Scott Sumner  

The Legacy of Milton Friedman

Reminder: My Debate at Oberlin... Would (Our) Open Borders Lead ...

Last month I attended a conference on Milton Friedman, in Austin, Texas. The final session discussed his legacy, and I thought I'd share a few of my remarks.

I argued that Friedman and Schwartz's Monetary History of the United States was Friedman's most influential contribution. Recall that the Great Depression had convinced most intellectuals that laissez-faire capitalism had two major drawbacks:

1. It was unfair (something they believed even before the Depression.)
2. Capitalism didn't even work very well; it was inefficient.

This change in the zeitgeist during the 1930s had caused governments all over the world to become far more statist. Note that there wasn't anything particularly "left wing" about this policy shift, right wing governments also became far more statist.

Friedman and Schwartz's book began to change this perception. But not right away. The world wasn't quite ready for their message in 1963, when the price of gold was still pegged at $35 an ounce. After the dollar was floated in April 1968, inflation and NGDP growth soared. And the rate at which inflation soared seemed to depend on how fast various countries printed money. Those that printed money at a faster pace than the US (such as Italy and Britain) had higher inflation than the US. Those that printed money at a slower rate (such as Germany and Switzerland) had lower inflation that the US. It looked like you could choose any long run inflation rate you wished, merely by adjusting the rate of money growth.

The 1970s saw a gradual acceptance of the view that money growth determines inflation and NGDP growth. And that meant that the Great Depression could have been prevented if the US had printed money at a faster rate. There was a minor dispute about how much could have been done under a gold standard, but that debate had no relevance for the fiat money world of the 1970s. The younger economists coming along at that time would have laughed at anyone suggesting that a fiat money central bank would not be able to create inflation.

Once this was accepted, the Great Depression went from being viewed as a failure of capitalism, to what it actually was, a failure of demand management, i.e. bad monetary policy.

That meant capitalism wasn't a dysfunctional system after all, just an unfair one.

At this point the left decided the best solution was free markets plus social insurance. By the late 1970s even Ted Kennedy was calling for deregulation, and for tax reforms that cut top income tax rates and simplified the tax code. By the 1980s left wing governments all over the world began deregulating and privatizing. Growth did not speed up in absolute terms (the golden age of growth ended in 1973, for unrelated reasons) but at least those who liberalized rapidly did better than those who moved slowly.

So the Monetary History helped usher in neoliberalism, which was a synthesis of classical liberalism and socialist 1930s liberalism---free markets plus social insurance.

The book also ushered in New Keynesianism. No longer was fiscal policy viewed as the appropriate stabilization tool. Later work by Friedman convinced economists that low rates didn't mean easy money (until they forget in 2008.) He convinced them that there is only a temporary trade-off between inflation and unemployment. New Keynesianism might just as well be called new monetarism. Brad DeLong did an excellent article on this.

New Keynesianism was a synthesis of the classical quantity theory of money and old Keynesianism. Thus the Monetary History contributed to two major sea changes in economics; neoliberalism and new Keynesianism.

Some argue that Friedman's free market ideas were utopian, out of touch with political reality. But you can never tell which policies that are viewed as crazy today, will still be viewed that way in the future. Who would have expected gay marriage and legalized pot in a number of states, even 30 years ago?

At the end of Capitalism and Freedom (1962) Friedman lists 14 government policies he didn't like. I was struck by the fact that substantial progress has been made in many of these policies:

1. AFAIK, the Texas Oil Commission no longer controls oil production.

2. Price controls are not very widespread, and Massachusetts voted to abolish all rent controls a few years back.

3. Legal restrictions on bank interest rates (loans and deposits) have mostly been removed.

4. The Interstate Commerce Commission has been abolished.

5. The FCC still regulates TV and radio, but entry is now far easier (perhaps technology gets credit here.)

6. The sort of public housing projects built in the early 1960s are no longer being built, and the voucher approach has gained ground. Of course things are still far from perfect with housing policy.

7. The military draft has been abolished.

8. Some toll roads have been privatized.

9. Tariffs are not gone, but are much lower than in 1951.

Of course there are lots of other free market policy reforms that are not called for on his 1962 list. Pot smoking wasn't a big issue in 1962, but when it became one he advocated legalization, which is now well underway. I'd guess he favored legalized gambling, another area where great progress has been made. He later advocated school vouchers, and there has been some progress on that front (much more in countries such as Sweden.)

His record is not perfect, I think he underestimated how much moral hazard was created by FDIC, and hence the risks of banking "deregulation." I use the scare quotes as banks were not deregulated, rather they were regulated in a way that encouraged more subprime loans, but banks did become freer to engage in risky loans to developers, and that's caused a lot of bank failures.

Friedman was my favorite economist, and the one who had the greatest impact on my beliefs. He left the University of Chicago just a few months before I arrived. :(

Comments and Sharing

CATEGORIES: Monetary Policy , Regulation

COMMENTS (22 to date)
Hana writes:

Friedman also was an inspiration for me. I was fortunate to be at UC during his time there.

One small point. I think you are wrong about the April 1968 date. My recollection was it took place under Nixon, not Johnson. Nixon first stopped convertibility of gold, and some time later the floating dollar began.

Hazel Meade writes:

IMO, what the 2008 crisis revealed is that capitalism *isn't* an unfair system, it's centralized banking that is. The fact that the banks got bailed out , and beyond that theat they are stabilized (and hence profits guarenteed) by the federal government, is the injustice.

I hate to sound like a Ron Paul fanatic, but minus the ability of large banks to get freshly printed dollars essentially for free from the government, where is the unfairness in the system?

I'm not convinced that central banking is even necessary to regulate demand or stabilize the money supply. The development of BitCoin and other cryptocurrencies shows that you can have currency without a central bank, and just set up a formula that determines the growth rate. And you can do that without selecting a group of favored people and institutions that get to receive the bounty of the freshly created currency. Anyone can be a BitCoin miner.

We're at a point with the internet where it's possible to see a vision of a capitalist economy that doesn't need a central currency, or even need property rights or courts. All you need is a website with user reviews to magnify reputation effects and increase imformation flow. The Silk Road survived for two years (before it was shut down by the feds) trading explicitly illegal products anonymously with no need for any of those things, or even a protection racket.

Don Geddis writes:

@Hazel Meade: "I'm not convinced that central banking is even necessary to regulate demand"

But then your examples are only about stabilizing supply (e.g. BitCoin). That part is easy. The gold standard stabilized money supply -- but it didn't prevent the Great Depression.

The real goal is to stabilize the value of the Medium of Account. (That's the part of money that matters for macroeconomics; not the Medium of Exchange part that is used in transactions.) And value comes from the intersection of supply and demand, not from supply alone.

BitCoin's value fluctuates tremendously. If it were used as an MOA, society would experience recessions and unsustainable booms on an almost monthly basis.

(P.S. Banks don't get dollars from the Fed for free.)

Scott Sumner writes:

Hana, I believe that private gold holders could no longer redeem dollars for gold after April of 1968, but perhaps my memory is off. Nixon closed the gold window for foreign central banks in 1971, but it's the free market price of gold that matters.

Hazel, The value of Bitcoin is very unstable, which makes it a very poor medium of account.

JP Koning writes:

In April '68 the authorities stopped selling gold in the free London market. Prices immediately rose to $40 or so. A two tiered market emerged in which the public could buy in London at $40 while central banks were required to buy at $35.

Hazel Meade writes:

Well, this is why I add "and other cryptocurrencies".

I agree that (at least right now) BitCoin's value is too unstable to be used as a currency. But that may simply be due to instability caused by it's introduction. If everyone switched back to using gold suddenly, I'm sure the value of gold would fluctuate wildly while that happened. But that wouldn't mean it would never settle down to a stable value. Eventually BitCoin's value may stabilize too.

Also, even though BitCoin's formula means there will eventually be a fixed number of BitCoin, that's not true of all cryptocurrencies. Theoretically, you could make the mining rate formula anything you wanted. You could tie it to the volume of exchange, so that the more people used it, the faster it would get mined.

However, my concern is mainly that with the central banking system we have created a system that is inherently unfair, since the people who have access to the freshly printed currencies are a limited and politically favored group. Me, or my next door neighbor, do not get to borrow money from the Fed at 0% interest. Goldman Sachs does. We don't get freshly printed dollars handed to us out of thin air. The point of BitCoin is that it removes that political favoritism from the equation. The generation of new currency is spread throughout society to anyone who chooses to run a mining program. It is not funneled to wealthy elites who lend it out at interest ot the rest of us.

I wish we had better measures of statism. I suppose Scott Sumner is correct when he says governments all over the world became far more statist in the 1930s, and that later in the 1980s left wing governments all over the world began deregulating and privatizing (meaning less statism I suppose). But that language is pretty vague.

I do not imply there is anything wrong with vague language; that's the best we can do with still-vague concepts. But I yearn for a day when clearer measures support our arguments for deregulation.

Benjamin Cole writes:

Nice post---and a rare reminder of the bad old days in regs and taxes...Don't forget stockbrocker commissions! is it with a unionized workforce, 90% MTR, limited trade, bad regs (ICC) the USA economy boomed, boomed, boomed in the 1960s?

If monetary policy is the answer...

Scott Sumner writes:

Thanks JP.

Hazel, You said;

"Me, or my next door neighbor, do not get to borrow money from the Fed at 0% interest. Goldman Sachs does."

I'm pretty sure this is inaccurate. Can anyone confirm? However I do agree that the Fed should not be in the business of lending money to anyone.

Richard, It's tough because there are so many types of statism. When I was young it was perfectly normal for steel companies and automakers to be owned by governments. Now that's considered kind of weird.

But on the other hand new types of regulations have sprung up. So it's sort of a "comparing apples to oranges" problem.

Ben, Yes, but I'd say that was a golden age for other reasons as well.

Andrew_FL writes:

If Milton Friedman's legacy on the intellectual front truly was the emergence of that horrid chimera of ostensibly free markets retaining legality of plunder by the ruling authority, then that's disappointing. If his intellectual legacy is forcing the Keynesians to become New Keynesians, that is even more disappointing.

There may be a great deal to celebrate in terms of policy progress, but clearly on the intellectual side there is a much longer way still to go.

James writes:

Don Geddis:

In order for the Fed's open market desk to get anyone to take the other side of their transactions, they have to buy instruments for a price higher than the current price or else no one would be willing to take the other side. If, on average, those instruments are efficiently priced before the Fed starts buying, then the extra increment paid by the Fed is a transfer.

Scott Sumner:

You write "The 1970s saw a gradual acceptance of the view that money growth determines inflation and NGDP growth. And that meant that the Great Depression could have been prevented if the US had printed money at a faster rate."

You second sentence does not follow from the first. Even granting the standard story about money, inflation and NGDP, the circumstances leading to the GD might have been so tied to real variables (as in the Ohanian and Cole paper on the GD) that no monetary or fiscal policy would have actually prevented the decline in real output.

I sometimes get the impression that you think all recessions can be prevented if the Fed would just print money fast enough. Are you quietly assuming that there can never be real shocks so severe that recessions are unavoidable? If so, it would be interesting to know why you think that.

Don Geddis writes:

@James: The Treasury market is about as liquid a market as exists in this economy. The "extra increment" is minimal. Worse (for your model), the actual price of Treasuries is dependent upon future NGDP/inflation, and the Fed's OMP actions can change these future expectations. So your intuitions about how the price would change, if it had been any other buyer, may not apply when the Fed is the buyer.

"all recessions can be prevented"

Of course, real output can decline, in the industries directly affected by the real shock. (E.g. the tsunami in Japan.) But a "recession" is economy-wide, where industries not directly affected by the real shock, see rising unemployment. Those are caused by nominal shocks, not real shocks. (And they can be prevented by NGDPLT.)

Enial Cattesi writes:


There was time when it was considered that faster money printing can prevent hyperinflation. Really ...

For the most part the GD was caused by FED's inflationary policy and then the continuous intervention of the Hoover and FDR administrations, but if you ask other people, they will say that there were excess savings. Really ...

David Witzel writes:

Scott, do you have an opinion about Friedman's position on the role of business & social responsibility? (e.g., his 1970 NYTimes article: )

I blame him for engendering a mindset that has over-simplified the role of business (i.e. that business exists to maximize shareholder value) which has resulted in huge damage to the environment and society.

Fair or not?


dwb writes:

Friedman was an enormous inspiration and frankly I think his loss is deeply felt in the GOP. His policies were what I would call pragmatic libertarianism - social problems exist but we need to use markets to address them - rather than the kooky stuff that's dominated since. When the progressives propose more command and control economics like increasing the minimum wage or adjusting overtime rules, or environmental regs, the GOP appears lost. The support for raising the EITC is weak at best. Half of them appear to deny income inequality or poverty is a problem, the other half appear lost as to how to use incentives to address issues. Even worse, some are aggregate demand shortfall deniers. Even now, on pot, we have republicans walking around saying don't raise tobacco taxes because doing so promotes black market smuggling between states, while those same individuals fail to recognize the silliness of pot prohibition. On monetary policy of course is where his influence is missed the most.

Jeff Frankel wrote a piece a month or so ago about how the Dems had returned to command and control style problem solving and republicans had returned to denying social problems exist. Could be a coincidence, but I think the loss of Friedman's influence on the right had a lot to do with that. The real legacy in my mind is that he was an analytical pragmatist who recognized that social problems (and nominal rigidities) exist. Not a kooky denier like we see on the right today.

Roger McKinney writes:
My recollection was it took place under Nixon, not Johnson.

Actually, FDR made it illegal for US citizens to own any but small amounts of gold. It was still illegal until Reagan repealed it in the mid-80s. No citizens in the world in all of history have been prevented from owning gold except US citizens. What kind of freedom is that?

The 1970s saw a gradual acceptance of the view that money growth determines inflation and NGDP growth.

That's sad, because Hayek and Mises had been telling them that for half a century. But as Mises wrote, the only mistake worse than ignoring the quantity theory of money is assuming it working mechanically. It doesn't. Ceteris Parabus it does, but Ceteris Parabus never holds.

For example, a great deal of the new money created by the Fed has gone overseas to buy imports and to invest in emerging markets. That keeps it from cause domestic inflation or goosing domestic gdp.

JP Koning writes:

It was Ford, not Reagan, who repealed the laws preventing private ownership of gold by Americans.

James writes:

Don Geddis:

On March 18, you wrote "Banks don't get dollars from the Fed for free." On March 19, you wrote "The extra increment is minimal," rather than zero.

Also, liquidity only addresses bid ask spreads. The extra increment that the Fed pays for paper also comes from the upward sloping supply curve for that paper.

Scott Sumner writes:

James, The huge fall in NGDP between 1929 and 1933 was mostly a demand shock. Cole and Ohanian are of course right that there were lots of really bad supply side policies, but most of those were adopted as a result of the fall in AD. If AD doesn't collapse, we don't get the horrible supply side policies (with the exception of Smoot-Hawley, but that played a modest role in the depression.)

David, I thought Friedman made some good points, but overall his argument had weak philosophical underpinnings. Do businesses have a moral obligation to lobby for tariff protection, if it is in the interest of their shareholders? I say no, and I imagine Friedman would as well. But that means business does have social responsibilities beyond maximizing profits. Perhaps obligations in areas like the environment, an idea which Friedman mocked.

dwb, Good points.

Roger, Your last point confuses money with other financial assets. We tend to buy imports with stocks and bonds and other assets, not money. Some money goes overseas and is hoarded in the form of $100 bills, but most of the Fed's recent injections stay in the US as excess reserves.

Roger McKinney writes:
We tend to buy imports with stocks and bonds and other assets, not money.

That's typical of mainstream econ's lack of time. When Walmart bought shoes from China, I doubt Walmart paid with stocks and bonds. I'll eat my hat if they didn't use cash to buy yuan. Of course, the Chinese government then turns around and uses the cash to buy US bonds and gold.

As the Institute for International Finance has demonstrated, $US trillions go overseas to be invested. I doubt many investors in emerging market ETF's or bonds are buying them with US bonds and stocks. All ETF's and bond funds require cash. Yes, much of that returns home as foreigners by US assets.

Your analysis would apply to domestic purchases, too. I pay cash to my hometown Walmart, but any cash they hold will be in the form of stocks or bonds. So you could say I buy Walmart goods with stocks or bonds.

B.R. Kelley writes:

Years before his death, Friedman did an interview with Brian Lamb of CSPAN. At the time, if memory serves, he put his works in different categories.

"Free to Choose" the best-selling.

"Capitalism and Freedom" the most influential.

"Theory of Permanent Consumption" the most original and groundbreaking, in addition to Nobel worthy.

"A Monetary History of the United States" most widely-accepted by academics.

James writes:

Scott: Do you believe there could be circumstances where real shocks cause macroeconomic problems that NGDPLT could not solve? Has this ever happened before?

Comments for this entry have been closed
Return to top