December 14, 2007

Break With Milton Friedman

It makes me nervous to part company with Milton Friedman. Only Hayek would give him a run as the man whose views most closely match my own. But I think I am ready to sign up for "New Monetarism." We've had a bit of monetarist persiflage around here. My contentions that productivity and trade are counter-inflationary have repeatedly been met with Friedman's assertion that "inflation is purely a monetary phenomenon."

I have to publicly break with that. The money supply is not being created by central banks and the deflationary effects of expanded trade, technology and productivity cannot be ignored. David Roche has written a book titled "New Monetarism" and he provides a brief summary on the Wall Street Journal ed page today.

The reason for the exponential growth in credit, but not in broad money, was simply that banks didn't keep their loans on their books any more -- and only loans on bank balance sheets get counted as money. Now, as soon as banks made a loan, they "securitized" it and moved it off their balance sheet.

There were two ways of doing this. One was to sell the securitized loan as a bond. The other was "synthetic" securitization: for example, using derivatives to get rid of the default risk (with credit default swaps) and lock in the interest rate due on the loan (with interest-rate swaps). Both forms of securitization meant that the lending bank was free to make new loans without using up any of its lending capacity once its existing loans had been "securitized."

So, to redefine liquidity under what I call New Monetarism, one must add, to the traditional definition of broad money, all the credit being created and moved off banks' balance sheets and onto the balance sheets of nonbank financial intermediaries. This new form of liquidity changed the very nature of the credit beast. What now determined credit growth was risk appetite: the readiness of companies and individuals to run their businesses with higher levels of debt.


Roche is not at all sanguine about this. This is a bubble-creatin' machine to him which will exacerbate the current credit crunch and contribute to a catastrophic burst of a China Bubble in 2008.

I cannot join his pessimism but I agree that we can no longer measure inflation by looking at the rate of growth for the money supply. Roche seems to only see the inflationary side in his editorial (I just ordered the book). To admit that this liquidity has been created off balance sheets and outside of reserve requirements is to assume an unignorable inflation rate. Even the worst inflation hawks do not claim any such thing.

I think you admit this ex nihilo money creation, and then admit that it is counterbalanced by the incredible efficiencies of trade and technology. It puts the Fed and Central Banks in the correct perspective: they influence but do not define monetary policy.

Economics and Markets Posted by jk at December 14, 2007 12:08 PM

Friedman is still right. Very bluntly now: I don't know who this guy is, but I can already tell that the book is a steaming pile of bull.

I wonder where Roche received his education. Even basic high school calculus (like when I learned it) teaches what he apparently needs a whole book to explain. Whether it's banks' paper ledgers, electronic ledgers or "securitization" of assets, there's no mystery about it at all: banks create money by lending. It's the simple principle of fractional reserve banking: you deposit some money, and the bank can lend out a significant fraction. Perhaps you'll deposit $100, and they'll feel safe to lend out $90. Under that ratio, $1, whether you use integral calculus or merely reason where the series converges, is potentially $10 to the money supply if at every stage it's spent and saved at 90%/10%. Even if no more than 50 cents of every $1 is lent out at every stage, that's still an increase of $2 to the money supply.

Nothing new about this, folks. Nothing more to see, please move on.

"deflationary effects of expanded trade, technology and productivity cannot be ignored."

As I've explained before, this is not deflationary. It exerts downward pressure on prices, but that is not deflation. Rising prices are not inflation. Again, you confuse inflation with price swings based on supply and demand. Oftentimes both are at work, but they are not the same.

Ultimately it is *fundamentally* about how much money the central bank creates. When I have $1, who made it? And what if they create $1 more? Look at the source, the entity responsible for the initial creation of money. You're looking at the wake, not the power of the propeller.

When the central bank creates 5% more dollars in a year, it doesn't do a damn bit of good if imports increase my purchasing power by 10%. And as I tried to get you to understand before, imagine how much better off we'd be if the government left us to be free to use sound money *and* if we could take advantage of cheaper prices.

Posted by: Perry Eidelbus at December 14, 2007 4:07 PM

Also,

"To admit that this liquidity has been created off balance sheets and outside of reserve requirements is to assume an unignorable inflation rate. Even the worst inflation hawks do not claim any such thing."

We don't "claim any such thing" because it's irrelevant. Let's say the money supply is $10, and fractional reserve banking turns it into an effective $100. That is *not* inflation. If the bank makes it $10, $50 or $100, that's based on supply and demand (interest rates and consumers' preference for borrowing). But no matter what they decide, it becomes $20, $100 or $200 if the central bank decides to double the money supply.

Posted by: Perry Eidelbus at December 14, 2007 4:28 PM

It's exactly fractional reserve banking. But Roche is pointing out that as soon as the loan is securitized it comes off the books and the bank is empowered to lend it out again, creating money that does not come from a central bank. This is not your grandpa's reserve banking.

Yes, Perry you've told me several times. I respectfully suggest that you are too closely wedded to a textbook theory that does not match what we are seeing in the global economy. The Everyday Economist linked to a paper called "The continuing muddles of monetary theory" by Charles Goodhart who believes that "monetary theory has a long way to go." I believe that to be true.

Posted by: jk at December 14, 2007 4:30 PM

It's still fractional reserve banking. It doesn't matter *how* the accounting is done, only that the bank is still lending out something that was deposited with it. And again, it doesn't matter what banks do with money *after* its creation. They only amplify the money supply that the central bank creates in the first place.

My strict usage of "inflation" and "deflation" are in fact *contrary* to what nearly all economics textbooks say. Most economics texts that discuss inflation to any degree are Keynesian in basis. I, however, go by the true definition, i.e. Milton Friedman pointing out that inflation is purely a monetary phenomenon. Look, it doesn't matter even if Kudlow says that trade and productivity are "deflationary": if that's how he terms it, he's wrong. He's said silly things before about liquidity, and Mises.org has called him on it.

By your definition, the jewelry industry, or at least one of its members, had some pretty big deflation this last weekend. Odds are pretty good that my fiancee won't read this, so I can mention something I have for her. Guys are worried about if they're "big enough," and also if jewelry is big enough. Not that my fiancee wants anything beyond my love, but I became dissatisfied with a pendant that I bought the previous weekend. So, I returned it this Saturday and exchanged it for one with exactly twice the carats. The first one cost $x, and the bigger one cost exactly $2x. This past weekend, though, the first was reduced to $.75X, and the bigger was reduced to $1.5x (which I didn't even know until they rang me up in the exchange and quoted me half of what I was expecting, what a happy surprise). Did Kay's suddenly experience tremendous deflation? Of course not. So I caution you again, don't confuse inflation with supply and demand.

As I wrote before, baskets of goods and services are ridiculous measures of money's value. You need to judge inflation by fungible things, like gold, crude oil, and copper. They won't be exact measures, because there is still supply and demand at work, but as I pointed out with crude oil prices, the price jumps are far more than what supply and demand can explain. The same holds true for copper and other commodities with industrial uses: China's and India's industrialization cannot explain the surge in dollar-denominated prices. Look at gold: the increase in its value cannot possibly be because it's suddenly so much more in demand, all else being equal. Gold's value in other currencies disproves that. It's because dollars used to purchase gold have lesser and lesser value.

As far as Goodheart, I took a brief look at his paper and had to laugh. Like Roche, and like Greenspan (who I blasted last night on my blog), Goodheart is saying stuff as if it were a new revelation, when it's been obvious for decades! Case in point, his footnote at the bottom of page 3:

"According to pure monetary theory, if the
quantity of money in circulation is increased and other things remain the same, the value of
money must fall.... But its applicability to a given situation depends upon a correct understanding of what things are to be regarded as money..."

Austrians have known that for decades, because we (I include myself even though I'm only 95% Austrian) know that money is not limited to any single thing. Money is merely a medium of exchange, a tool, and fundamentally it's only information. Remember what I said before, that if a Scrooge McDuck decided to hoard almost all the gold, then the rest of us would simply find something else to use as money. A tool cannot be too easy to obtain if it's to have any value in acquiring (it can still have value in momentary utility), but it's useful only if it's available in sufficient quantities.

Economists have a bad tendency to make theories of money and banking, and other theories, far more complicated than they need to be. It's like Christians arguing about trivial things like how to cross themselves (or whether to) or if Christ really is mysticaly present in the Eucharist. Economists reinvent the wheel not because they feel "they're right," but because presenting something rehashed is key to sustaining their livelihoods: they get to sit around "theorizing," instead of doing actual productive things like analyzing. I gave a friend an explanation of Japan's deflation in a few paragraphs that was far more clear than Bernanke's convoluted non-explanation explanation a few years back. The difference is that I'm not a professional economist, and my time spent explaining is limited and therefore valuable.

Posted by: Perry Eidelbus at December 17, 2007 3:39 PM

I appreciate your views on a more absolutist monetary theory (FWIW, Ich bin auch ein österreicher) but I am troubled that you not see securitization as a huge shift in reserve banking. I can lend $10 on a dollar, fine. But if I securitize it, I can lend another $10. This is not a series with a limit, like .999999... this is complete absolution from reserve requirements 9 + 9 + 9 + 9 + 9... -- nothing to see here?

Posted by: jk at December 17, 2007 6:55 PM

It still makes no difference. Banks will be constrained by reserve requirements, not only by legislation, but to meet the demands of their depositors. Could a bank securitize 100% of its deposits? Certainly, if it wanted to risk failure (or in these days of moral hazards, being bailed out).

Banks might securitize deposits at a lower reserve ration than regular commercial lending, but either way, banks cannot increase the money supply infinitely. If they could, they would have already devised such ways hundreds of years ago. But no matter what ledger legerdemain they employ, the series still converges. They're merely amplifying the money supply that already exists. It's still the central bank, and the central bank alone, that controls the money supply.

Anyway, securitizing a deposit is no different, from either an accounting perspective or an investment one, than the regular commercial lending of money. What do you do when you purchase a security, but lend someone money?

Are you familiar with how to determine the total potential money supply from the initial monetary base and the average reserve requirement? If you do, consider this: when the Fed pumps in another $10, $20 or $50 billion, how much is it *really* creating?

Posted by: Perry Eidelbus at December 18, 2007 2:40 PM

On a related note, I didn't realize till this morning that the ECB is "injecting liquidity" to the tune of 348 *billion* euros. The Fed at least required an auction for a finite amount. The ECB simply said, "Here's 4.21%, borrow all you want."

And Europeans call Americans stupid?

Posted by: Perry Eidelbus at December 19, 2007 11:25 AM | What do you think? [7]