdf
If you look at section 5.3, I discuss in the paper exactly the point you make. The increase in debt seen in U.S. data can be attributed to the rise in income inequality only if the rise in inequality comes from shock that are not permanent. To quote from my paper
"The increase in debt during the period 1984–2003
can be explained by the model only if the increased earnings dispersion comes from income shocks that are mean reverting. Put differently, the increase in inequality that has taken place in the 1980s and 1990s can quantitatively explain the increase in household debt insofar as it has resulted from an increase in the variance of the non-permanent component of earnings."

What about the world savings glut? Wouldn't an outward shift in the supply of savings from Asia, due to their rapid growth (and, say, sticky consumption), reduce interest rates and increase the quantity of debt domestically? Too simple? But very real, no?

I don't understand any of this.

I do understand that securitization, and leverage expansion created by low interest rates, (hey who wants low interest bearing AAA stuff when you can buy higher interest bearing AAA stuff)expanded the money supply and a great amount of it backwashed into the real economy. And that process allowed many people and businesses to expand their debt.

I have no clue as to where in this process equilibrium occured (if ever), but when the underlying assets rolled over a large amount of leverage and wealth evaporated. What remains is a good portion of "legacy" debt at all levels.

Again, sorry for my ignorance of basic econ 101.

Michael: Yes, but that's part of "1. People became more different in their willingness to borrow or lend." Doesn't seem to work for Canada though, because there wasn't a lot of borrowing from abroad.

Matteo: Thanks! So, if I read you correctly, there was empirically an increase in the dispersion of transitory income, and it was a big enough increase to explain the rise in debt? That's an important finding. Or are you not sure if it was transitory or permanent, but your model says (correctly in my view) that it would only explain the increase in debt if it were transitory?

Neil: "Isn't there the distinct possibility that there was an excess of people willing to lend, but unable to find borrowers - a problem which the lower interest rates solved, thus making the original explanation correct?" But if everyone were identical, interest rates would fall until nobody were willing to lend. You need an increase in the dispersion of willingness to lend and borrow to get more lending and borrowing.

Don: OK, I get you now. The government guarantee was indeed a subsidy, not just metaphorically. But did the guarantee become stronger during this period, or was it always there? If it was always there, it wouldn't explain the increase in debt, just its high level.

Mark: That is the sort of analysis of debt that needs to be done; the sort I praised the Bank of Canada for doing (in its recent Report) in a post a month or two back.

A post on "Money and Debt" is percolating through my mind, to try to address the other comments.

Beezer: "I do understand that securitization, and leverage expansion created by low interest rates, (hey who wants low interest bearing AAA stuff when you can buy higher interest bearing AAA stuff)expanded the money supply and a great amount of it backwashed into the real economy. And that process allowed many people and businesses to expand their debt."

How can you possibly hope to understand that complicated stuff, that you say you understand, when you say you don't understand the simple economics of supply and demand?

That illustrates exactly the problem. Before we claim to understand all that poncy stuff (I don't), let's start with some basics.

I suppose this is related to (1), but rather than people becoming more different from each other, they could be identical but have changes in their life cycle earning profiles, which could explain increased debt. Im thinking some sort of an OG model where income is higher during teh "old" period, and so the old lend to the young. Agents are identical, but if the earning difference between old and young increases, debt will increase.

Im not sure if this holds any water empirically, but could the increase in returns to education not explain increased debt through this model? Young people would be more likely to stay "poor" and borrow when their young but obtain more income and lend more when old.

Just a thought

Thinking of debt as a barter exchange between borrower and lender is kind of misleading, because it abstracts away from money and financial intermediation. Borrowers and lenders are not being matched in a market, where the interest rate is a price arrived at by tatonnement. Borrowing and lending are equal, because of an accounting identity, not a market-price equilibrium.

I don't believe in the quantity theory of money, but it provides a healthy shorthand. The quantity of money can increase, and debt is just money in an extended form. It might be useful to explore why a Society might benefit economically from having more money-debt in more diverse forms.

And, debt is a contingent contract for money flows over time. So, money debt has an insurance function, redistributing risk and income.

As you almost say, intermediaries may be borrowing and lending to/from the same people.

Intermediaries are not necessarily tapping a flow of "savings", so much as they are mining reservoirs of transaction balances, and offering insurance contracts. And, yes, the amount of debt may be related to changes in the distribution of income, if the distribution of risk is changing as a result. (We southerners refer to these phenomena as Casino America, aka the trick-and-trap economy.)

The amount of debt is not constrained as a matching problem between willing borrowers and lenders, but as a tradeoff of risk and expected income along a yield curve, with money-currency as its left-most fulcrum.

Hope this helps.

Nick:
Having worked for an investment bank, I get the investment side, the flow of money side, the influence of interest rates. But frankly what I learned there doesn't match up well with economics. I understand supply and demand when it comes to securities, but even with that knowledge your explanation of supply and demand befuddles me.

Again, sorry for my imbalance. Been around debt too much. Maybe I'll go have a beer and read the thing again.

We have all had some experience with debt so we think we understand it. It is clear from this thread that there is a lot of confusion.

Let me start with some facts. To a bank or a financial institution, debt is an asset. They grow by increasing their debt assets.

While fractional reserve banks have grown enormous, they are not the greatest source of debt increase.

GSEs have $5 trillion in debt assets, the Federal Reserve estimates that the total US debt is $50 trillion, and there are estimates that financial derivatives total $200 trillion.

Debt has a time dimension that is easily overlooked. A borrower may use future income to increase current consumption. Even if that future income does not materialize, the borrower may enjoy that current consumption. Financial brokers receive fees for creating / intermediating the debt. Investors grow by using borrowed funds to buy the debt.

The financial market does not behave like the economic market. The price of apples may tend toward an equilibrium due to competing desires of buyers and sellers in the economic market place. In the financial market place, buyers buy more when prices increase and sellers are pleased when they get a higher price. The concept of equilibrium does not apply.

There is no savings glut. There is a glut of debt! Real assets have not been created as fast as the debt was created but now it is hard to know who has a real claim on those assets because the real (economic) market has been obscured by the financial market. All those excess financial assets amounts to a house of cards that is only held up by the belief it matters.

It is all so simply absurd.

The million dollar question: where are debt levels headed from here?

OK OK, I had my beer(s). So let's take #3. The left hand right hand one.
Right hand (China) has a high savings rate. It's a forced one actually because it basically is in retained earnings. China doesn't trickle down as well as it could. But then again, neither do we.

We're left hand (US). We borrow, right? No, no, that can't be correct because...
"Even if everybody is identical, there can still be debt, because each individual both borrows and lends money. He has a credit in his pension plan, and a debit in his mortgage."
.....because I definitely got the debit on the mortgage but my pension went into the toilet.

But then I kind of like #4 and #5 too. And #1 makes sense now that I get it. And I'm positive #2 happened.

Could it have been a little of each?

Cheers.

Curtis: Yes. I think that the life-cycle is the most important reason why debt exists. And even if people don't change (in their preferences), increased returns to education, and so increased investment in education, would definitely be one of the things I ought to mean in 1. There is a bigger difference between 20-year olds and 50 year olds today than there was in the past, because in the past both would be working and earning, while today the 20 year old is more likely to be at school and borrowing.

So increased returns to and hence investment in education would be one cause/explanation of rising debt. (My guess is it's not big enough to explain a lot, but that's just my guess).

Bruce: Yes, I wasn't always clear on the distinction between "borrowing = lending" as an accounting identity, and "desired borrowing = desired lending" as an equilibrium condition. I think I need both, because I also want to talk about the relation between changes in interest rates and the quantity of debt, and I can't get anything about changes in interest rates out of an accounting identity.

Yep. I need to bring in money explicitly. I feel a next post on that very subject welling up inside me....

Could changes in risk etc., or attitudes towards risk, explain (at least in principle, and maybe in practice) the increase in debt. Hmmm. Yes, probably, almost certainly, I think. Though my brain is not giving me any clear intuitive signals yet on how it would work, in a simple and clear way. Let's see...if the world became less risky, risk of default and moral hazard etc. should lessen, so transactions costs should fall, so debt would increase? But then the demand for insurance would also fall, and that might interact with debt??? Dunno. But it's gotta belong in there somewhere and somehow. At least theoretically.

"As you almost say, intermediaries may be borrowing and lending to/from the same people." Yep. I should have said that explicitly.

"The amount of debt is not constrained as a matching problem between willing borrowers and lenders, but as a tradeoff of risk and expected income along a yield curve, with money-currency as its left-most fulcrum." I disagree with you here. Sure, your yield curve picture makes sense, but at each point on that curve there is a particular financial instrument, and if we want to explain why the quantity of that financial instrument increases (why debt increases) we need to understand BOTH why more people want to sell it AND why more people want to buy it. AND, if total debt (all those instruments) increases in quantity, why this happens at all or most points along the curve. So we are back to the same question that I posed in its simplest form, by ignoring those differences between types of debt.

Beezer: Hmmm. Maybe you DO understand all that "poncy stuff"! ;)
I think it comes down to the distinction between partial equilibrium vs. general equilibrium analysis (micro vs macro). There are certain adding-up constraints that apply to the system as a whole that do not apply to each of the parts. Like income = expenditure, total borrowing = total lending, savings = investment, etc. I find in order to get a clear consistent picture of the whole, I am forced to simplify drastically each of the parts. But those who understand the complexities of some of the parts often ignore those adding-up constraints on the system as a whole.

Richard: But why did debt grow? Why were people willing to borrow more, and other people (or the same people?) willing to lend more? And both at the same time?

mh: "The million dollar question: where are debt levels headed from here?" EXACTLY! That is exactly where I wanted my post to lead. If we could understand why debt had grown in the past, we MIGHT (maybe) understand where it's going in future. I had wanted to continue my post to discuss that question, but ran out of steam. (Probably just as well, given the arguments around what could have caused it to rise in the past, let alone what did in fact cause it to rise in the past).

Beezer: The China/US theory of rising debt is a classic example of explanation 1: differences. If China were like the US, everyone would want to borrow, nobody would want to lend, so there could be no debt, and interest rates would rise until people stopped wanting to borrow. If the US were like China, everyone would want to lend, and nobody would want to borrow, so there could be no debt, and interest rates would fall. You need differences to get debt; the difference between China and the US, in that theory. (But it can't explain rising debt in Canada, where we weren't, in aggregate, borrowing from abroad).

Each of those explanations (plus some I have missed) could be part of the reason debt increased. They aren't mutually exclusive.

Nick Rowe: "at each point on that curve there is a particular financial instrument . . . we need to understand BOTH why more people want to sell it AND why more people want to buy it. "

The thing about a yield curve is that it is the foundation of intermediation, the age-old process of borrowing short and lending long. Intermediaries, like banks, borrow short and lend long, to balance the overabundance of people, who want to lend short and borrow long. So, there's no "each point" because intermediaries are engaged in (what they hope is) arbitrage between people, who want to lend short and people, who want to borrow long.

In your barter-loan model, if everyone were exactly alike in their financial preferences, there'd be no debt, because everyone might want to lend short and borrow long, and no one could be matched with anyone. But, here's the thing: Everyone can want the same thing -- pretty much every business wants to have cash-in-hand and working capital to do business, and to finance that with promises of future cashflow (long-term loans and a marketable equity stock). Intermediaries make that possible. To some extent, it is a confidence game -- banks insist that your deposits with them are as liquid as cash; then, they use those deposits to fund long-term loans to business, which are very illiquid.

It isn't a desire to save or borrow that drives this market, it is the differential utilities of risk. Confidence game or no, the banks economize on risk at both ends, by vast aggregation of deposits from people and firms with independent and predictable needs for cash, and by building a manageable portfolio of loans.

It is easier to discuss what a commercial bank appears to do, then to deal with the esoterica of trading in financial securities, intermediation and insurance and leverage and diversification are still at the heart of what is a process of risk arbitrage and "money" creation.

I wouldn't insist that all debt is money, but, clearly, a lot of debt functions as an extension of money in facilitating transactions. So, how much debt an economy "needs" is a lot like the question of how much money the economy needs to keep fully employed. And, how much debt an economy can sustain is very similar to how much fiscal capacity is necessary to credibly sustain the value of a fiat currency, or a national debt.

Fans of the model of financial markets as matching savers with borrowers like to imagine that interest rates equilibrate flows of savings from one to the flows of investment by the other. I don't think that's what goes on. It is more like there are vast reservoirs of cash or near-cash always piling up. Inflation poses a danger, and puts hydraulic pressure on those reservoirs, forcing them to seek a return in the financial markets. Then, the challenge to intermediaries is to avoid making (too many) bad loans, by following systematic underwriting standards and procedures. Interest rates are just a long chain of minimum offers from potential borrowers -- the minimum to best the expected return (or, more likely, slow loss) from the perfectly safe alternative. Every participant is both borrower and lender, saver and investor, engaged in managing a portfolio of risks against the opportunities presented by the financial markets to insure against risk and profit from escaping budget constraints.

"Both at the same time" isn't necessary to increase debt. One possibility is some combination of "easy money" on the short-end and a systematic relaxing of underwriting on the long-end. Then, people and firms will engage in counter-arbitrage to try to take advantage. Home owners, for example, might take out cheap home equity loans, for example. Pretty soon, the financial system is generating more near-money debt that can be funded from expected cash flows.

Let's take the simplest example of a financial intermediary: one that simply pools risky assets into (relatively) safe assets. And let's suppose all borrowing goes through that FI.

There are two ways of looking at it:

View A: The total supply of (risky) debt by the ultimate borrowers has to match the total demand for (safe) debt by the ultimate lenders, but the FI is a black box that can convert risky into safe debt, so the exact type of debt supplied by the ultimate borrowers does not need to match the exact type of debt demanded by the ultimate lenders.

View B: The supply of risky debt by the ultimate borrowers must match the demand for risky debt by the FI, and the supply of safe debt by the FI must match the demand for safe debt by the ultimate lenders. So there must be supply = demand for each type of debt along the spectrum.

I think Bruce is talking in terms of View A, and I am talking in terms of View B. They should be equivalent, provided we adjust who we include when we talk about "borrowers" and "lenders". Does it mean ultimate or all borrowers and lenders.

Either the lender or the borrower have been pricing credit/debt wrong.

AIG exploded, so to some degree the lender's insurer got it wrong. But lenders suffer the pain.

The rating agencies rated securities too favorably. They however kept all their fees, while Banks' balance sheets exploded. In this case, the rating agencies got it wrong, but the lenders took the hit.

If the lenders learned their lesson, going forward they should be more risk adversed so future borrowers will have trouble rolling their debt. Credit/debt will inevitable deflate... unless lenders believe that the government will bail them out.

Who is willing to continue to lend?

According to Brad Setser's posts/blog, recently it has been central bankers.

Is debt being concentrated in fewer and fewer entities?

I think you need to concentrate on more. That would be price inflation, wage income, asset prices, debt, ability to make interest payments, retirement dates, and living standards.

"Well, it is possible that what limited the amount of borrowing and lending in the old equilibrium was the requirement that borrowers be able to service the interest on their debt."

Yes and finally. This is why the fed and/or even the market (rich people trading interest rates) would have interest rates high. That would be so that more debt (future demand brought to the present) is NOT produced and therefore will NOT produce price inflation and if the labor market is tight enough wage inflation.

Assume one world bank that is a monopoly with a 0% or near 0% reserve requirement and a 0% or near 0% capital requirement. Also assume full recourse loans (no default and I think that is the correct terminology) and no bankruptcy.

Would this entity be able to produce UNLIMITED debt?

"My brain can't yet focus on the money-debt-loanable funds-fractional reserve nexus. I need to say something right, and also say it simply. And it's an important enough topic to maybe deserve a post in its own right, rather than buried in the comments.

But on the meme: "increased income/wealth inequality caused the increase in debt". I have heard this argument a lot, and it too makes little sense to me."

Keep thinking about more debt, fractional reserve, and income/wealth inequality. I believe that eventaully it will make sense to you.

Some reference posts for me:

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/07/why-the-bank-of-canada-should-rise-interest-rates.html

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/07/fiscal-multipliers-in-new-keynesian-models.html

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/07/greenspan-and-his-critics-again.html

http://worthwhile.typepad.com/worthwhile_canadian_initi/2008/12/hair-of-the-dog.html

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/06/nonsuperneutralities-an-open-invitation-to-austrians-especially.html

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/07/income-expenditure.html

"I still think income is a misnomer because when I think about wage income, corporate profit income, rent income, and interest income there is no debt repayment involved.

It seems to me that debt has affected both national income and national expenditure."

And, "Closed economy with 1,000 people. 1 is the fed, 1 is a bank CEO, 1 bank employee is for consumer debt, 1 bank employee is a trader who borrows short to speculate in financial assets, 1 is an owner of a corporation that produces everything, 5 are on the bank's board, and 990 are employed at the corporation. Throw a "fungible" money supply on the closed economy.

Everyone plans to retire at 65 at a certain standard of living. Productivity shock happens and everyone can retire at 62. The 10 get greedy. Instead of increasing CURRCIR and real wages, they "open" the economy to another economy with cheaper labor. The productivity shock and cheaper labor lead to better margins (either thru fewer employees, fewer hours, or lower wages), but the 10 need to convince the other 990 go into debt to keep quantities and prices slightly positive. If the 990 go into debt to maintain their standard of living in the present, they now will need to work longer.

It is somewhat negative real earnings growth (lower standard of living) and more debt on the 990 and extremely positive real earnings growth and excess savings on the 10. The 990 have to work longer to retire, and the 10 could retire at about any time but won't. Eventually, the 990 may get so far in debt they can never retire. They become permanent debt slaves to the interest payments of the 10.

Notice the "fungible" money supply skews towards too much debt. I think the closed economy's money supply also becomes "more fractional".

Does that make any sense? I typed it in a hurry."

Nick:

If there is an excess supply of money, which means that people have accepted media of exchange in payment beyond what they want to hold, and they are fixing to spend it, then given the right institutional framework, that can be matched by an increase in the real supply of credit. Yes, lending (money holdings) still matches borrowing (those borrowing money from the banks that issued the money,) but desired lending doesn't match desired borrowing (because those holding the money received it in payment and it is excess balances that they will spend.) Once the price level rises so that the real quantity of money falls to meet the real demand, then the real supply of credit falls too. The nominal credit supply matched by the quantity of money that was excess now buys less.

I think that is pretty much it regarding how fractional reserve banking interacts with your analysis.

Otherwise, I think your comments before about people who understand the complications (or don't) just fall into the fallacy of composition and fail to see how things add up.

People irresponsibly borrow more than they can afford and spend. Total spending rises. Total sales rise. Total production rises. Oh.. but who lent the money?

Those indebted people can't borrow anymore or don't want to. They want to pay down debts. They spend less. Business sell less. Production falls. Oh.. but what do those who would have lent to these borrowers do with the money? What do those receiving debt repayments do with the money?

An individual firm can "leverage" its "capital" by borrowing. It can purchase more capital goods and expand so much more. So, when all the firms leverage, then all spend more on real assets. Capital goods spending rises. More production, more employment....

But when the firms all start deleveraging and paying down debt, then they buy less capital goods and pay down debt. Less spending on capital goods, less production, less employment.

So simple....

But when the firms "leverage" by borrowing, who lends them the money? What would they have done with it? Maybe the firms should have expanded by selling stock to these people rather than borrowed. Maybe these people consume less and buy the debt of the firms. But it had to come from somewhere.

When the firms deleverage, who is receiving the money being repaid? What do they do with it?

It always comes down to an excess supply or demand for money. The implicit assumption is that people who lend are reducing their demand for money. When funds are repaid, those receiving payment hold the money. They are increasing their demand for money.

The epitomy of money is currency, with no market price and no explicit yield. Monetary disequilibrium is just assumed to be the other side of the coin of problems with debt.

And, of course, in reality, the only possible disequilibrium in credit is going to be the difference between the real quantity of money and the demand to hold it before prices and wages have adjusted enough to return to monetary equilibrium.

And the credit is just a side show to the real problem of the impact of monetary disequilibrium on the flow of nominal expenditurres on income and output, with the resulting disruption in real output and employment.

Of course, those focused on financial markets may have different interests--financial and intellectual.


I thought I was agreeing with Jim Rootham. The model is too simple, and too simple in the analytic sense of being insufficient, of failing to take account of necessary elements. Risk arbitrage and attenuation is such a necessary element, being a critical element of what money and money-debt accomplish, functionally, for the economy.

Imagine that I anticipated greater short-term variation in my labor income in the future. I'd want to prepare for that, financially, by holding more cash. I go to my bank and mortgage my house for $10,000, and deposit the $10,000 in my checking account. I've increased leverage, in financing my ownership of my home.

Total debt in the economy has increased. My balance sheet, and the balance sheet of the bank have changed, in mirror image fashion. There has been no meaningful flow of funds that I can see. No funds have passed a Warlrasian auctioneer, calling out "the" interest rate. (In fact, the interest rate I pay on the mortgage, and the interest rate I'm paid on my bank deposit, are set by the bank in an administrative schedule, with some reference to "market" interest rates. I don't necessarily expect anyone, other than bankers and fans of Tanta at Calculated Risk, to grasp the significance of administrative pricing.)

A lot of banking and business finance is exactly this kind of portfolio rebalancing. Banks, or the finance sector in general, by practicing arbitrage in borrowing short to lend long, are selling insurance as well as diverting savings to investment. It has great significance to the economy, because leverage concentrates residual risk on the owners of a business. One could argue that finance, in the Hayekian economy of coordination by "market" price for an economy in an uncertain world, is generating the incentive structure, transforming market prices for resources into contingent contracts. Finance enables the organization of a firm specializing in building houses, and enables that firm to offer the carpenter building houses an hourly wage at an efficiency rate, with a credible threat of being fired for not showing up on time or for slacking. But, I digress.

Unless you are confusing "fund flows" with accounting identity, your view and my view are not equivalent. Real fund flow -- by which I mean an actual deferral of consumption from current income to supply claims on current consumption to others in return for promises of future claims -- is insufficient to account for the amount of money and debt.

In addition to the fund flow, there's an oceanic pool of money and debt sloshing around, providing an insurance function. I don't know if you can see it, but your simple, fund-flow model implies a tightly-coupled system, not unlike a barter economy. Money and money-debt enable a loosely-coupled system.

(And, yes, selling insurance can slip-slide into running a casino, with all the usual implications for the distribution of income on full display at a payday lender or in the hopeless sink of a credit card balance. But, that's another rant.)

mh: I think your explanation is a variant of my 2. above, where financial markets get "better", and so reduce transactions costs, that are like tax on borrowing and lending. Only you argue that they acted like a subsidy, by mis-representing the quality of the loans, so the lenders thought the loans were better quality (lower risk) than they really were. Polishing the apples to fool buyers into thinking they will taste better.

Too much Fed: Have a look at my latest post, where I bring in money.

Bill: you and I think much the same way on these questions. That means we must be right;). Yes, a lot of the false reasoning is like a fallacy of composition, or what I referred to above as ignoring adding-up constraints.

I think you are right on the long-run flexible price implications of introducing fractional reserve banks into the model. No effect on the results for a change in monetary policy. But it may change the short-run effects. And the introduction of fractional reserve banking may itself change the equilibrium amount of debt. I'm going to try that in a future post.

Bruce: I see where you are going now, I think. In a closed economy, an increase in gross debt does not change aggregate net debt, which stays at zero (of course). But what you are saying is that an increase in gross debt might not even change any individual's net debt. Each individual increases both his borrowing and his lending, and by the same amount. Agreed. I considered a similar case in my "3. Forced Savings" example. But your example gets the same results without it being forced; it's a voluntary response to increased risk. (Presumably you could also get the same results through a tax system that encouraged people to be both borrowers and lenders at the same time).

I agree. I would see it as an increased demand for liquidity. It also matters, because if your interpretation is the whole truth (or a large part of it) of the increase in gross debt, then it's not at all true that even some of us are "increasingly living beyond our means". I'm not sure whether or not it works empirically. I don't think there was an increased perception of risk in the last few years (until the last few months).

I'm sure that what Bruce Wilder is (as ever) at least part of the truth. But I think it misses the big picture (and here I am old enough and have moved around enough to have a good perspective). What really has massively increased is mortgage borrowing, consumer credit is relatively small beer.

And it is very much a shell game.

My parents generation (the lucky generation born in the 1930s) benefitted massively by borrowing modestly to buy homes which were cheaply paid off (as inflation inflated wages in the 1970s in particular) and appreciated greatly - at least in nominal terms.

New entrants in the housing market, have much greater entry costs (as a number of factors - not least urbanisation and greater equality - have pushed up real housing values - against disposable income if not against raw family income). Lower NOMINAL interest rates have had some effect, because people forget that you have to pay off the real debt - not just service it. And then along comes lower real income growth and people suddenly people are in potentially big trouble.

I think most people just extrapolate from historical experience and have failed to notice major secular trend changes, let alone more dramatic changes (such as peak oil). People have bought the shadow of house ownership, where they only in reality get to own the liabilities (such as maintenance), but the bank owns the rest.

So perhaps I would put it like this is demand and supply terms, on the demand side there is a certain amount of irrationality, with unrealistically optimistic expectations being a big factor, on the supply side there is a big agency problem with incentives being to lend and pass on the "asset" no matter how dirty it is to someone else.

That made me realise of course - the real big point that is missing in the list above - speculative demand for money! People are borrowing to gamble!

reason: but on the supply side there must also be more people willing to spend (on goods and services) less than their incomes so that the people on the demand side can spend more than their incomes. Who are these savers, the ultimate lenders? Or rather, why are they willing to save more than they did in the past? Because otherwise, the borrowers can't borrow more and spend more?

But Nick, with fractional reserve banking couldn't banks have reduced reserve ratios to expand credit instead of relying on increased savings from other agents?

Moreover, to the extent that the additional loans needed lenders who already had cash, isn't the answer to the question "Who are these savers, the ultimate lenders?" quite obviously the Chinese (and other Asian savers)?

Hi Adam:
I took up fractional reserve banks in my latest post. They don't affect the accounting identity S=I. That's why I insist on some sort of general equilibrium perspective.

I was about to say "the Chinese?" at the end of the last comment. In a GE (i.e. closed world economy) they are clearly part of the picture. If people with a high propensity to save get richer, the total supply of savings goes up.

That might be part of the answer for the US, but I don't think it works for Canada, since we didn't have much in the way of a capital account surplus. Net foreign debt is approx zero.

Nick,
you are of course correct, but I could then point to increased profit share and increased inequality of income and ask which came first the chicken or the egg (since the increased share of profit is ENABLED by wage earners spending more than they earn).

I'm sort of wondering if we aren't drifting into some sort of vague Marxism here?

Profit, that is, the rate of return on capital investment, is part of the yield curve, or should be regarded as such. Since the spread between profits and interest bearing assets is due to pyschological factors like time-preference and risk aversion/seeking, it should be noted that since the monetary policy of the fed cannot change these values, it follows that it cannot change the spread, which, ceteris paribus, stays constant, regardless of fed induced changes in interest rates. That is, interest rates have no affect on investment demand whatsoever.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad