Monday, September 19, 2011

Krugman.   Me.


The big rise in prices during and after WWII arguably did a lot to eliminate the debt overhang, making it possible for the economy to enter a sustained, non-inflationary boom.


Print money and use it to pay off debt. Paying off debt destroys debt. It also destroys the money used to pay the debt. So if we print money and use it to pay off debt, the debt is destroyed, and the money is destroyed, and we do not cause inflation.

UPDATE: The debt that *must* be paid off is private-sector debt.


Sackerson said...

But if you pay off debt, someone else has just had that payment, and the money eventually goes somewhere else. The newly-printed money is not sterilised.

The Arthurian said...

Hi, Sackerson... They have to be careful to pay off debt that was created by fractional reserve banking, I think, not debt at random.

But it seems to me that the newly printed dollar would go back into someone's savings account, filling a "hole" where the dollar was, before it was lent out.

The problem we have today -- the threat of deflation -- arises because of deleverage: People are paying off debt, which removes money from circulation, which is why there is the threat of deflation.

If we print a dollar and use it to pay off debt, the newly printed dollar goes out of circulation (and thus threatens deflation; but no, because it was newly printed!)

Where the newly printed dollar goes, is into somebody's savings. It does not *ADD* to the fellow's savings. It only frees up credit that was in use, making the credit again available for lending.

The threat of inflation, ironically, arises not from the newly printed money, but from the already existing surfeit of savings.

Sackerson, I acknowledge that this is a crazy idea. And I definitely need help thinking about it. Perhaps the above notes will convince you give it a second thought?

Q: What happens to banks if the debts get paid off? I think they get hungry to make more loans.

Q: What happens to debtors if the debts get paid off? I think they get willing to borrow, again.

Calgacus said...

Sackerson, I acknowledge that this is a crazy idea. It's not a crazy idea at all. In fact, it was basically how "the debt overhang" was eliminated by the New Deal & the War. What was striking about the war was how little inflation there was, considering the size of the war and the expansion. Your method was much more important than Krugman's. Sometimes he seems to want to become a target of the Austrian web-loons.

People paid their debts the old-fashioned way. With money from their jobs, many of which came directly from the government. Which is the real old-fashioned way, before this capitalism/private enterprise intermediation. And with lots of money going around, and memories of the Depression seared into psyches, reckless lending was not resumed.

Debt repayment, along with taxation, is the other major method by which credit/money is "refluxed", and by which the value of money is maintained. In times like these "it" will not "eventually go somewhere else", the newly printed money tends to be "sterilized" by being used to repay debt. Post-Depression, both borrowers and lenders are more careful.

It is sobering to contemplate just how far backwards things went in the decades since the 40s. Until the 70s, sometimes theory &/or practice advanced, but it was usually accompanied by at least equal retrogression, especially in theory, until we had the last several decades of a perfect storm of utter, intelligence-insulting nonsense.

Sackerson said...

I meant, paying off debt with freshly-printed money.

Paying it with existing money ends the debt without increasing the total currency in the system; so does writing off debt as unrecoverable.

If we're into thinking laterally, I'd like to see say 50% of all mortgage debts written off; that would instantly double the deposit-to-loan ratio (ignoring other types of loan). But then that would tank house prices and doubtless there'd be further consequences.

And as you say, let people off the hook and they'll do it again. Unless you introduce strict rules e.g. on loan-to-income. Not to mention that I don't trust rule makers any more, either.

The Arthurian said...

Sackerson said...
I meant, paying off debt with freshly-printed money.

Yeah, so did I.

jim said...

Hi Art,

A question you might ask is where did the $1.6 trillion come from that the banks produced to purchase the treasuries that were then sold to the Fed.

To sackerson:

The newly printed money does not go "somewhere else" The printed money that buys treasuries that were purchased with bank excess reserves is sterilized in that those reserves have no place to go in a world where nobody is willing to borrow.


The Arthurian said...


I didn't say it in the post, I see, but the debt I want to pay off is private-sector debt. I think it is private sector debt that hinders private sector growth. And I think the primary objective is to restore growth.

But I don't know: Where did the $1.6 trillion come from?

jim said...

I'm not sure why or how you want to pay off private sector debt. What specifically does that entail? Are you just going to give money to people with debts? And what are the people without debt going to think of that?

The problem with the private sector debt is that the overhang of debt is causing high unemployment due to the low demand created by diversion of money into saving and repayment of loans.

In normal times, as Sackerson said the money saved or used to pay debt would be recycled back into the economy (as loans). But what happens when (as we have today) no one is borrowing?

That is how the banks ended up with $1.6 trillion in excess reserves.

I mean there is all this M2 money that is money supposedly in deposit accounts, but it is not really in those accounts. It has been loaned out (more than once). The accounts are actually close to empty and you would like the loans (something like $42 trillion) to be repaid so that the deposits are back where they belong (something like $8 trillion in deposits).

Now how are you going to accomplish that without unwinding the income derived from those debts (which will mean many more unemployed)? Or if you pay debt with printed money without unjustly rewarding those who imprudently went deep into debt on a gamble and punishing those who didn't behave badly?

The Arthurian said...

Hi, jim. This is what I'm looking for... thoughtful challenges. Thanks.

"Are you just going to give money to people with debts?"

No. Most people would most likely use the money to pay down debts (as happened before with stimulus plans). But that is not good enough. My objective is to *assure* that debt is reduced. I guess people could go to something like an unemployment office to do the paperwork... Or, to their lender!! That works for me.

"And what are the people without debt going to think of that?"

If it costs nothing to create the money used to pay down debt, and if the money is destroyed by paying off debt (so it is not inflationary) then why would anyone care? And if this "anyone" is a creditor, well, his savings are less at risk because debt is being paid back. He should be happy.

"The problem with the private sector debt is that the overhang of debt is causing high unemployment due to the low demand created by diversion of money into saving and repayment of loans."

Yes, this is exactly what I have been looking at.

"That is how the banks ended up with $1.6 trillion in excess reserves."

Nice, Thank you, jim, for laying this out for me.

$42/$8 is five-plus. In a normal economy, people need to get paid so they can make their payments. Same thing here, why not? So print one dollar and [hopefully] it is used to pay off 5 or 6 dollars of debt.... But, I suppose that wouldn't work?

Dunno about unwinding the income. But "unjustness" of policy does not concern me here. Stopping the decline concerns me. Preventing a Dark Age is the primary issue.

I have to think about this a lot more. Thank you, sir.

jim said...

Hi Art,

One thing that is a given is that paying down debt levels will result in large deposit holdings by lenders. If the govt borrows the money it distributes then lenders have some place to put all the money they are being repaid. If the govt prints money it won't be issuing treasuries so the lenders won't recycle the money back to treasuries. That seems potentially dangerous.

There was a time after WW2 when people could remember that assets can lose value. Borrowing was done as an investment to preserve income or produce new income. You borrowed to buy a house because in the long run it was cheaper than renting. You didn't borrow because you expected to get rich off the increasing value of the house. That meant borrowing produced growth from increased income and not from inflation of assets prices. The inflation of asset prices is the result of widepread speculative borrowing that wouldn't occur naturally with productive borrowing.

Printing money to pay down debt is likely to just restore the unproductive borrowing resulting in inflation of asset type growth.

The way I see it is private debt is currently being slowly paid down. Total debt is no longer growing. I don't see a problem with stimulus based on reduced taxes that allow people to pay back debt or save or spend as they choose. All three of those are needed to get back to a point where credit expansion becomes productive again.
But I don't think it is wise to remove all pain from the process.

Borrowers have once again learned that assets can lose value. It is the knowledge that assets may still lose more value that is inhibiting borrowing the most. You don't want that lesson to be lost and encourage the return to speculative borrowing.

The Arthurian said...

Hi, jim. Thoughtful stuff.

I can't defend this, but my impression is that investors turned to speculation as productive-sector profits declined, making speculation more profitable. By the 1980s people had forgotten "that assets can lose value" and policies were put in place that should never have been.

Now, as you say, "Borrowers have once again learned that assets can lose value." But I think private debt is currently being paid down much too slowly. The longer it takes, the lower the "new normal" becomes. I think this is dangerous for the survival of the nation and the civilization.

Make it as painful as you like; before long we are all dead. The next generations will not remember the pain. When their profits fall, they will elect a Reagan and he will say a little speculation might be a good thing.

jim said...

This graph shows how debt and GDP have have grown relative to each other

[increase in GDP]/[increase in debt]

Prior to the meltdown there seem to be a steady trend of declining productivity increase per dollar borrowed.
That may now have improved considerably, although it is still too early to tell what the long term trend will be.

I see that as a lot more positive step forward rather than reflating some old bubble to get out of this.

The Arthurian said...

Nice graph!

Pretty regular for a long time. Then the big surprise. Hopefully, yes: It would be great if declining productivity increase per dollar borrowed showed some marked improvement.

For the record, I do not see my goal as "reflating some old bubble to get out of this"... :)

I think the first priority is to stop the decline. I think the second priority is to reduce our reliance on credit. I hope to see the financial sector whither away.

I'm thinking about using your graph and most of your comment as a new post, if that's okay with you.

I do like that graph. It shows a thing more clearly than Grandfather Hodges or Ron Robins or The Economist, for sure.

jim said...

Hi Art

I'm looking forward to see what you do with that graph.

I didn't mean to imply that your goal was to reflate bubbles. I suspect the print money method would have that unintended effect.

I think the ideal solution is the debt gets paid off the old fashioned way of earning money to pay it off. That means the only real priority should be jobs.

The Arthurian said...

Hi jim.
If you're like me, you might have remembered to check the checkbox so that you receive follow-up comments (like this one) in your email. That would be good.

I am looking at your deltaGDP-per-debtaTCMDO graph and starting to get my thoughts together to do a post or two about it.

In my evaluation of the graph, a question arises: Since the graph looks at growth of GDP, why doesn't the graph use real GDP?

I thought about it and came up with some reasons why showing "nominal" GDP might be better in this situation, but I think my reasons are not very strong. Or (if strong) not well-stated.

Do you have any thoughts on this? Can you give me solid reasons for sticking with nominal GDP? Other than "It makes an interesting graph," I mean.

Warning... I may quote you in my post!!! :)

Thanks jim.