Afternoon Commentary

ARNOLD: By "dropping the representative agent who works at the GDP factory" do you mean continuing to ignore financial accounting, balance sheets, and Federal Reserve Bank operations?

Also, given that the private sector is short on aggregate demand, and that the household is the ultimate driver of aggregate demand (and overleveraged), why do you restrict a payroll tax holiday to employers?

Wouldn't it make more sense to recapitalize households (or at the very least, de-capitalize them at a slower rate)?

Businesses add capacity when they see profitable market opportunities, just as banks extend credit when they see profitable loan opportunities.

Your approach of "giving money to business, so they can hire people" seems to miss the point almost as much as "give reserves to banks so they can make loans". But at least you are recapitalizing part of the right sector, which is miles ahead of Sumner and company. I would love to understand the logic you used to select the part of the sector you did.

That said this new position of yours is much better than "go recalculate yourself!"

Do you really think securitization is optional? How will capital get where it's needed? In all seriousness, do you expect endowments, pensions, government agencies, foundations, and individuals to just put their cash into illiquid CDs and subsidize the banks? Oh, I can't wait to get my 1% return!

Most people want the banks to increase their capital. That means *less* lending, not more. So how would small businesses, commercial developers, and mortgage borrowers get the cash?

I usually expect otherworldly non-seriousness from Krugman, but it's just silly to assume that securitization can just go away. It is a crucial means by which capital can be deployed in a liquid manner to sectors that are willing to pay incrementally more.

Unless you think we should all buy Treasuries and let the Federal govt allocate. Because they do such a better job than the market.

"cognitive capture"

Great expression.

It sounds like Solow is attacking his own work and his own influence on the profession:

"currently fashionable macroeconomics likes to formulate things in a way that inevitably endows the economy with more coherence and purpose than we have any right to assume. I certainly hope this is obvious enough to the younger people in the profession, the graduate students and even junior faculty. I expect there will be a revival of doing macroeconomics that does not push that kind of coherence on aggregate economic behavior.'

"let's drop the notion of the representative agent who works at the GDP factory"

Didn't Solow's work on "capital theory" do as much as anything to _encourage_ this picture?

The one capital goods model IS the "GDP" factory ...

Doug T: I think securitization is optional. Securitization for mortgages didn't start until the 70s. Did the US lack for affordable housing in the 60s? 50s? 40s?

Banks should be focused on credit risk.

winterspeak: We cannot afford to go back to 50's and 60's productivity. The cost of bank-led financial intermediation is huge. Quite frankly, I need more than 3% from my savings.

1960s institutions were fine for a 1960s level economy. But since the world is more complex, with more varied players, we need ways to intermediate Chinese capital into Brazilian manufacturers to provide airlines for Israeli executives at American software giants.

Put another way, you can't put the genie back into the bottle. "How you gonna keep 'em down on the farm, once they've seen Gay Payree?"

"let's drop the notion of the representative agent who works at the GDP factory"

Doesn't Solow's one capital good model = the GDP factory?

Doug T: I think we can go back to 50s and 60s productivity in housing starts. There were about 4x as many housing starts in the 60s as there were in the 90s.

It's good to look at how finance actually worked in the past. Google did not need securitization, neither did the iPod.

Bank-based financial intermediation requires that banks borrow short (your checking and savings accounts) and lend long (30 year home mortgaes and 10 year commercial loans). This works fine so long as interest rates are stable. Fixed exchange rates, substantial product market rigidities, and a currency convertible to gold are necessary to maintain these stable interest rates. We don't have those things any more.

With volatile interest rates, no bank can possibly lend out the money from your savings account for thirty years. So to finance long-term loans in a world without securitization, the banks would have to issue bonds. Those bonds would have to backed by either (i) the operating business of the banks, which would make the bonds very risky or (ii) defined pools of bank assets, which would mean that we had re-invented securitization, only with less transparency and efficiency.