« This is what a balance-sheet recession looks like, and it's not pretty | Main | Thomas K Rymes, 1932 - 2011. »

Comments

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

"If age were the only thing that mattered, and if people were otherwise identical, and if everyone were age 35, debt would be zero. That's because there would be nobody for the 35 year-olds to borrow from. It takes two to tango. A borrower and a lender. Dollars borrowed equals dollars lent."

Doesn't this break down if banks change their leverage ratios. IE they start lending the 65 year old's savings out 30 times instead of 20 times?

Al: Short answer: No. Banks are (approximately) irrelevant. With (say) 10% capital, a bank borrows $90 debt (including savings account + chequing account) plus $10 equity from the 65 year old, and lends $100 debt to the 35 year old.

Nick, time inconsistency? Social norms "honour thy mother and father..." are one way of making the younger generation pay the older one back, government pension plans, backed by the power of the state, are another.

How does the global financial system prevent the young generation from doing an Eric Cartman and just walking away?

That's why the blather about "protecting the young" from the " pension burden" by going from "répartition" or pay-as-you-go to " capitalisation" or funded scheme is so wrong. And that the nonsense often comes from some who should know better is so dispiriting.

On the micro level, capitalisation is a legal and accounting trick. You put aside part of the wage bill in a segregated fund so that the employee have a garantee that the "funds" will be there even if the employer is no longer extant.
On the macro level , of course,( in a money not a real-coconut economy) there is no such thing as " putting away" savings to be "drawn down" later. You just send the unused purchasing power to the financial system in the hope it will find some borrower.And hoping the lessened demand will not crash AD. Basic Kn stuff.
What will happen in a demographically unbalanced world is that the amount of savings will be lower than the amount of desired borrowing. Interst rates will rises as ther will be more bond sellers than buyers. The same for equities. The young will face higher borrowing cost. So will their employers who will find themselves less able to invest. Scarce labor would see their wages rise but decreased investment will not allow higher productivity.
Mortgage rates will rise but housing price will go down as the old unload ther now-too-big houses. Good clean fun will be had by all...

How to "avoid the Cartman"? For you conspiracy fans ( and you know who you are),that's here that the real purpose of going from pay-as-you-go to fully-funded comes in. Taxes can be fought in the political arena. What can you do about "market forces" wreaking havoc to interest rates?

"How does the global financial system prevent the young generation from doing an Eric Cartman and just walking away?"

Ah, but dangle a marshmallow in front of Cartman and what's he gonna do? And those over 65 have an awfully big bag of marshmallows.

Patrick: "dangle a marshmallow in front of Cartman and what's he gonna do?"

Eat the marshmallow and then say "Scr** you guys I'm going home."

:)

"If age were the only thing that mattered, and if people were otherwise identical, and if everyone were age 35, debt would be zero. That's because there would be nobody for the 35 year-olds to borrow from. It takes two to tango. A borrower and a lender. Dollars borrowed equals dollars lent."

I don't know, I think if that was the case, every 35 year old would go to the bank and borrow money. And the bank would print the necessary funds and deposit it into their bank accounts. Of course, since they are all identical, they'd all end up maintaining an equal balance in the chequing account to offset their loans. The bank would siphon off its usual spread of a few percentage points but as long as the bank spent its profits back into the economy all would be stable - it would be like the scenario with no debt, only the owners of the bank would capture a portion of the economy' spending power.

But if the bank decided to increase its own retained earnings, then the economy would run out of money and you'd get a debt-deflation type scenario, which could be staved off temporarily by more borrowing or an increase in velocity, but these would be stopgap measures that would just end up increasing the size of the eventual deflationary pullback.

In seeking to compute a global index, you are assuming that the propensity to borrow of each cohort is identical in all societies.

Also in the specific sense that the propensity to borrow is of identical magnitude and opposite sign between the 35-yr cohort and the 65-yr cohort. This makes your result rather less remarkable...

Frances: actually, what I had in mind was voluntarily undertaken debt. The borrower goes to the lender and says "If you give me $100 I promise to pay you $105 next year". So the response to Cartman would be "Because you promised". There is always a time-consistency problem with promises, that's why the social institution of promising exists. And why laws enforcing contracts, and debtors' prisons (er, OK) exist. But yes, there is always the risk that Cartman will refuse to pay.

(We used to say "welch on his debts", but the Cymru anti-defamation league frowns on that expression, understandably, and I have no idea of its origins. I wonder if the Germans will coin a new expression "Greichen"?)

Jacques: I disagree. I think there is a real difference between a fully-funded and a PAYGO scheme. In the former, real resources get transferred away from current 60 year olds to current 30 year olds. The 60 year old builds a car that the 30 year old drives, and then 10 years later the latter, now 40, builds a second car that the former, now 70, gets to drive, to repay the debt. In a PAYGO scheme, we skip the first car.

Declan: Suppose that banks create money to lend to the young a la the traditional story, without the old wanting to save and hold those extra deposits. We are now in disequilibrium, and that extra money will start to hot potato, and will create inflation. It is then the inflation tax, and the transfer of real resources away from existing holders of now inflated money that finances the resources transferred to the young.

I'm assuming, correctly I think, that the Bank of Canada won't let that happen. If there's an increased desire to borrow and spend by the young, and no matching increased desire to lend and save by the old, that means the natural rate of interest has increased, and the Bank of Canada will increase (allow to increase) the actual rate of interest to match this. So the rate of interest (and possibly credit rationing) adjusts to equilibrate the demand to borrow with the supply of lending, without inflation or an excess supply of money.

My simple story in the post simply ignores the role of interest rates in matching the number of 35 and 65 year olds who get to dance. If I re-told the story, with interest rates playing their full role, it wouldn't change what happens (very much).

(As an aside, it was depressing reading the comments on Stephen's Globe and Mail piece. So many commenters said "they just borrow from the banks", without asking how the banks finance their lending. When the bank makes a loan the bank has an asset. And the bank in turn has a liability to the people who lent to the bank and who own the bank. So, ultimately, one group of people have an asset and another group of people have a liability, and the bank just washes out.)

Jon: "In seeking to compute a global index, you are assuming that the propensity to borrow of each cohort is identical in all societies."

Agreed. Or, another way of saying it, I am wanting to see how much of the increase in debt can be explained by demographics alone, without having to resort to other variables that might explain debt. My hypothesis is that age is far more important than anything else, and we can explain most of what's happening with age alone. I could be wrong, of course.

"Also in the specific sense that the propensity to borrow is of identical magnitude and opposite sign between the 35-yr cohort and the 65-yr cohort. This makes your result rather less remarkable..."

Well, in telling the story I did skip over the crucial role of interest rates in equilibrating the desires to borrow and lend. I talked about shifts in the demand and supply curves of loans, and posited a very crude credit-rationing mechanism Q=min{Qd,Qs}, rather than r adjusting the equilibrate Qd and Qs. But the two stories should have roughly the same effects, if both Qs and Qd increase over time.

I think Jacques is correct. Real assets are transferred from the 60 year old to the 30 year old in both cases, in the fully funded case within the retirement plan as the 30 year old buys them, in the paygo case outside of it as the 60 year old spends them. There is a difference only when first establishing it and not even then when paygo is phased in over time. In your case a car is produced that otherwise wouldn't be, but if the economy is operating at capacity, it can't be, so one has to ask why it is operating below potential and why this allows it to reach it. If savings equals investment, investment may be nominally higher, but not really higher as savings displaces consumption and interest rates and returns adjust to equalize them.

Lord: "There is a difference only when first establishing it and not even then when paygo is phased in over time."

That difference when first establishing it is, to my mind, is a big difference. A PAYGO phased in over time is still a PAYGO, just a smaller one initially.

"In your case a car is produced that otherwise wouldn't be, but if the economy is operating at capacity, it can't be, so one has to ask why it is operating below potential and why this allows it to reach it."

The economy can be operating at capacity. The car (or something) would have been produced anyway, but the 60 year old would have been driving it, rather than the 60 year old lending the 30 year old the money to buy it from the 60 year old.

"As an aside, it was depressing reading the comments on Stephen's Globe and Mail piece. So many commenters said "they just borrow from the banks", without asking how the banks finance their lending. When the bank makes a loan the bank has an asset. And the bank in turn has a liability to the people who lent to the bank and who own the bank."

Now you are making *me* depressed. A person borrows to buy a house. He sells his IOU to the bank -- that is the bank's asset. In exchange, the bank gives him its liability -- it's IOU, which is a deposit. The buyer then buys the house with the bank's liability. The homebuilder turns around and pays his suppliers and shareholders, etc. So other people in the economy end up with the bank's liabilities -- with deposits.

Ex-post -- these are the "lenders" that offset the borrower. But that accounting identity is not a behavioral relationship. The behavioral relationship is that the bank agreed that the borrower could repay the loan, the borrower was willing to take out the loan at the given interest rate, and as a result the (nominal) wealth of everyone else in the economy *increased* by some amount.

The reason why banks were necessary is that a priori the borrower's liability would not be accepted as a means of payment, whereas the bank's liability is accepted as a means of payment. That is why you need the bank to buy the borrower's liability in exchange for its own liability.

The right way to get rid of banks is to assume that *everyone*'s IOU is accepted as a means of payment. That would be the equivalent model in a monetary world without banks, as we are cutting out the middle step where the bank buys the borrower's IOU and issues its own IOU (a deposit).

The wrong way to get rid of banks is to assume that there is no money and the bank is lending out chickens, in which case the bank must first obtain chickens from a lender before it can lend the chickens out. That approach not only strips out banks, it also strips out money and leads to fundamentally different debt dynamics.

In the first approach, the wealth of the (ex-post) lenders is a function of how much the borrower borrows. If borrowers borrow more, than everyone else in the economy has more money to lend to them (ex-post) as a result of the borrowing occurring. As the real assets are unchanged as a result of the loan, this means that nominal wealth moves with a different law of motion from real wealth. As long as the borrower is willing to borrow at the interest rate posted by the banks, then the act of borrowing creates enough money to ensure that the ex-post identity holds.

In the second approach, only real wealth is being transacted, and you can talk of lenders having a level of wealth that is exogenous of how much people borrow. You can literally talk of lenders looking for borrowers and setting the interest rate in the process. But that is not our world.

We need to consider the breadth of a country's social safety net. A case in point: China's median age is 35. But we know that the country has been a (huge) LENDER over the past decade. Age by itself doesn't do the job here.

I agree that people of a certain age have a certain propensity to save, but I don't think global finance obeys such prima facie laws.

RSJ: OK. But a third approach is to assume that the Bank of Canada gets it right, so we can ignore monetary shocks because there aren't any. And a fourth approach would be to assume that this is long run analysis so we can ignore the effects of monetary shocks even if there are some.

Colin: Yep, age alone is not the only thing that affects individuals' borrowing or lending propensities. (If it were, then all people of the same age would have exactly the same debt/income or loan/income ratios, which they don't.) But how many of these other things cancel out in aggregate? Have these other things been changing over time, on average, in a big enough way to matter? Sure, China's high propensity to save is a big story. But my hunch is that global aging (which includes China's aging) is a bigger story. It's unprecedented.

What was it somebody said? Something like "Normally, countries get rich, and then later get old. China is getting old before it has got rich."

"RSJ: OK. But a third approach is to assume that the Bank of Canada gets it right, so we can ignore monetary shocks because there aren't any. "

But that gives you a different model, with different outcomes. In other words, it makes a difference. And whether or not it all washes out in "the long run" is a huge open question. You need some theorems to say that all these approaches lead to the same long term outcome -- I don't think these theorems exist.


In terms of China -- another example of how the details are important -- China is not a private ownership economy. The vast majority of savings are in the state and state owned enterprises. You cannot argue that households have these savings via their ownership claims on the banks and state, because there are no such claims.

If you just subtract consumption from GDP-TX and call that household savings, you get a 20% savings rate, which is 4 times the U.S. savings rate of 5%. But if you actually poll households and ask them how much they save, then you find that urban households in china save 5% of their disposable income (rural households save more, but have much lower incomes). The difference is the retained earnings of government enterprises and state owned enterprises, which are never delivered as capital and wage income to households. So here demographics are irrelevant. If the government were to deliver that income to households, then perhaps they would save it and perhaps they wouldn't. But as they don't have the option, it doesn't matter.

So, if there were a big bulge in the population at around age 35, and very few people around age 65, the average level of debt would be highest, right?

1) You should be clear that you are abstracting from the income/wealth distribution. If a few people hold (nearly) all the wealth (or all the wealth except assets saved for retirement, and the former is much larger than the latter) are lending it to everyone else, it is likely that their lending and borrowing behavior will be relatively age invariant. If everyone else's indebtedness peaks at age 35, then yes, a society with a demographic bulge will have higher average per-capita debt than others.

2) I suspect that distinguishing between the different definitions of "average" - mean, median and mode - is important, but I mislaid this train of thought while typing up point 1 above.

I see that RSJ has already pretty much said what I was going to regarding banks being 'a wash'. I hear what you're saying Nick, but it seems a stretch to think that the dynamics of debt creation and interest flows can be safely ignored or that they won't make much difference to the story.

Two things:

1) RSJ, claiming "the vast majority of savings are in the state and state-owned enterprises" is wrong. Households save two-fifths of the Chinese 50 percent saving rate, as do corporations, and the government saves the balance--ie, Chinese households matter, as the American ones do. These statistics were made in a BIS report by two Chinese economists that cites other Chinese authors, not in a poll. And, yes, I understand that corroborates my point about precautionary saving, which brings me back to it.

2) Nick, you ask: "Have these other things been changing over time, on average, in a big enough way to matter?" My answer is yes. China’s dependence dropped from 70 to 40 percent within a generation, and the working-age share of the population consequently jumped from 60 to 74 percent. There are more eastern savers than there use to be. And let's not forget about the other giant in region either.

The global economic centre of attention is being shifted to Asia not because of its aging, but because of its development and subsequent accumulation of wealth. The first decade in the new millenium was China's coming of age, not its increasing of age.

What Marcel said.

The distribution of wealth is such that, as rough approximation, only those in the top 10% of the wealth distribution hold anything more than housing wealth, employer pensions, and government pension entitlements. They might hold some other wealth, but it's generally trivial. See, for example, this 2002 article .

Wealth inequality has risen considerably over the past 20 years.

"Young people borrow, to finance school, house, car, and kids. Their debt reaches a maximum, somewhere around age 35."

I'll skip the kids part. If the young people's real earnings and real earnings growth are high enough (maybe even the parent's saved for college), why should they need to borrow at all? Why not just pay for it with savings and/or current wage income?

Colin,

My source is "Household Savings in China" by Art Kraay,

"Between 1978 and 1995, rural saving averaged 15.9 percent of household income, while urban saving rates were a much lower 5.2 percent." (p.11)

Note that it is common for rural households to save more -- due to the precariousness of farm income and lack of social benefits. Chinese rural households do not save more than Indian rural households, for example. And chinese urban households do not save more than U.S. urban households. This is by measures of individual behavior. At the aggregate level, the large savings of the Chinese are fully accounted for by actions of the state.

Both measures of savings, of course, are done by surveys, but it is the difference between looking at NIPA data and at SCF data. In the case of China -- where the vast majority of savings are forced by the state (e.g. are involuntary), it is important to look at SCF-type data if you want to assess the result of voluntary behavior.

Btw, when you say "corporations", what you really mean is state-owned enterprises. Households do not take delivery of these savings, as they belong to the state.

RSJ,

Point taken. But you agree that those data are old. Yes, I understand that there are capital controls and alternative names for corporations in China. But corporations, too, can save in precaution.

My point to you was that the Chinese household isn't to be labelled as a small fry. They matter, a lot. It wasn't until they started moving to the city and earning more that the non-household sector was able to save (by expropriation, as you say) so much. And this happened in a country with a median age of 35 years; ie, in a country of not-your-typical "maximum indebted" consumers.

It was an increase in eastern earning power, not in age, that created the savings glut and lowered interest rates that reduced the westerner's marginal debt cost. Nick actually touched on this point in his first post: "A rich country, with twice the per capita income across the whole age distribution as an otherwise identical poor country, would be like a country with twice the population." I'm just flipping the argument and (loosely) saying that a country with twice the population is like a country with twice the income. One-point-three-billion people is a lot, and their combined savings are nothing to snub.

Nick:
In fully-funded, my" contributions" and the from my employers ( that is not tken-home pay) are loned to a 30 year-old to buy a car. It might also be loent to a 75- years old to buy an electric walker, In a paygo my taxes go to the 75 to buy his walker. The rest of my personnal savings (the same amount as before since I pay in taxes what I don't need to contribute) go to the 30-year-old. There might be a slight difference at the macro level on the precise uses of resources but on the macro side, there can't be any difference.

This from my post on August 26 2010 on the thread "Why "everyone" should be forced to take Intro Economics" ( about the Kotcherlatoka incident incident ) a.k.a. the Christ Oh Christ ! thread

"Third, and so related to the first it is almost the same thing, is why in a money economy there is no difference between a capitalized ans a pay-as-you-go retirement system at the macro level. (At the micro level it is of course a different story) Nobody reads "Capital accumulation with or without the social contrivance of money" anymore?" You didn't disagree at the time...

Correction
"There might be a slight difference at the macro level on the precise uses of resources."
Of course I meant micro...
As for the other fat fingers, the celebration of the Fête Nationale has alredy begun at the office...

"Al: Short answer: No. Banks are (approximately) irrelevant. With (say) 10% capital, a bank borrows $90 debt (including savings account + chequing account) plus $10 equity from the 65 year old, and lends $100 debt to the 35 year old."

Here is what I think I have picked up about accounting from JKH, Billy Blog, and others.

Same scenario.

Asset side:

$100 NEW loan
$10 Capital requirement

Liability side:

$100 NEW demand deposit
$10 Equity (Net Worth)

The new loan gets a capital requirement. The new demand deposit gets a reserve requirement. I skipped the reserve requirement. The demand deposit is new, and if it becomes a means of payment, is new medium of exchange.

The key is the demand deposit is new (new/more medium of exchange, capital multiplier), and whoever holds the liability (demand deposit) is a "lender" ex-post.

Frances: "The distribution of wealth is such that, as rough approximation, only those in the top 10% of the wealth distribution hold anything more than housing wealth, employer pensions, and government pension entitlements."

OK, but isn't that top 10% disproportionately old? And, those savings for government and employer pensions are exactly the sort of savings for retirement I'm talking about. It shouldn't matter for my model whether the individuals save for their retirement, or whether their employers and governments do it for them. It's still savings, and has to be lent.

TMF: if you think people should save to buy a house, and then buy it without a mortgage, who do they lend their savings to? Are you saving for your retirement? Who did you lend your savings to? You bad person -- you just created debt!

Colin

-- Yes, the data is old -- it takes time to read these papers, and that was the one I read a while ago, back when I was more interested in these things. It may well be that now (e.g. 10 years later) household savings have increased, which would suggest a downward sloping function of consumption with respect to wealth. It's hard for me to believe...

I agree that Chinese households are important. But households in other nations are also important. But why does China, as a nation, have consumption equal to 1/3 of GDP whereas other nations have consumption equal to 2/3 of GDP.

Is it something special about the Chinese character --something not shared by Japan, India, Korea, Vietnam, Thailand, or even the previous generations of Chinese? Or is it government policy, forcing a certain amount of savings to occur?

Chinese households, given the chance, would prefer to consume more than 1/3 of national income. But they aren't given the chance because they never take delivery of that income.

"If age were the only thing that mattered, and if people were otherwise identical, and if everyone were age 35, debt would be zero. That's because there would be nobody for the 35 year-olds to borrow from."

Aren't you describing a depressed economy? That is, the 35 year olds are able and willing to produce the goods and services for each other that they all want, but they cannot afford? (Why they can't afford them is an interesting question, but if they could afford them, they would not need to borrow, right?) In that case, couldn't, wouldn't, shouldn't banks or the gov't create the needed cash?

IOW, if the rich uncle does not exist, let's create him?

Min: if anything, I'm describing the opposite to a depressed economy. The 30 year olds want to spend more than their income. If the central bank just printed the cash, and lent it to them, there would be excess demand for output. We need some other people, the 60 year olds, who want to spend less than their income, so that desired total spending equals "full employment" income.

(BTW, I have talked about 30 and 60 year olds, not 35 and 65 year olds, because by assumption 35 is the age at which desired debt is maximised, before which they want to spend more than their income and increase debt, and after which they want to spend less than their income and decrease debt, so at 35 we hit the inflexion point, where desired spending equals income. And 65 is the other inflexion point, at which negative debt is maximised.)

Jacques: ignoring capital and investment, then you are right that at the macro level we would see no difference in consumption between a fully funded and a PAYGO pension plan. But at the micro level we would, as you say. The first generation that starts the PAYGO plan consumes more.

The 80 year old is consuming both assets and transfers. If transfers increase, the 30 year old has less to save and buy the 60 year old's assets, so what the 60 year old gains on transfers he loses on assets. In the absence of transfers, assets would higher but returns on them would be lower. There would be significant distributional effects, I grant you, as assets are much more highly skewed than transfers, but these are primarlly intra rather than inter generational.

Nick Rowe: "if anything, I'm describing the opposite to a depressed economy. The 30 year olds want to spend more than their income."

Yes, they want to spend more than their income. But what are they going to buy with the money they borrow, and who is going to produce it? They are going to produce what they consume. (Ain't nobody else to do so, by assumption.) Since they are under producing and under consuming, why isn't that a depressed economy?

It seems to me that what you are describing is like the situation in the British colonies in America in the 17th century, a society starved of money, in that case because of the British view that the colonies existed for the benefit of Britannia. The colonists made up for their lack with barter, store credit, commodity money, wampum, and Spanish dollars, etc. Later, creating their own money brought them prosperity, but it also brought them into conflict with the Crown because of seignorage.

"If age were the only thing that mattered, and if people were otherwise identical, and if everyone were age 35, debt would be zero. That's because there would be nobody for the 35 year-olds to borrow from. It takes two to tango. A borrower and a lender. Dollars borrowed equals dollars lent."

Uh, but what about government borrowing? Corporate borrowing? And then what about foreign investors lending money or buying fixed income assets? (Or fromt he perspective of people in china and 3rd world contries, using savings to lend across borders)

RSJ:
There is nothing special in the Chinese character. Income is rising faster than the customary standard of consumption. They can't believe their luck and the perceived permanent income is still way below the real one. That's why you need to use that short window of opportunity to invest as much as you can. Once that stage of growth is over, it won't come back. Yes the chinese gunmint is doing his best to channel GDP into investment but it would happen anyway ( they are the "mouche du coche" the fly who thinks it is pulling the cart)
Britain never could never start up investment after the 1900's when the middle-class got to consume most of its income. And no government policy could change that.

Min: let's create a rich uncle. That's the Hamiltonian argument about getting rich through the creation of national debt.


Nick: yes the nominal holders of wealth are old. But the heirs are young and know the moolah is theirs. Being the hereditary son of billionaire is almost ass good as being one.

"Being the hereditary son of billionaire is almost ass good as being one"

That may be true on an individual level. On the macro level though, I wonder ... Something tells me that 1 billionaire and 99 serfs is not the same as, say, a normal distribution of wealth. Don't ask me to prove it though.

Inherited wealth is just redistributive socialism within the dynasty. Richer generations transfer some of their consumption to poorer generations ;)

TMF: if you think people should save to buy a house, and then buy it without a mortgage, who do they lend their savings to? Are you saving for your retirement? Who did you lend your savings to? You bad person -- you just created debt!

That probably depends on the definition of debt. I'm going to use it has to be currency denominated, has an interest rate attached, has repayment terms attached, and brings something from the future to the present.

So if I save $1,000, take the currency to a brokerage firm, and buy a stock, then there is no debt involved. The other thing is it should be possible to earn interest on my savings with no one else going into debt/having to make interest payments.

Patrick: yes 1 bil and 99 serfs is not the same as equality in about every conceivable way. I was just talking about certainty of your income. And being a serf's son is as bad as being one ( if only in that being the son a serf means you are one yourself...)

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad