Robert Hall on Recession, Stagnation, and Monetary Policy
Jan 9 2017

recession.jpg Economist Robert Hall of Stanford University talks with EconTalk host Russ Roberts about the current state of the U.S. economy and what we know and don't know about the recovery from the Great Recession. Much of the conversation focuses on the choices facing the Federal Reserve and the policy instruments the Fed has available. The conversation includes a discussion of Hall's experience as chair of the National Bureau of Economic Research Committee on Business Cycle Dating.

Robert Higgs on the Great Depression
Robert Higgs, of the Independent Institute, talks with EconTalk host Russ Roberts about the Great Depression, the New Deal, and the effect of World War II on the American economy. Using survey results, financial data, and the pattern of investment...
Barry Ritholtz on Bailouts, the Fed, and the Crisis
Barry Ritholtz, author of Bailout Nation: How Greed and Easy Money Corrupted Wall Street and Shook the World Economy, talks with EconTalk host Russ Roberts about the history of bailouts in recent times, beginning with Lockheed and Chrysler in the...
Explore audio transcript, further reading that will help you delve deeper into this week’s episode, and vigorous conversations in the form of our comments section below.


Josh Sher
Jan 9 2017 at 8:31am

The podcast is not working this morning.

[Please reload the browser page or your iTunes feed. I’ve fixed the problem, thanks to a commenter who doesn’t want to be named who pointed it out privately. The mp3 file accidentally still was listed as 2016 rather than 2017 in some of our records. The correct mp3 file is available at
My apologies!–Econlib Ed.]

Jan 9 2017 at 10:33am

I was as puzzled as Russ about the reserves staying the same.

I took econ 101 in the 50s and recall the quantity of money was determined by the required reserve amount, which produced a leak in the lend-deposit-lend-deposit cycle, so that the quantity of money way the sum of a geometric series with the given leak rate.

Back then the required reserve ratio was the control on the quantity of money.

Since then the Fed went to open market operations, deciding on a slight change in a target interest rate at each meeting (based on leading indicators of inflation) and automatically selling or buying govt short term debt, and hence respectively decreasing or increasing the money supply, to keep it trading at the target. The old reserve requirement was left in place but didn’t change. It was no longer the tool. (note that you want your leading indicators of inflation to be orthogonal to short term interest rates, lest the Fed blind itself.)

Since the mortgage meltdown, they recapitalized the banks by buying bad stuff from them and sequestering the added money in reserves, kept there by paying interest. How they were going to unwind this position was always a question.

Hall is talking about this new situation as the permanent one, where the reserve requirement is long forgotten. They still have the interest rate target, but they also have the interest rate on reserves, to play with, to keep money there.

Whether this gives two dimensions or only one in fact remains to be seen. I think it will produce more of a conductor vs his crazed orchestra effect as the one affects the other in new ways not yet explored.

My idea was always just raise the reserve requirement and pay no interest. It still counts as capital on the banks’ balance sheet, which was the intended effect, and you’re back to the nice one-dimensional open market operations control.

Jan 9 2017 at 10:58am

Finally, an economist who understands that the quantity of reserves is fixed by the Fed and can’t (in the aggregate) be “lent out”, and that banks aren’t just “sitting” on these reserves, for some reason not lending them!

I’m a central banker who has worked with payment systems, and it has frustrated me endlessly to hear good economists complaining about banks not lending out their reserves.

The reason is pretty simple and purely down to accounting. Suppose that there are only two banks, A and B, in the economy, both with reserves (i.e. account-balances) of 100 with the Fed. Now, suppose that a firm is granted a loan of 10 from bank A. At first, this is just an electronic record at bank A. It gets a loan, an asset, of 10, and deposits the firm’s account, a liability to the bank. No reserves are involved. The firm, of course, hasn’t borrowed the money to keep them idle, but to spend them on something, say a machine. If it purchases the machine from another company, which also happens to be a client of bank A, nothing happens – no reserves. However, if it transfers the money to a company whose bank is bank B, bank A will have to make a transfer through the Fed’s payment system, and afterwards bank A has reserves of 90 while bank B has reserves of 110. The total amount of reserves, however, can’t be changed by the banks. Only the Fed can do that by creating new money.

Here’s a more detailed explanation by two Fed economists:

Todd Kreider
Jan 9 2017 at 11:51am


Or building houses is a great example. It’s a very large budget share; … And, houses are being built today with exactly the same technology virtually as they were 100 years ago.

Not exactly with the same technology:

The portable electric drill was developed and patented by Black & Decker in 1917. The first Black & Decker factory opened the same year.

Jan 9 2017 at 6:44pm

Does Hall report on recession after it occurs? Or predicts recessions?

I remember Bernanke prior to the fall 08 down turn saying the housing market was fine. I personally don’t think these guys saw it coming.

Why are the banks getting paid 25 basis points over the current rate of 1/2% by the fed. As hall said its a subsidy.
As a taxpayer I find that unacceptable.

The expansion of the deficit to nearly 20trillion….seems hardly likely that will be decreasing by the fed waving a magic wand.
And the amt of debt the U.S. Has built up results in a Neg mult effect on GDP. Plz read…
Lacy hunt was at the Dallas Fed at one point, now runs hoisington. Russ, he would be a great guest. He has to answer to clients with skin in the game!

Russ Roberts
Jan 9 2017 at 7:02pm


The NBER committee that Hall chairs reports on recessions after they begin. Their task is to date the beginning and the end of each downturn and upturn of the business cycle.

Jan 9 2017 at 8:24pm

[Comment removed. Please consult our comment policies and check your email for explanation.–Econlib Ed.]

Jan 9 2017 at 10:48pm

Mr. Roberts,
Good podcast I always learn something new. This podcast was a little over my head but enjoyable nonetheless.

Is there any plan down the road to do a podcast on the national debt and how to address? The USA is plagued with many problems but in my opinion none bigger than this.


Thomas Schlosser
Jan 9 2017 at 11:01pm

I listen to your show via Pocketcasts and there is an error that prevents me from hearing your Jan 9th show;contact me and I can email you the screenshot of the error.

Jan 10 2017 at 2:24am

I did not understand why the excess reserves could not go down.

Later in the interview the guest appeared to agree that eliminating interest on reserves would be expansionary because the Fed is following the market rate so that it would prevent the banks chasing down borrowers for higher returns. OK so what happens if tomorrow the Fed says the interest on excess reserves is going down from 50 bpp to -150 bpp (essentially creating Russ’ example but shifted down). Wouldn’t that lead to a massive burst of growth in the economy?

Stéphane Couvreur
Jan 10 2017 at 4:12am

I was confused by the discussion about reserves. I understand how the amount of reserves is under the Fed’s control, in the sense that if it didn’t bother about interest rates and asset prices it could always achieve any target by buying and selling stuff.

But R. Hall’s analogy with “borrowing” does not work for me. Sure, reserves are on the “liability” side of the Fed. But, in fact, this is an illusion due to the resemblance with private accounting in which the libilities side represents contractual obligations, debts, promises of all kinds. My understanding is that there are no such promises in the Fed’s balance sheet: no obligation to redeem or pay anything on the reserves, for example.

At one point, R. Hall mentions that “they have this automatic right to borrow from the banking system by creating more reserves”. In short, since the Fed decides the quantity, the maturity and the rate on reserves, how can this be analyzed economically as “borrowing”?

Jan 10 2017 at 11:12am

Another great podcast. Thanks. It’s too bad it ended when it did, because it seemed like Roberts and Hall were just getting into the meat of it in the last five minutes.

It is interesting that there is still so much disagreement on the cause of the downturn, what the response accomplished, and how it could have been better. Maybe that’s because it’s ultimately impossible to rationalize a system where an entity like the Fed has near unlimited ability to create money (which ought to be a relatively stable proxy for work and physical stuff) from nothing….and does it to the tune of trillions of $.

The discussion about photography etc. and what we can all do with a few clicks on a home computer, was also interesting. Maybe we have reached a level of technology and automation where, at least in the developed world, significant inflation is a now thing of the past, even when trillions are injected.

As an aside, I wonder when work itself will become a thing of the past in the developed world, and how happy that will make us…Star Trek? Blade Runner? Morlocks and Eloi? Eh, I’ll thankfully be long gone.

Jan 10 2017 at 4:36pm

Share some of the confusion from the discussion about the link between the interest paid on excess reserves and the interest rate on government debt.

Prof. Hall referenced a speech by Chair Bernanke on this topic in the immediate post-crisis period. Would appreicate it if you could identify the speech and post a link.

Thanks much.

K Banaian
Jan 10 2017 at 5:58pm

I was stuck in an airport van for 2.5 hours last night, and this helped get me through an hour of it. I rewound it a couple of times.

The point about reserves is that total reserves are indeed completely under the Fed’s control. Excess reserves are not. When excess reserves are converted to loans to start a deposit expansion, little by little those reserves go from being excess to being required. Eventually the system runs its course.

I was interested in his point on state government spending. I served in a state legislature in 2011-12, and I distinctly remember us having to demonstrate “maintenance of effort” to receive that money. That meant that money our state received had to go to K-12 or higher education to not decrease from pre-recession levels. This of course was an invitation to many states to simply divert funds elsewhere. The reason states were given money was that, due to constitutional provisions against borrowing for current expenditures, spending in these areas would have fallen dramatically had ARRA not backfilled federal dollars into the states. For Prof. Hall to say ARRA didn’t raise state and local government spending avoids the fact that it would have fallen without it.

Jan 10 2017 at 6:54pm

“They borrowed the funds from the commercial banks and used the proceeds to buy bonds. Because a lot of people don’t understand that the process that you just described is simply borrowing in the capital market. The Fed greatly expanded the national debt by issuing what are Federal obligations, borrowing–which are reserves.”

1- maybe it is technically true that this increased natl debt but when the funds were used to buy commercial bonds, it doesn’t change the sum total of credits and debits.

2- NOOOO!!!! Hall says if a bank lends/invests more funds another bank must hold more funds. There is some truth in this but that totally ignores what he says later about if interest rates go up, Banks will be lending as much as they can instead of parking it with the fed, and money will be moving like a hot potato. Said differently, it ignores the difference in M1, M2, etc. money supply. It also ignores the inflationary effect of banks lending more instead of parking with the fed.

Jan 11 2017 at 12:02am

The point about reserves is that total reserves are indeed completely under the Fed’s control. Excess reserves are not. When excess reserves are converted to loans to start a deposit expansion, little by little those reserves go from being excess to being required. Eventually the system runs its course.

How does this work? I thought we had the following:

Total Reserves = Excess Reserves + Required Reserves

JK Brown
Jan 11 2017 at 11:18am

Robert Hall: Yeah. Exactly. So that’s a very concrete example. But I think it’s fair to say, when you look at what the economy produces and what we consume, there’s just an awfully large number of things where we know nothing much has changed. Like the generation of electricity. Or building houses is a great example.

I’m afraid Mr. Hall is out of touch with that statement. There have been very large improvements in both how we build houses and electrical generation/motors in the last century. Note this depiction of the number of workers on site in ‘Mr. Blanding builds his Dream House’. Also note the absence of nail guns and while not clear, sheet goods such as plywood. Today we have factory, and thus possibly built by robots in the future, -made trusses, structural insulated panels, even concrete wall sections.

I remembered a post by Timothy Taylor on nails from 2012. The price of nails declined until the 1950s and has even risen since then. However, many attributes of nails have changed for the better and it is argued that the cost of an installed nail has declined with the advent of nail guns. The home construction job site is now a far less populated place than even 30 years ago.

As for power generation/motors, the motor/generator has been mature technology for a very long time in the windings area, but control was difficult and expensive. The separate motor controller has undergone a dramatic change of late with improvements and cost reductions in power electronics and computer switching control. But this happens even as the “ancient” motor remains. Just the guts of the controller box are switched out.

As a consumer example of the change, have you noted the move from the large, heavy bricks that used to be on your laptop, electronic device power cord to tiny, lightweight plugs. Many of which have standardized on the USB standard creating a secondary market for chargers and integration of the capability into everyday objects?

Jan 11 2017 at 12:09pm

When the guest mentioned that having increased the reserves from 18 billion to 3000 billion and it was no problem and nothing could ever go wrong, that primal buzzer went off in my mind that someday, when something happens and we find out that something really bad could go wrong, quotes like this will be laughed at. But, primal impulses are often wrong.

I did not quite follow the ending discussion about the government multiplier. I don’t see why a negative multiplier for a giant group taking a bunch of money out of the economy and then trying to spend it might not be negative. I think Dr. Roberts even identified how it could be – the fed tried to boost the economy and the states paid down debt. Despite the claim they were “doing it wrong” it certainly appears that multiplier could be negative.

Jan 11 2017 at 12:44pm

I think that you and Robert Hall were perhaps talking past each other when you discussed the multiplier.

Taken in a very literal sense, if the government spends, it is impossible for the multiplier to be less than 1 in the short run. A dollar spent is a dollar spent- even if the dollar comes from debt.

You seemed to be questioning where that dollar came from.

In the short run, the answer is debt. In the long run, the answer is taxpayers servicing debt.

Only in the long run can your measure the “marginal” multiplier which could of course be negative. (ie money could be directed better than servicing debt).

I think a useful analogy for this is how a city may spend money on a stadium for a sports team. In the short term, stadiums always seem like a great deal. In the long run, not so much. The multiplier generated by the incomes generated by the stadium doesn’t offset the increased debt loads borne by all other taxpayers.

Jan 11 2017 at 12:55pm

The NBER reports on business cycle after it begins. With all due respect isn’t that like saying it’s raining after the rain starts? Other than a historical record I don’t see the importance of their work. Who pays the NBER? Hope it’s not the taxpayer.

And seeing the confusion above as to how reserves work, I wonder if Mr. Hall even has a full understanding. Being that this is his area of expertise he should have had a clear message on reserves and the impact.

Being a huge fan of this program, I can only ask if a knowledgable guest could clarify the reserve, money supply and velocity, and multiplier effects in the near future.

Russ, I do appreciate your push backs if you don’t think something is correct!

john penfold
Jan 11 2017 at 4:49pm

I don’t understand and will listen again and read comments again in a few days. But what happened to the old money multiplier? Yes the reserves are determined by the Fed but isn’t total money supply determined by the credit expansion new reserves allow? The Fed either directly or indirectly monetized the debt. The debt was acquired by spending more than receipts. So new Treasuries absorb the spending and interest rates should go up but reserves and money remain the same. Now the Fed buys the new treasuries which had absorbed the new spending, and the increase finally takes place. Of course if banks again buy Treasuries absorbing continued Federal spending it’s just a gradual shift of resources from the private sector to the public sector. What did our teachers have wrong? What on earth do reserve requirements mean if banks cannot lend based on new reserve creation? And are we saying shifting these resources to political control is meaningless?. That it has no affect on investment expectations, entrepreneurial activity, productivity? Is this a Keynesian view of the world, or just assertion that neo classical, austrian school and monetary economists have always been wrong.

Jan 11 2017 at 9:04pm

Interesting discussion. The discussion about the reserves seemed to ignore two issues. What if the monies go out to banks in other countries. And, if they are always balanced, why die the overall quantity need to be raised. Also, the interest payments made to the banks for the reserves (and on to the bank executives) comes from somewhere – taxes out of my pocket – why?

Ajit Kirpekar
Jan 12 2017 at 1:26pm

@ Hannah

I’m glad you made this point because I think Hall was answering the question in the literal sense that spending money will have an immediate impact.

The question really should have been – “Ok, but is that a way to get us out of a recession?”

One of the things that bothers me is that the pain of stimulus spending is usually felt way after the fact or not at all if the debt is slowly paid back from future productivity, so it keeps cropping up as an anti recession solution.

I agree – following some kind of shock, recessions are the result of a deep drop in demand for purchases. And in theory – if one could return spending to its normal levels, it would “cure the recession”. But this in theory stuff ignores a whole lot of issues.

For one, no one can even properly explain how its supposed to work in a practical sense. Money is spent through the government and that apparently keeps purchasing power constant and thus employers don’t fire people because they don’t see a loss in revenue. Leave aside expectations and the fact that businesses are generally forward looking, just how do the checks get to the person to spend in the first place?

This issue never seems to be explained. Congress goes through several months of deliberation before the spending is dolled out and even here there are tons of lags and its deeply uncertain who gets what and how much. And its not even clear(as in this case) that money will even be spent. As Hall pointed out, a lot of that money was spent to pay off debt. Ok, but then what did the debt holders do with the cash? They mostly plopped back into the banks where it sat doing nothing.

Then again, why is it assumed that if they had instead bought tvs and clothes that somehow that would have solved the problem? Recessions aren’t some uniform hit to employment across all sectors. Real estate agents, construction workers, maybe accountants and engineers unemployed in Nevada and New Jersey won’t exactly be instantly hired in areas like Pennsylvania and Oregon if we suddenly spend more on Nike shoes or diamond rings. And whats true of Pennsylvania and Oregon is true for Tokyo and Shanghai. If you really believe in Stimulus, you ought to restrict it to goods that hire mostly Us Workers.

Another dubious assumption – demand for goods created by stimulus will go to areas that are most likely hit by recessions. Ie – if there’s a sudden recession and people stop buying durable goods, the stimulus money will be spent on durable goods. Is this true? Has it been researched? Is it time independent? After all, in the old days, when you were hit by a shock, you probably bought less sugar and meat and spent the money on heating and cheap grains.

Today – sugar is practically everywhere and our standard consumption cutbacks are probably not even consistent across individuals within the same income levels!

In conclusion, the whole thing relies on a top of assumptions and wishful thinking. I can imagine if someone tried a sales pitch this way they’d be laughed out of the room, yet the malady lingers and will continue to do so for reasons that are beyond me.

Jan 12 2017 at 5:18pm

I agree with many here that this was an especially dense podcast. For the first time ever, I had to print out the transcript and carefully read it (BTW, we get about 20 pages of content in these podcasts, which is amazing. It is also understandable as to why the transcript comes out in sections). Notes:

– Hall is correct to point out that monetary policy can only do so much, it cannot solve an out of control fiscal policy.

– Since now everyone agrees that there is no such thing as “shovel ready” (Obama quote on biggest first term surprise – “there’s no such thing as shovel-ready projects.” – something a lot of us knew already), infrastructure spending cannot quickly stimulate an economy, which severely limits government’s tools.

What goes unsaid is that the death of “shovel ready” is due to the YEARS of regulatory compliance and hoops and environmentalist lawsuits that have to happen before the first dirt is turned over. This is a correctable problem (IF you want to use stimulus).

– One of the areas that wasn’t clearly explained is that government spending is in the definition of GDP, so debt driven stimulus spending (there is no other kind, although Keynes stated that his policies would only work with surplus based stimulus, a fact ignored by modern Keynesians) must, as Hall states, increase GDP at a multiplier of 1. But that is just in the short term. The longer term effect is much less than 1, doing more harm than good to the long term economy. Keynesians never state that in order for an experiment to be complete, the starting conditions have to return, that is, the debt from the stimulus must be repaid. This has never happened (WWII doesn’t count).

– The headline unemployment rate has become a political football. The discretionary adjustments to the data (business birth/death, participation rates) have become so large that it has no value left. Watch it start moving up after the inauguration.

– The labor market did not follow normal patterns in the recovery. Russ was right in that we should have seen a much quicker recovery to pre-recession levels. The recovery was hampered by tax increases, ACA and hundreds of thousands of pages of new regs.

We are not in record uncertain times and technological advancements have not stopped. The explosion in regs also helps to explain a lot of the decrease in the level of productivity as well. Every new regulation has a cost.

– Agreed that disability benefits are being widely abused and reform is desperately needed.

– The discussion on monetary policy was the hardest to follow as I think it could have been made clearer if total vs excess terms for reserves were more precisely used. Hall agreed that if excess reserves were loosed upon the economy (via the potential $30T in fractional lending capacity it enabled that we discussed in earlier podcasts) that it would be extremely inflationary and that the Fed would move to prevent it.

But saying that there are Fed papers that describe how this would theoretically be done does not mean that these techniques are without risk, quite the opposite. Using reverse repo’s (RRP’s) in the overnight markets for non-banks to retire it has never been done, let alone for $3T, and could lead to more major disruptions. Paying market interest on excess reserves could get very expensive if one goes back to the ancient history of 2007 when Tbills were paying 5%. That’s a $150B/yr cost on top of any other federal spending, including $1T in interest payments alone if the entire debt was at 5%.

– Although Hall didn’t quite admit that QE1 in 2008 was purely a bank bailout (it was), it was refreshing to hear him admit that most of QE was a transparent example of a central bank buying its own government debt. Refreshing to hear, but not reassuring. Whenever a CB buys its own government’s debt, you are approaching (or in) the debt cycle end game.

Ajit Kirpekar
Jan 12 2017 at 5:32pm

One thing he said that made me think about was – reserves are the same as debt.

What I think that statement means is – without QE, banks would have just bought all of the US debt and that would be on the books instead of reserves being there and the debt being held by the Fed. Since the rate of interest is nothing, reserves or debt makes no difference.

The real question is, long term, are the consequences the same for both stories?

Here its worth thinking about. In our current situation, the Fed will need to reverse repo or either continue to pay higher interest as JW pointed out.

In the alternative case – the banks hold a lot of treasures and dump them on the open market in exchange for cash to make loans to. That drives up interest rates and the cost to roll over debt; unless the Fed is the buyer in that situation.

I guess this gets back to – when debts are intractable, there’s nothing monetary policy can do.

Jan 13 2017 at 7:27am

Ajit Kirpekar,

What I think that statement means is – without QE, banks would have just bought all of the US debt and that would be on the books instead of reserves being there and the debt being held by the Fed. Since the rate of interest is nothing, reserves or debt makes no difference.

That is not the case. If there were no QE, the banks would not have had enough liquidity to purchase anything, they may not have had enough solvency to even stay in business. (This is not necessarily a bad thing, they would have failed, their assets would have been sold to other or new banks, and life would eventually go on. There would have been a deeper recession, but the moral hazard would have been eliminated and the misguided concept of “Too Big To Fail” would be dead instead of codified into Dodd Frank. See Schumpeter.)

As discussed on the podcast, the crisis trigger – subprime loans – were “only” a few hundred billion. But since everyone’s cool new risk models had the same strategically flawed variable – housing price growth is always positive – they created trillions in derivatives that also failed.

And without QE, an additional $3T of US debt would have had to have been financed in the open market, possibly raising interest rates further into debt spiral territory. We can never know if this counterfactual would have happened, but the Fed was certainly concerned about it.

The headline explanation for the effort to keep interest rates low was always to stimulate the economy. After eight years of failure, it becomes clearer that the real reason was to keep US debt financing costs low. Over the past eight years, the US AVERAGED $1.2T of additional debt each year (this is not the same as the deficit, there are government “off budget” accounting tricks involved).

If anyone’s current risk models assume long term low inflation and low interest rates, they may want to rethink them.

Ajit Kirpekar
Jan 13 2017 at 1:32pm

@ JW

I agree in principle, but remember TARP was s treasury act. In absence of the FED, the Federal Government would have stepped in and bought even more of those troubled assets or found some other way to provide a liquidity subsidy to the banks. They would have done so by issuing bonds on the open market which I’m assuming a lot people(including the Chinese) would have bought along with banks with healthy reserves.

Assuming there was general flight to safe assets from everyone – this would have flooded into banks who would then have either held them as reserves or bought said treasuries.

Walter Antoniotti
Jan 13 2017 at 2:58pm

Question on the last two minutes as related to success of Fiscal Policy used to negate the Great Recession

Federal Government used debt it as part of their Fiscal Policy plan to stimulate the slow economy caused by the Great Recession.
Much of the money was given to states governments. States used much of the money to lower their existing debts rather than spending it to increase Aggregate Demand. Mr. Hall believes the fiscal policy was not a failure because the states did not spend the money. Does this mean that advice given by economists is not a failure if the “economic behavior“ that follows wasn’t as predicted by said economists? Is this related to the “Undoing Project” predicted by “The Black Swan?”

Jan 13 2017 at 10:32pm

As I understand it, there seems to be a lot of confusion caused by referring to credits at the Fed as “reserves” when “excess reserves” entirely unrelated to the classic reserve function have come to entirely dominate required reserves in quantity.

There is also the confusion of the Two Multipliers. One is the legacy Money Multiplier which used to quantify the increase in the money supply caused by the ability to loan repeatedly money freed by a reduction in the required reserve, something rendered irrelevant by the insignificance of reserve requirements relative to now-interest-bearing credits.

Then there a the entirely different GDP and real-GDP Multipliers where it is claimed that government spending increases (or in the case of real-GDP, maybe decreases) the relevant variable by some factor.

Jeff Enes
Jan 14 2017 at 1:23pm

Thank you Russ for the push-back at the end of the program. It is unclear how government spending necessarily increases GDP when during the Great Recession federal stimulus was provided to states and the states chose not to spend it.

Brian Mason
Jan 16 2017 at 3:02am

Hall made several very weak claims.

One of the more glaring was testing of the uncertainty hypothesis. Hall claimed essentially that looking at a one off event, namely shortly after 9/11, showed no uncertaintity effect of broad not well-known persistent policy changes backed by 10,000s of bureaucrats, namely stimulus and ACA. shockingly weak

Bob Anderson
Jan 17 2017 at 10:44am

I’m one of the listeners who has urged Russ to explore with guests that had predicted inflation following the stimuli packages why their predictions proved inaccurate. Dr. Hall isn’t quite in that category but Russ does raise the issue with him generally. I’m afraid any prediction is “accurate” if it is subject to after-the-fact “not enough time has elapsed” or “it would have been worse/better.” I’d love for future guests making predictions to be asked, “what result would cause you to rethink the opinions that led to your prediction?” As always, thanks for the best interviews on the internet.

Robert Swan
Jan 17 2017 at 6:10pm

I came away from this with as muddy a picture as ever about the power of monetary policy. Like commenter rhhardin, I can’t see the benefit of the extra “lever” provided by interest rate on reserves. I was picturing the old scheme as a tap running water into a basin with a fixed drain. The new scheme adds a control to the drain but doesn’t give you any more actual control.

I sensed Russ was angling for the guest’s age, but he missed a clue. Prof. Hall said that interest rates were rising for the first half of his career, and falling ever since (1981). That makes the second half of his career ~35 years, which would make his age at least 90. Sounded pretty well preserved too. No need to thank me.

Daniel Barkalow
Jan 17 2017 at 8:17pm

The problem I have with the idea of interest on reserves being used to control the pace of the economy is that it seems unstable. If the system is limited by the quantity of reserves, a small change to the control value (putting in a bit of excess reserves or taking out a bit) has a directly proportional effect on the amount of economic activity. If the system is limited by interest on reserves, however, a small change to the control value (lowering or raising the interest on reserves) pushes it past an unknown quantity of possible safe-enough investments, leading banks to use reserves to make all possible safe-enough loans that pay between the old rate and the new rate. If they gradually reduced the rate, they’d initially have no effect (since it’s so high currently), and at an unknown point, they’ve have a large unknown effect.

In the worst case, every safe-enough investment would be a loan that could afford to pay the same amount of interest. If the Fed is paying more than that, every bank wants all the excess reserves and no loans are made to main street. If the Fed is paying less than that, every bank wants no excess reserves and every possible loan is made to main street. If the Fed crosses that rate, banks suddenly make $12 trillion in loans (or enough that their total capital requirements are $3 trillion). If it goes back, all those new businesses fail when they can’t keep their credit. There wouldn’t be any rate the Fed could set that would get banks to make an intermediate total amount of loans.

It’s a lousy control system, and I think that the Fed agrees with that assessment, because they’re keeping the rate so high that they can be sure that there are practically no loans worth making. If you had a car where the pedals were so sensitive that you could either floor the gas or slam on the brakes, effectively, but nothing else, you’d stay on the brakes, and press hard enough you were sure it wouldn’t slip.

Marilyne Tolle
Jan 19 2017 at 1:33pm


Below are a few points which will hopefully help to clear up some of the confusion about the determination, role and remuneration of central bank reserves.

On the system-level determination of the total amount of reserves held by banks at the Fed, a New York Fed briefing called “Why Are Banks Holding So Many Excess Reserves?” (2009) is the clearest exposition I’ve seen on the topic.

To understand the underlying mechanism, it helps to remember the basics of double-entry book-keeping.

When the Fed expands the asset side of its balance sheet – whether by providing emergency loans to banks (e.g. after the collapse of Lehman’s in September 2008) or by buying MBSs and Treasuries (known as Quantitative Easing, starting in December 2008), the counterpart to that must be an increase on the liability side of its balance sheet, in the form of bank (excess) reserves (I have to admit that I didn’t quite understand Robert Hall’s characterisation of reserves as being “borrowed” from the banks; as far as I know, reserves are usually thought of as being “created” by the central bank, in much the same way as commercial banks ‘create’ deposits when they make a loan).

Looking under the hood, when the Fed buys $1bn of Treasuries from, say, a pension fund, it finances the purchase by crediting $1bn of reserves to the pension fund’s bank (assuming the pension fund doesn’t have its own account at the Fed). At the same time, the pension fund’s bank gives the pension fund $1bn to deposit in its bank account.

So, from the Fed’s perspective, its assets (Treasury holdings) have gone up by $1bn, and its liabilities (the pension fund’s bank’s reserves with the Fed) have also gone up by $1bn. From the perspective of the pension fund’s bank, its $1bn of additional reserves at the Fed represent an expansion of the asset side of its balance sheet, matched by the extra $1bn in the pension fund’s deposit account on the liability side.

The point is that both central bank reserves (which make up the bulk of “base money”) and bank deposits (the electronic money that makes up the bulk of “broad money”) have increased, but the creation of central bank reserves is simply a by-product of the transaction.

Reserves don’t play a central role here, because banks cannot lend out those reserves directly to households and firms; they can only lend these reserves to other banks in the interbank money market (keep that in mind for later as it’s central to the implementation of monetary policy).

The new reserves are not somehow mechanically “multiplied” up into new loans (under this system, the money multiplier theory is a construct with no empirical grounding). In fact, no new loan has been created as part of this transaction.

What matters is the new bank deposit held by the pension fund, which it can choose to invest in corporate bonds or shares – effectively substitutes for the Treasuries it used to own (that’s what’s referred to as the “portfolio rebalancing” channel of QE, which is meant to boost asset values and lower the cost of corporate finance and so increase nominal spending in the economy).

A corollary is that the new reserves created by the Fed’s QE do not directly change the incentives for banks to lend more to the real economy. Banks decide to lend when there are profitable opportunities, which depends on several factors, including how fast the economy is growing and their cost of funding (and also financial regulation but that’s a topic for another day).

Here QE can have an indirect effect on bank lending via:

1) portfolio rebalancing: if the sellers of Treasuries to the Fed use the proceeds to buy private-sector assets or goods and services, the boost to the economy could encourage banks to extend more loans to households and companies (but it’s also possible that the sellers will use the proceeds to repay bank loans, in which case there would be loan destruction, not creation);


2) funding costs: reserves are the means by which central banks implement monetary policy, by influencing the short-term interest rate at which banks are willing to lend to each other in the interbank money market (remember the point made a few paragraphs back). And changes in interbank interest rates in turn affect banks’ choices of interest rates on customer deposits and loans, which influences the profitability of loans and so banks’ incentives to lend.

So, overall, there’s no a priori reason why Treasury purchases by the Fed should necessarily translate into more bank lending. It’s an empirical matter that depends on the relative magnitudes of opportunities for profitable lending vs corporate deleveraging.

In any case, I think that this explanation is a fair representation of the ex-post account given by the central banks that did QE after the financial crisis; the stress has been on portfolio rebalancing and the associated wealth effect, not bank lending.

Let’s now turn to monetary policy and the remuneration of reserves.

My sense is that the Fed’s decision to pay interest on different types of reserves reveals a deep-seated concern about losing control of short-term interest rates.

In the US, the interest rate at which banks lend reserves to one another is called the effective federal funds rate. The FOMC sets the target federal funds rate, which is the Fed’s policy rate. But the target rate and the effective (realised) fed funds rate have differed since the financial crisis. Why?

At this point, it helps to get into the difference between required and excess reserves.

Required reserves are funds that banks must hold at the Fed, in order to back a fraction of their deposit liabilities, and thus meet any demands for deposit withdrawals by bank customers (as an aside, that’s where fractional reserve banking comes from). Excess reserves are funds that banks hold at the Fed beyond what is mandatory.

The Fed pays a positive interest rate on required reserves to compensate the banks for the opportunity cost of being forced to hold these reserves.

Before the financial crisis, banks’ reserves at the Fed pretty much consisted entirely of required reserves. And the effective funds rate tracked the target rate pretty closely.

But after the collapse of Lehman’s, the Fed started to ramp up the size of its balance sheet, first by providing emergency liquidity to banks, and then through large-scale purchases of Treasuries and securities from the Government Sponsored Enterprises (GSEs). That translated into an explosion of excess reserves on the liability side of its balance sheet.

By definition, banks are not required to hold these excess reserves at the Fed – they can lend them to other banks. You’d expect the large amounts of excess reserves to bear down on the effective federal funds rate (the overnight rate at which banks are willing to lend to each other). And that’s exactly what happened.

So, to prevent the effective funds rate from falling too far below the policy target rate, the Fed began paying interest on excess reserves (IOER) in October 2008.

The idea was that banks should have been unwilling to lend to each other overnight at a rate lower than that they could receive from the Fed. In other words, the IOER would establish a floor under the effective funds rate, and allow the Fed more control of short-term interests.

But there’s a catch. By law, the Fed is not allowed to pay interest on excess balances held by non-depository institutions such as the GSEs, who are willing to accept any non-zero return on their deposits. As a result, the effective fed funds rate has remained below the IOER since its introduction, undermining the Fed’s control of short-term interest rates.

That’s where the “new kind of reserves” mentioned by Robert Hall – the overnight reverse repurchase (ON RRP) agreements – comes into play. Basically, when the Fed conducts an overnight RRP, it sells a security to a financial institution and simultaneously agrees to buy it back the next day. The difference between the sale price and the repurchase price, together with the length of time between the sale and purchase, implies a rate of interest paid by the Fed on the transaction. And that interest rate (the ON RRP rate) plays the same role for the RRP counterparties as that of the IOER for depository institutions. In other words, the counterparties that use the RRP facility should be unwilling to invest funds overnight with other counterparties at a rate below the ON RRP rate. So the ON RRP also establishes a lower bound for the effective fed funds rate. And you will have guessed it, the RRP counterparties include…the GSEs.

So, all these different interest rates on reserves are just tools to give the Fed greater control of monetary policy as it lays the ground for policy normalisation.

Robert Hall is absolutely right to say that Bernanke said early on (probably a reference to his speech “Federal Reserve’s exit strategy” given during the false dawn of 2010) that the Fed has two ways to normalise policy: either sell the assets and drain the excess reserves from the financial system, or raise the federal funds rate.

But raising the target rate first is pretty pointless if the Fed can’t control the effective rate because of the excess reserves sloshing around the system. In fact the Fed looks so concerned that in addition to remunerating the excess reserves of depository and non-depository financials, it’s also created a Term Deposit Facility as an option to reduce the quantity of reserves, by getting banks to convert a portion of their excess reserves into interest-bearing fixed-term deposits.

This is pretty technical stuff but I’m hoping it will add value to the discussion.


Comments are closed.


EconTalk Extra, conversation starters for this podcast episode:

This week's guest:

This week's focus:

Additional ideas and people mentioned in this podcast episode:

A few more readings and background resources:

A few more EconTalk podcast episodes:



Podcast Episode Highlights

Intro. [Recording date: December 8, 2016.]

Russ Roberts: Before introducing today's guest, I want to encourage listeners to go to'll find a link there [in the upper left corner] to vote for your favorite episodes from 2016 and give us some additional feedback. Thank you so much.


Russ Roberts: Now, on to today's guest, who is Robert Hall.... I want to start by discussing a recent paper of yours that we'll link to, "The Anatomy of Stagnation in a Modern Economy." And I'm going to read the opening paragraph. It's pretty blunt and dramatic.

In the years following the global financial crisis of 2008, many modern, advanced economies suffered stagnation. Unemployment rose sharply and declined slowly, output fell substantially, and growth remained substandard even eight years later. Investment in plant, equipment, software, and research and development languished. Productivity grew well below its historical rate. Monetary and fiscal measures to offset these developments were aggressive, but were only partially successful. This paper studies these events as they occurred in the single largest advanced economy, that of the USA.

A lot of people have argued that, while fiscal and monetary policy were only partially successful, that's simply because they just weren't aggressive enough. Do you agree?

Robert Hall: Well, first of all, with respect to monetary policy, the standard policy of lowering interest rates was done about as aggressively as possible. Maybe not quite, but pretty close. With respect to fiscal policy, really a lot happened; but it happened mainly because of what are called the automatic stabilizers. Discretionary fiscal policy, as in previous contractions, was not politically feasible, it turned out. So, I think I'd probably agree that in some ideal world more would have been done on the fiscal side.

Russ Roberts: So, when you look at the Stimulus Package of the early days of the Obama Administration, which was, I don't know, what ended up being about $800 billion--why was that not a significant amount?

Robert Hall: Well, first of all, a large amount of that money went to state and local governments. And rather than expanding their fiscal effects, a lot of them, a lot of the governments used them to pay off debt. Understandably. If you stare at the National Income Accounts on the government purchases of state and local governments, you don't see any sign whatsoever. In fact, you see a contraction. So, calling it $800 billion is a bit of a myth, in terms of what was actually delivered as stimulus to the economy.

Russ Roberts: And another good chunk of it was tax rebate, that just--it's interesting to me. There's this myth that it was all these shovel-ready infrastructure projects. I don't particularly think that would have been a good idea or successful. But it's not what we did.

Robert Hall: I think that's right. If you look at the National Income and Product Accounts, at the category that would include these supposedly shovel-ready projects, you just don't see any expansion. Remember that most of that type of spending is done by state and local governments, not by the Federal government. So, the fact that [?] the Federal government couldn't articulate a change in what the other government units were doing came through very clearly. But, all research of every contraction assumes that the same thing happens. So, this is not really a surprise.

Russ Roberts: When you say, "the same thing," what do you mean?

Robert Hall: I mean, the Federal government would like to expand the economy by inducing state and local governments to spend more. But they don't have the tools to do it--at least, they have neither the political will nor the tools to make it happen. So, there's a long literature, as long as I've been an economist, which is a long time, that shows that discretionary policy aimed at increasing government purchases like infrastructure just don't work. Or, if they do work, they don't start stimulating the economy until it no longer needs it. There's some work that shows this. It's actually perverse, because it takes so long to get spending cranked up; if it ever does, it's when the economy is back into a boom condition.


Russ Roberts: So, what do you see as the set of factors that explains the mediocrity of this recovery, the fact that, as you say, eight years into it, into the recovery, we see very little recovery or improvement in investment? In the labor market, a lot of people have suggested it's been very disappointing. Even though unemployment is low, employment, labor force participation is low. What do you think explains it?

Robert Hall: Well, the two big factors you just mentioned, one of them, labor force participation totally unexpectedly fell several percentage points--at least 3 percentage points. More than you'd expect, even given the deep recession. So that's one factor. Even more important is simply the slowdown in productivity growth. We had productivity growth, but at much lower rates. And that's just fundamental to the amount of output. You know, I've studied the labor market pretty carefully, and I'm convinced--and other papers that I've written recently make the argument that the labor market is really is back to absolutely normal conditions. The unemployment rate of 4.6% is well below the long-run average of 5.9% for unemployment. Key factors like how long it takes an employer to fill a job--it takes longer than it ever has before to fill a typical job. That's a sign of a tight labor market. So, at this point, we can't say that there's a lingering effect in the labor market. But there are other lingering effects, which are--to quantify them, that paper you mentioned.


Russ Roberts: What do you think of the work of Casey Mulligan, who has been a guest on the program talking about the impact of changes in various welfare programs and their implicit marginal tax rates that have discouraged workers?

Robert Hall: Well, I think, philosophically, I agree with a lot of the points you've made, which is that we've created a system where people become dependent on programs, lose their benefits so aggressively that they become trapped. It just doesn't make sense to enter the labor market even though it's back to normal because of losing benefits. And there was a big expansion of benefits which has not completely gone back to normal, as a result. Casey's numbers are in some ways hard to interpret. I wouldn't want to endorse--certainly, the theme of his book, which was the entire recession was caused by it--does not square with anything I looked at. What I see is that about 8 million people suddenly lost their jobs, mostly in the late Fall of 2008 and the first half of 2009. And that was a major factor. Working that off was actually a--very time-consuming. And there's no way that it makes sense to say that those people only lost their jobs because there was an increase in benefits and they quit their jobs. We know they didn't quit their jobs. So, that part of it, I could--in my analysis, overblown. But the need for reform of benefit programs--Disability probably the most important one--is very acute. And there's just agreement across the board on economists, that point.

Russ Roberts: Related to that: Do you see the long secular decline--secular meaning over time--decline over time in prime age male labor force participation to be explained in part or in large part by that expansion of Disability opportunities? Or is something else going on? Is it people's skills--people are worried that, you know, these remaining factory workers, a lot of whom will not find good alternatives and are not returning to the labor market in the normal way?

Robert Hall: You know, it's a combination of all the things you mentioned. Probably some more. First of all, among men, it's a very long-term phenomenon.

Russ Roberts: Yeah.

Robert Hall: Labor force participation was much higher in 1950. And then it's--among men--and it's declined, a pretty smooth trend. Whereas the situation for women is quite different. There was a big increase in participation among women; and then, about 10 to 15 years ago that changed and you see some decline in participation by women. But, you know, work has become less common. And particularly work among young people--and most [?] factors that about--in the year 2000, not that long ago, about half of all teenagers at any given time were working. Today, less than 25%. So there's been a decline in half of the fraction of teenagers who are employed. That's the most extreme number. But you see the same thing among the 20- to 24-year olds. And you see it among older men, especially. And that's very tied in with evidence about the declining health moves[?] age groups, and the absorption by the Disability. Which there was bulge[?] of people going on to Disability, which has not reversed itself--going on Disability is pretty much a trap, in the sense that relatively few people make that decision and are qualified than exit.

Russ Roberts: We had Erik Hurst on recently talking about the--you mentioned teenagers. He found in his work with some co-authors, I think it's 21-30-year-olds who were non-schooled, I think 18% did not work in the previous year, compared to 8--I can't remember the time frame but it's a short time frame, and a lot of these folks are living at home.

Robert Hall: Playing video games.

Russ Roberts: Yep. That's what he says. I'm not sure it's--

Robert Hall: I followed the progress of that paper in detail.

Russ Roberts: I'm a little bit skeptical. Let's move to--or at least that video games are driving the change, which is what he's--I'm sure it's some of the [?].


Russ Roberts: Let's go to productivity growth, which you said is sort of the crux of the matter. A lot of people have argued recently, or at least a lot of prominent people have argued recently that the reason investment is down and productivity is down is because we've figured out all the interesting and productive things to invest in. What's your view on that hypothesis? And if you disagree, what do you look for to think about it?

Robert Hall: Yes. So, Greg Ip had an article in yesterday's Wall Street Journal. It's such a tired old story. And, you know, we have surges in productivity. There was productivity pessimism, say, around 1990, similar to what there is today. And a similar thing--all the good ideas have already been implanted[?]. And then the Internet came along and gave this very remarkable, or things like it.

Russ Roberts: Well, 'That was the last one! That was it. There was one left. We found it.'

Robert Hall: Okay. I gave a talk recently to a popular audience about computer vision and the changes that it makes. And for one example, I showed in the video the way trash used to be collected from households. Which, you'd have a truck with a team of 3 guys, 2 of whom rode outside and jumped off--

Russ Roberts: Yep--

Robert Hall: and emptied a trash bin and then pushed a button; and the thing went up; and then they would go to the next house. Today, the trash in most places, certainly where I live, is picked up by a truck that has 1 guy. And he barely stops. And then computer vision takes over. It spots your trash bin, wherever it is; reaches over. It adjusts for any mistakes the driver made, flips the trash--at least 3 times more quickly than the guys can do it. So, they have a 3-fold reduction in the number of people, from 3 to 1. And you have a tripling of the speed. So, the productivity of collecting trash has gone up by a factor of 9. And that's just scratching the surface of what computer vision can do. You know, the excitement about computer vision today is mainly a self-driving car, but there's so many other places. And it's a very, very intensive area of work by computer scientists. There's many, many tech startups right now whose main mission is to improve computer vision, and the improvements that have been made in the last few years are just stunning. So, I, you know, I don't try to forecast; but I think this productivity pessimism--I emphasize productivity uncertainty. Episodes like we've had recently, of a slowdown in, say, over a 6-year period, have been common. There's nothing statistically surprising about it. And they've typically been followed by productivity growth going to normal or even above normal. So, we have waves of productivity pessimism, and we have waves of productivity optimism. And that's what I've seen. And I've been watching this for quite a while.

Russ Roberts: Well, your remark about the trash pickup reminds me, not--at my current weight level [?]--to stand near the curb as the trucks go by. Could be mistaken. They are still getting some of the bugs out, I'm sure. I wouldn't want to be picked up by mistake.


Russ Roberts: But it's an example of where technology is being applied. And it may or may not be showing up in the data. What do you think of this argument that, say, in GDP (Gross Domestic Product), so many of the pleasures of life of the last 10 years are not very monetized, and we are not measuring a lot of the gains correctly in the National Income Accounts?

Robert Hall: Okay. Well, Chad Syverson, Chicago, has written a very interesting paper. We're[?] giving I think what started out as a kind of a neutral evaluation at that point. And he showed, using various arguments that sort of put it in a prelimit[?], that upper limit is pretty limited. The idea that we're somehow missing some huge benefits. Of course we are--

Russ Roberts: But we always have--

Robert Hall: but, you know, for example, the whole revolution in photography is not captured at all in our data. They still collect only--they base statistics on the assumption that people are using film and sending it off to be developed. And I wouldn't know where to go with film to be developed.

Russ Roberts: Right; I've spent $100 for software program to develop on my computer. It's a one-time payment. They upgrade it every once in a while. But I take 10,000 photographs on my computer--I have about 10,000. And it's glorious.

Robert Hall: I do exactly the same thing, probably the same software. Exactly.

Russ Roberts: And it's not being measured. Certainly my pleasure from it is not being measured. Certainly the pleasure I get from not having to re-take--only take 12 pictures and worry I got the right ones that I can take 1000 is just wonderful.

Robert Hall: Yeah. Exactly. So that's a very concrete example. But I think it's fair to say, when you look at what the economy produces and what we consume, there's just an awfully large number of things where we know nothing much has changed. Like the generation of electricity. Or building houses is a great example. It's a very large budget share; it's a growing budget share for housing in the United States. And, houses are being built today with exactly the same technology virtually as they were 100 years ago. It's a very big sector, much bigger than any of the ones that people are excited about, like photography, but it's not one where productivity growth has occurred. You have to get realistic. If you stare at the numbers for what we make, it's an awful lot of totally boring stuff where you know there's not much change in productivity going on or even possible.

Russ Roberts: So, when you say you're not a pessimist or an optimist, that you are a realist, the way I'd describe it, the fact that the last 8 years have been so disappointing--you're saying that's just a standard kind of slowdown that happens in the data now and then? And not to be worried about it?

Robert Hall: Well, I don't know--I hesitate to say that it's 'just,' because I've received a lot of study--like John Fernald, at the San Francisco Fed has been kind of the highly knowledgeable expert on that point, and he's written a number of papers there that everybody should read who is interested in this subject. But, it seems like there was this period of rapid adoption of revolutionary information processing technology, and that's somewhat slowed down oddly because there are so many places that you deal with every day that could be immensely improved. Like the airline that I fly that I won't name has a ridiculously bad website, a notoriously bad website. And if they could just hire the right people, who are numerous, especially in Silicon Valley, to write a decent website it would be a huge step forward for all of us. So it just hasn't happened. And we sit there waiting--every single time you enter anything it does a whole page load. Nobody writes websites that way today. That's just pathetic. That's the website of 1995. Twenty years ago.

Russ Roberts: Well, the cost of revamping that website, the time it would take and the--I guess they don't think those 3 seconds for that re-load, which are annoying to us, are going to be that valuable. And they don't think they can capture that monetary gain, maybe. I don't know.

Robert Hall: But it's odd because the companies that grew up with the web, especially Amazon--why doesn't Amazon start an airline? It seems to me that would just be perfect.

Russ Roberts: Well, they're close. They are really close, actually.

Robert Hall: It wouldn't surprise me, if they did.

Russ Roberts: Nor I.


Russ Roberts: One thing you haven't mentioned that we haven't talked about yet is interest rates. And one of the things that people have noticed--it's funny, it's been going on for decades but we just sort of noticed it, at least some of us have, at least some people just started writing about it--they've been falling for a very, very long time, through both good times and bad times.

Robert Hall: Well, that shows that you're younger than I am.

Russ Roberts: I don't know about that.

Robert Hall: In the first half of my career, they went up.

Russ Roberts: Well, that's true.

Robert Hall: They reached a spectacular peak, around 1981. And then it's been down from there.

Russ Roberts: But that's a long time--

Robert Hall: Of course, a lot of that--we are talking about nominal interest rates, though they were boosted a lot in the 1970s by growing inflation. They don't [?] monetary policy that further increased, just a double-whammy of high interest rates. And then we got back to normal, during the 1980s. But then other things took over: We got inflation down, so that was no longer an important factor; but there were other things which resolved into declines in real interest rates. It's normally that case that--first of all, interest rates are global. This phenomenon all over the world. And world economic growth has slowed recently. It's not just the United States, but China, for example--a big factor is China. So, the decline in growth is one thing. And there's others. I've just written a paper on this--you're invited to read, of course.

Russ Roberts: Yeah; we'll put a link up to it. But isn't that just the flip side of this concern that--this slowing of growth is the flip side of--I'll just say it's a puzzle. Normally after a recession, certainly a recession of the size we had in 2008, certainly the worldwide nature of it, it would be followed by a very brisk and dramatically recovery. We haven't seen that. Growth seems sluggish. What do we think is going on? People ask me; I just say 'I don't know.' It's easy for me. But 'I have no idea,' is my answer.

Robert Hall: I know something.

Russ Roberts: Well, tell me what you know.

Robert Hall: Okay. So, what I do know is that the labor market has behaved very much the way that it has historically. There was nothing very surprising about what happened in the labor market, once you come to grips with the fact that it was a very large shock. So, in terms of immediate job loss, it was by a considerable margin the worst.

Russ Roberts: In a long time.

Robert Hall: Yes, well, yeah, over the period since we measured unemployment, which started in 1948. So, that triggered this normally, fairly long period in which unemployment gradually declines. But that's always been true. People always get excited about it because recessions don't happen very often. For example, most reporters who are reporting on the Great Recession were not reporters when the previous recession occurred in 2001.

Russ Roberts: Correct.

Robert Hall: But, if you take--if you look--here are the numbers on job destruction. There were 8 million jobs destroyed, as I mentioned before in late 2008 and early 2009. There were about 5 million jobs destroyed in the 2001 recession. Which was actually a pretty bad recession from the job-destruction point of view.

Russ Roberts: Mild enough [?]

Robert Hall: Other things--it was followed--so, the reason that--so, we had what amounts to, structurally, a pretty normal recession/recovery, except that it was really bad. So, it was just scaled up. But, qualitatively it was all the same things would have happened previous recession. The thing that makes everybody think that recovery was slow is not what we should be calling a recovery at all. It's what happens to productivity. Productivity doesn't systematically do one thing or another over the business cycle. Here, we had low productivity growth at the same time that we had a normally-structured recovery from a very bad shock. But the combination gave an extended period, going up to today, of disappointing results. As of today, all the disappointment comes from things that are not related to recovery. The recovery is complete. But we are not where we'd like to be because productivity growth has been bad. And also because we have a legacy of not having formed much capital since 2007. And that's really hurting, too. And that was another of the big factors in the paper that you started the discussion with.

Russ Roberts: And to summarize that: So, that's a nice way to think about it, right? Low capital formation and low productivity; and of course in a way they are almost saying the same thing. They are certainly related.

Robert Hall: No, not the way--if we are talking about what economists call 'total factor productivity.'

Russ Roberts: Explain.

Robert Hall: So, it's the--normally the economy grows faster than you'd expect, given the fact that normally employment is rising and capital is rising. But output rises even more than you expect. And that's what's called a 'residual.' And the residual is also called 'total factor productivity'. The 'total factor' means that it's considering and giving credit to all factors, capital and labor, and not just labor. So, when you talk about total factor productivity it's already allowing--in this case it's allowing for the fact that--

Russ Roberts: [?] or less--

Robert Hall: capital formation had been very weak. By some measures it's back to normal today. It's very tricky to figure out what the right measure is. But there's still a legacy of capital that would have been formed if we hadn't had a severe recession which wasn't. And that's a nontrivial part of the total.


Russ Roberts: So, a lot of people want to blame a lack of investment on uncertainty of various kinds--policy uncertainty. We just had a 2-term President who came in and overhauled the health care sector and financial regulation, two significant parts of the U.S. economy, and overhauled the ways where the impact of it wasn't clear for a while. I think the legislation is still somewhat incomplete, even today.

Robert Hall: There's a question of whether it's viable, the premiums are going up so rapidly that nonviability is definitely an issue.

Russ Roberts: So, a lot of people, certainly on this program, who share my free market views have, you know, said, 'Well, of course the economy is lousy, or investment is lousy. It's because there's a lot of uncertainty.' My thought is: I'd love for that to be true, for my biases. But there's always a lot of uncertainty. Now we have a President who is really uncertain: what policies will be put in place. We'll see what is going to happen.

Robert Hall: Surely there is more uncertainty now than ever before--

Russ Roberts: Than ever.

Robert Hall: But the stock market loves it. So, it's a--I'm very familiar with that literature and I certainly respect the people who have worked on it. But, the connection they--getting a direct sort of causal connection has proven to be elusive and then in the literature, what is true is that--actually, 9/11, which my colleague has worked on, is actually a very interesting test of that proposition. In fact it was the first thing that Bloom looked at. And what happened there was there was clearly an enormous rise in uncertainty after 9/11. There was a brief period of diminished spending, investment and consumption spending. And then, you know, people looked around and said, 'I guess not.' And it occurred toward the end of a recession. And the economy behaved quite normally after that; and then we had a pretty good expansion through the end of 2007. So that was kind of a laboratory for that idea. It shows that the idea is valid; but on the other hand, the effect of what was kind of a transforming experience for many of us, on the economy was remarkably small just after a few months. By the way, there's another hypothesis floating around. The Council of Economic Advisers has a paper on its website questioning this idea that there's been a substantial increase in market power--

Russ Roberts: Yeah, I saw that--

Robert Hall: in this economy, and that translates directly into bad news on several fronts, including capital formation: Market power functions like a tax to discourage economic activity. I don't know if it's true. So far, as market power is one of the topics I worked on 30 years ago, and found it remarkably elusive to measure. And I don't--I think there's kind of a presumptive case in favor of it; there has been some increase in concentration. But I think it's going to be an area of research. I think that the policy uncertainty research has sort of done everything they can do--and they've done a lot and I think they've convinced some people, perhaps not me. But I think this market power hypothesis is going to get a lot of attention as another thing that's holding back, and maybe even holding back productivity. Now, to me it's a somewhat more promising idea.


Russ Roberts: Let's shift gears to monetary policy. I ask my guests, and I get asked a lot by my friends, why the Fed, why banks are holding excess reserves in the magnitude that they are at the Fed. And I get asked this question--it's usually a related question, which is, 'If the Fed was so aggressive, why didn't we have inflation? And does that mean that Milton Friedman and others were wrong?' And then people respond, 'No, because the banks are holding the money. And when the money goes out in the economy it will cause inflation.' And yet--I have not heard a good theory for why the banks are holding onto massive amounts of excess reserves rather than investing them. This is related somewhat to our previous discussion. So, what are your thoughts on these issues?

Robert Hall: Okay. Well, first of all, the first question is easy to answer. The banking system as a whole has no choice about the level of reserves. It's completely dictated by the Federal Reserve. If one bank decides to invest its reserves, the reserve just land at another bank. It's a sealed system [?]. There's a lot of misunderstanding of this point, but it's absolutely central to understanding monetary policy to understand that banks collectively have no choice whatsoever about their holding of reserves. It is completely determined on Constitution Avenue [i.e., the Board of Governors of the Federal Reserve--Econlib Ed.] So that's the first point to make.

Russ Roberts: You want to explain that? So, the normal way that this story gets told is that the Fed intervened in 2008, 2009, I think even kept going--quantitative easing of various kinds. And they bought a lot of assets that banks held, both government Treasuries as well as mortgage-backed securities. And they paid for those by crediting those banks on the balance sheets that they have at the Fed, with reserves that they could then lend out.

Robert Hall: Yeah, you know a more instructive way to say it: They borrowed the funds from the commercial banks and used the proceeds to buy bonds. Because a lot of people don't understand that the process that you just described is simply borrowing in the capital market. The Fed greatly expanded the national debt by issuing what are Federal obligations, borrowing--which are reserves. It's as simple as that. It's just a bank. It gets funds from one source and it uses them to buy securities of a different type.

Russ Roberts: And so the expansion? The so-called [?] where did it come from? Just--that's electronic magic.

Robert Hall: No, it's just as you described it. It's a conscious thing that the Federal Open Market Committee directs, [?] like to say, the Trading Desk to start buying stuff. Every time they start buying stuff, every time they buy stuff, as you pointed out, they do it because they have this automatic right to borrow from the banking system by creating more reserves. So they write a check, so to speak, which is increasing reserves. And they use it to buy bonds.

Russ Roberts: But you are suggesting that the banks, which have statutory limits--minimums--of what they have to hold at the Fed, now hold, because of those actions, they have much larger amounts on the Fed balance sheet. And if they try to lend those out to, say, homeowners and home purchasers, they can't do that--they can't change that, net?

Robert Hall: The system can't. Any one bank can. So, one bank can say, 'Oh, we don't like what they are getting today which is 50 basis points, 1/2 of 1%'--

Russ Roberts: That's the interest that the Fed is paying them for holding them there.

Robert Hall: Exactly. Right. And they look around; and that's completely safe; and they say, 'What can we get on other safe investments?' Well, less than that. Okay. So, they have absolutely no incentive, given the current state of the money market to do anything. Now, suppose that--

Russ Roberts: That's a separate point. That's a good point. I was going to get to that. But that's not the main point. Keep going.

Robert Hall: Okay. But then if you ask, 'What would happen if Alan Blinder,' who has been very vocal in The Wall Street Journal on this, and others, including me, but not so much in print, said, 'Well, that's a silly policy. Let's change the 50 basis points to 0.' Then banks would look around and they'd say, 'Ah! Now I'd really like to invest.' And so they would start, as you were saying. It's, 'I'm not getting the market return from reserves any more. I'll try to get rid of them.' But every time they try to get rid of them--which they do aggressively. But the reserves always land at some other bank. If you buy something--if you make a loan, then the funds are deposited in someone else's bank account; and that means that the corresponding reserves are moved to that person's bank. There's nothing the system can do. One of the anomalies here is that, for crazy reasons which I won't go into and don't fully understand myself but I know that my colleagues do--a lot of the banks that--the reserves under certain circumstances can be held by foreign banks, like Deutsche Bank in particular. So, Deutsche Bank, at one time--I don't know if it's true now--was actually the single largest bank holder of excess reserves. And they were harvesting money from the American taxpayers because the Federal Reserve for completely unexplained reasons was paying more than the market rate. This paying excess--being over the market rate--went into effect in October of 2008. At exactly the wrong time. Blinder just sputters; and he's a very middle-of-the-road kind of a guy. He's not political. Just economics.


Russ Roberts: Well, it's a mystery to me, too. When I--that was my next question, which is: there are two ways to ask the question, which is: One, the question is why the Fed started paying interest on reserves. The standard economist's answer is--and I know you've written a paper on it recently and you're going to be talking about it shortly--is: It was new way, it was a new instrument for the Fed to use in effecting monetary policy. For those of us who might not be economists, who might be political scientists or just everyday people, it looks like a way to take care of the banks through a tough time when they had very troubled balance sheets. Economists don't like to give that answer. Most of them. What's your explanation? In October of 2008, what were they thinking? What were either the economic ideas or the political forces that made that happen?

Robert Hall: Hard to say. Again, I'm not the--I somewhat stay away from that because--I never thought that, during the time especially when it was 25 basis points that it was the most important policy issue that some people do, like Alan Blinder did. I don't think we have a good answer to that. I think everyone is puzzled. Everyone applauded the Fed using the rate they pay on reserves as an instrument: in fact, certainly my feeling is that it should be the instrument of monetary policy. But if you asked then what I thought the rate should be, I would have said it should be negative. For example, the European Central Bank currently pays -40 basis points. There's actually a big difference. Obviously Europe needs to be stimulated more than the United States today, but even today, the 50 basis points they are paying, quite a bit more than they should. And they know that because they've actually created a new kind of reserves, the RRP. This is technical stuff; but they recognize it so fully that they've actually got two kinds of reserves now, and they pay the right interest rate, which is currently 25 basis points, on this new kind of reserves. We're going to, I think, gradually see a swing toward a more reasonable use of this policy instrument, phasing out old-fashioned reserves and then using these new kind of reserves.

Russ Roberts: Well, let me ask one more question to capture this idea that the amount of reserves is fixed at the Fed and they in fact just move it around. So, if real interest rates, real productivity in the economy grow to, say, 2, 3%, historical levels while the Fed continue to pay a half a percent--

Robert Hall: Oh, well, that would be a big surprise.

Russ Roberts: Right. Well, if that happened, walk me through the chain as banks then tried to get rid of their reserves. What would happen?

Robert Hall: So, an individual could say, 'Wow. I can now make a lot more money lending out, even adjusting for risk. For one thing, the interest rates you are talking about would apply to Treasury Bonds, and then there--

Russ Roberts: [?] and then 5 or 6% with riskier things.

Robert Hall: Okay. But let's say there's a safe interest rate to 50--to 1/2 of 1 percent, the Fed is paying. And if you go to the Treasury and buy a Treasury Bill, which is equally safe and is short term, and you get, say, 3%--well, it's a no-brainer to spend your reserves to buy Treasury bills from somebody. But then what happens? That person puts it in the bank and then that bank is saddled with excess reserves. Okay. So, the way we used to teach, why monetary policy can expand the economy, was the same logic: that if there were reserves around which banks regarded as hot potatoes and wanted to get rid of them, then as they all collectively started lending more that would expand the economy. That was the basic theory of why boosting the quantity of reserves, which was traditional monetary policy--this is Milton Friedman monetary policy--

Russ Roberts: That's the way I was taught--

Robert Hall: Quantity of money.

Russ Roberts: Yeah, not interest rates, not that it makes investment look more attractive. Just, yeah, print money; spend it.

Robert Hall: But that whole mentality was developed during a time when reserves received zero interest. So, everything was based on the quantity of reserves: if the Fed increased the quantity of reserves it would put hot potatoes in the hands of banks; they would try to lend it out; that would expand the economy. As the economy expanded, then we'd settle into a new equilibrium where the quantity of reserves was what the banks needed to use and the expansion would stop. Today, we need to re-map that into thinking about what happens if you change the interest rate that's paid on reserves. And then the corresponding thing to increasing the quantity of reserves is to lower the interest rate; that low interest rate makes reserves hot potatoes; that expands the economy. So, you are describing a situation that would be intensely expansionary.

Russ Roberts: Correct.

Robert Hall: Way more than I think anyone would choose.


Russ Roberts: Well, the only thing I'm confused about--it's probably more than one thing--is that, I thought you were going to say, as they tried to acquire those Treasuries, that demand is going to push the return on those back down, and make it hard for those interest rates--I'm wondering what the interest rate the Fed pays on reserves, how that affects the real interest rate in the economy and the rest of the economy.

Robert Hall: Well, it's very persuasive. That is, it simply has to be the case that Treasury Bills, which are essentially the same thing as reserves--they are safe obligations of the Federal government--they can't pay a return that's very different.

Russ Roberts: Correct.

Robert Hall: So, all kinds of things happen, as you were saying. But you are starting from a point where the Fed would never be. They would never be in a situation where market interest rates were that different from what they had chosen to pay on reserves in the new regime, because, remember that with $3 trillion dollars of reserves outstanding, they have to make those attractive enough. The stresses that would be caused in the situation you are describing would be--the economy would practically explode because of this difference. So, what you'd see is the Fed would never get into that situation. They would be inching up, just as they did; and what they're going to do next week [note: episode recorded December 8, 2016--Econlib Ed.] is probably add another 25 basis points, because they see market rates rising and they say, 'What matters to us is the difference between what we pay on reserves and market rates and we have to keep that differential fairly small because we've got $3 trillion dollars of reserves out there.' In contrast to the days when there were $17 billion dollars of reserves, where they have a kind of a free hand and in fact in those days didn't pay any interest on reserves and sometimes had a very large differential. But $3 trillion is a very different number from $17 billion.


Russ Roberts: Well, the only thing I'm confused about--it's probably more than one thing--is that, I thought you were going to say, as they tried to acquire those Treasuries, that demand is going to push the return on those back down, and make it hard for those interest rates--I'm wondering what the interest rate the Fed pays on reserves, how that affects the real interest rate in the economy and the rest of the economy.

Robert Hall: Well, it's very persuasive. That is, it simply has to be the case that Treasury Bills, which are essentially the same thing as reserves--they are safe obligations of the Federal government--they can't pay a return that's very different.

Russ Roberts: Correct.

Robert Hall: So, all kinds of things happen, as you were saying. But you are starting from a point where the Fed would never be. They would never be in a situation where market interest rates were that different from what they had chosen to pay on reserves in the new regime, because, remember that with $3 trillion dollars of reserves outstanding, they have to make those attractive enough. The stresses that would be caused in the situation you are describing would be--the economy would practically explode because of this difference. So, what you'd see is the Fed would never get into that situation. They would be inching up, just as they did; and what they're going to do next week [note: episode recorded December 8, 2016--Econlib Ed.] is probably add another 25 basis points, because they see market rates rising and they say, 'What matters to us is the difference between what we pay on reserves and market rates and we have to keep that differential fairly small because we've got $3 trillion dollars of reserves out there.' In contrast to the days when there were $17 billion dollars of reserves, where they have a kind of a free hand and in fact in those days didn't pay any interest on reserves and sometimes had a very large differential. But $3 trillion is a very different number from $17 billion.


Russ Roberts: So, what's the significance of that $3 trillion? Is it irrelevant? Is it important?

Robert Hall: It's a huge piece of the Federal debt.

Russ Roberts: I guess that's the right way to think about it.

Robert Hall: It would create what economists call a disequilibrium if the identical components of the debt paid different interest rates.

Russ Roberts: But don't people think--but that's a little bit--the Treasuries aside. Like you say, let's say those are just the same thing. Out in the real world there's some level of productivity that normally would be driving interest rates. I think what people are worried about is whether this $3 trillion, 3 point whatever-it-is in the Fed's books, is somehow affecting the real economy. You are saying it's no different from any other debt that the government runs.

Robert Hall: Well, no, it could be because the Fed does have a choice about what interest rate to pay. They do have that choice. I believe that what we'll see, and what we are seeing, and what we will see next week, is they'll track the rise in interest rates and keep the interest rate on reserves, in their case, a little bit above market rates. Blinder and I would say, we got that wrong: it should be a little bit below. But to the extent that they start switching reserves, which they could do at any time, from old-fashioned reserves to the modern reserves which pay a lower interest rate, 25 basis points less, then that would be a sign that they were reducing the subsidy that Blinder and others have been upset about. But I want to emphasize that's not a huge deal. This doesn't compare to some of the important policy decisions that are going to have to be made in our nation's capital soon, which is: Can we possibly finance our government?

Russ Roberts: Well, we're going to save that for another episode. But just to make sure I understand this: A lot of people say, 'Well, when are we going to get back to normal?' So, you are saying that it's not so unnormal to have $3 trillion compared to $17 billion, because it's just a different form of government debt?

Robert Hall: Absolutely. This question, when Bernanke started expanding the portfolio, very early in that process he put an essay on the Fed's website with sort of a speech that he'd given, which said, 'Look, we have two things we can do when normalization occurs. We could either keep the rate we pay on reserves at a low level and reduce our sales back to $17 billion of reserves. That's Strategy One; that's a completely viable strategy. Or, Strategy Two is we can pay market rates on reserves and we can continue to have trillions. And either one of those works fine. Don't worry: we've got two. We've got belt and suspenders.' And he's right. And, you know, Yellen has continued to make the same point as often as necessary. But [?] this idea that there was some kind of inflationary [?]--they've got two ways to prevent it from happening.

Russ Roberts: Yeah. I'm just not sure about the political economy there; and I'm not sure the political economy of having fiscal policy, the way you've described it, centered in the Central Bank rather than in the Congress--there's less accountability. I don't know.

Robert Hall: Yeah. Yeah. I'm not going to argue [?].


Russ Roberts: Okay. Well, let's turn--I'm fascinated by the NBER (National Bureau of Economic Research) dating business cycle--business cycle dating. 'I'm fascinated' is a bit strong. But I'm very interested in it. So I want to turn to that now. As far as I understand from the web, you've been the Chair of that committee since 1978, when it started--the official, of course they are dating for other recessions and booms, healthy times--that economists have created, and the NBER has created. But since 1978, you've been the Chair of this committee that decides when did a recession start; when did it end? What is that like? How often does that committee meet? Do you sit around for 20 minutes? Do you sit around for 3 hours? Do you argue a lot? How do you proceed to do that? Because a lot of us thought--I did, for example--that, 'Oh, it's easy. When you have two consecutive quarters of GDP (Gross Domestic Product) going down, that's a recession.' And then you see how long that goes for; and then it ends. That's not what you do, exactly. So, what do you do?

Robert Hall: Well, first of all the two quarters is a generalization that isn't bad. In particular, our dating of the turning point when the Recession began, the most recent recession in December of 2007, fits perfectly: there's just no controversy about it. On the other hand, if you go back to the recession before that in 2001, that idea kind of breaks down. And the two-quarter thing was in the days when GDP was only measured quarterly. But now GDP is measured monthly. And that was true actually when we made the decision, 2001; and it didn't help. There was kind of a zigzagging of monthly real GDP--not quarterly, but monthly--that still created a big challenge.

Russ Roberts: And if I remember, there was a revision of the GDP data that changed one of the quarters from negative to positive, I think, net, even though some of the months were negative, maybe.

Robert Hall: Yeah. That's of course something we have to live with. One thing about the committee is that in principle, if we get persuasive new information, we can change the chronology. That's never happened under the time that I've chaired the committee, or when the committee existed; it was done more informally before 1978. But we do have that. But anyway, you asked some other questions--

Russ Roberts: The logistics. Tell me.

Robert Hall: So, first of all, most of the time the committee does absolutely nothing. That is, it does not meet. It meets as soon as, say, some small number of members of the committee say, 'We'd better get started here: it looks like this expansion is finally coming to an end,' and we're going to have to eventually--often, you know, a year later or sometimes even two years later, we wait to be sure that we're right. For one thing, as the revisions of the data tend to be quite aggressive over the first year. So, we wait for the data to calm down. And we also have to deal with the fact that there's a possibility that, for example, some time in the next year that there might be a pause in the growth of the economy, but then growth resumes. If it resumes quickly enough, we'd say, 'Well, we don't call it a recession.' There's a criteria that we apply to decide whether, say, a small negative development is big enough to be called a recession or not. That's something we, in principle, have to think about. It's never actually been--it's always been clear to the committee in the time that I've run it--we've never had something that was just barely a recession. Either a recession is not a recession at all, clearly, or clearly a recession. So in that respect, although that's a theoretical problem, it hasn't been a practical problem. The work of the committee when it's active is almost all done by email: email comes in to me. So, spreadsheets go back and forth; opinions go back and forth. But then, before we actually make a decision, we have a conference call. And then work it out. Also, another important thing is we have a long discussion of the exact wording of the announcement. That's what clarifies the whole process--because we put out a press release. The whole thing is very disciplined: it's sort of like what the FOMC (Federal Open Market Committee) does. And we put out a press release, and it explains the logic: what data we're looking at, things like that.

Russ Roberts: The FOMC being the Federal Open Market Committee. So, I guess, [?] some of my illusions: I assumed you sat in a beautiful boardroom somewhere in Cambridge, possibly New York, and pounded the table and argued for June over July. But it sounds like a civilized process where spreadsheets do most of the heavy lifting. Is there ever any disagreement? Do you guys ever fight?

Robert Hall: Oh, sure. Oh, absolutely. Yeah. Yeah. Yeah. Yah. There's a certain amount of confidentiality that our operating rules require, though I can say, yes, there's a frank discussion; but I would be forbidden to say who said what.

Russ Roberts: I understand.


Russ Roberts: We're almost out of time. I want to get to a big-picture question. For me, the Great Recession was a big learning experience, a lot of dimensions. Mainly about the role of finance in the economy and the role of how investment banks work and what leverage did--that interaction between the financial sector and the real economy was just something I didn't know much about and had to learn a lot about. And probably a good chunk of what I learned was not true. But that was one of the things I had to learn. And certainly there was a lot of optimism in October of 2007, or November, before the December onset of that Recession that you mentioned. And a lot of people did claim at the time that we'd kind of figured it all out. And in the aftermath, it seems to me--there's two groups. There's people who said, 'Yeah, we really did understand it the whole time,' and they explain why the last 10 years or 8 years are explicable. And then there's another group who say, 'We've got to start from scratch,' or 'I've got to radically revise my view of the world.' One: Where do you fall on that group? And, two: How do you see the profession as a whole in its response to this event?

Robert Hall: I think, for the first observation I'd make is that there are papers that have been published and were well known in the literature that formed a basis for economists' attention to this question of how financially that's affect the whole economy. In particular, the Bernanke, Gertler, and Gilchrist paper, which was already famous, became more famous as a result of being kind of the backbone of the modeling and the upsurge of interest in the subject that obviously occurred after the Crisis in September of 2008. So, on the one hand I think it's just completely false to say that we were clueless and had never thought about this and that we had to start from scratch. On the other hand, of course a huge amount of interesting work has been done. Stimulated. And it's been a huge stimulus to further thinking and further integration of finance. The integration of financial thinking into macro modeling in general has advanced over this period. Faster than it would have otherwise, because it was so clear that the financial crisis triggered a big contraction of activity across the board. But, you know, I'm kind of in the middle of that. I don't--I reject this idea that the profession was completely unprepared; I never thought of it before. But I also recognize that we didn't see it coming. My wife is an economist who knows a lot about mortgage finance. And we kept a sticker on the refrigerator saying, 'There's only $250 billion of the prime mortgages out there. It can't be such a big deal.'

Russ Roberts: How important could it be?

Robert Hall: Exactly. How important could it be? Well, we had to take the sticker down. Some time even before September of 2008--we were actually past--the whole spring and summer of '08 was a time was everyone was getting nervous and the economy was in fact contracting. But it was a gradual process.


Russ Roberts: Well, let me ask a related question, which is: I think--I don't know if I'm younger than you, but I'm not so young. I'm 62. And I have to remind my guests that I went to graduate school in the late 1970s; and I went to Chicago, so I was trained with skepticism toward Keynesianism, which seemed to pervade an increasing portion of the profession as well, outside of Chicago. And then, with the view that--and certainly our colleague John Taylor will say, 'Oh, yeah, everyone knew in 1990 or 1997,' I don't know what year you want to pick, 'that Keynesian stimulus is ineffective. Short term stimulus is ineffective.' And then we come to the Great Recession. All of a sudden people are Keynesians again. And to some varying amounts, depending on where you go. And I did not realize until preparing for this interview that you are the originator of the 'Saltwater/freshwater' distinction, in a 1976 paper--Saltwater being the coasts of Cambridge and Berkeley and Stanford and MIT that are more sympathetic to the Keynesian models. I mean, there's a lot of differences in the distinction; I'm not going to put words in your mouth. But my real question is: Do you think we've made progress in our understanding of the business cycle? I was taught by Robert Lucas; he was a Mitchell fan. I got some taste of that history. It felt like, again, in 2006, we had really kind of "solved it." It seems like, to me, we are to some extent back to Square One. Do you think we've--but maybe not. So, tell me your view.

Robert Hall: No. I think a lot of the ideas that have permeated the consensus macro modeling--first of all, the term 'Keynesian' is--it means so many different things to different people. If you look at how most macroeconomists build a model, it has a huge number of things in common. They--a parting point, if, what's called the New Keynesian Model, isn't related to this question that you mentioned at all, whether or not fiscal policy is effective. It all has to do with whether the economy reaches its equilibrium quickly or whether there's a time when prices are wrong--that is, prices are sticky and wages may be sticky. But then there's a lot of work--and this is, certainly on the labor market side, this is something that I've contributed to, is applying sort of more standard analysis: nothing sticky but other things that sort of mimic stickiness that represent equilibrium. So, that kind of work is, I think, has taken us quite a ways. And it's been in place for a while. You know, Lucas's ideas, which just seemed so dramatically different from the way we've been brought up. Became fully integrated. And now, our scene is, 'Well, that's sort of the old-fashioned way of thinking. Rational expectations, that's old-fashioned.' The idea is sweeping now that we should model expectations as beliefs that can be different across people, heterogeneous beliefs. And people at Chicago are just as enthusiastic as anybody about pursuing that kind of idea. There's--the idea of schools of macro has pretty much disappeared. Especially in the younger generation. It would be impossible find, you know, an economist in a top university today who could be reasonably identified to be--well, especially a monetarist. But even a Keynesian in the old-fashioned sense. We all build models in which if the government purchases more, there's higher GDP. It just has to be. And no one would disagree about that today. So if that's the--if that's the question about what Keynesian means, then Keynes is gone completely. We all agree that an important determinant of total activity is how much the government chooses to buy.

Russ Roberts: Well, I was going to end there. But I can't. I can't. Because I have to challenge that statement. Certainly there are many, many illustrious economists, far more illustrious than I am. Two that come to mind are Valerie Ramey and Robert Barro who have suggested that that relationship is not very reliable or doesn't hold at all. Or it could be negative.

Robert Hall: No, no, no, no, no.

Russ Roberts: Do you think that's true?

Robert Hall: No, no. You are talking about--first of all, Valerie was a student of mine. And I follow her work very carefully.

Russ Roberts: Doesn't she say that government spending has little effect on--that the multiplier is quite small?

Robert Hall: Okay. So, her most recent paper says--well, first of all you have to decide whether you are [?] or at a lower bound or not. And if you aren't, then a billion dollars of government purchases adds about a billion dollars to GDP. So, the so-called multiplier is 1. And I think that's within the consensus range. It's certainly within my beliefs.

Russ Roberts: That doesn't stimulate anything. That's not offset--that's not worrying whether it's offset by monetary policy?

Robert Hall: No. That's assuming that it is offset. Nonetheless, there's an effect. By historical pattern. This is poring over historical data.

Russ Roberts: Correct.

Robert Hall: And some old examples of rapid change of government purchases are the main source of the information. So, since we haven't had much of it in the last 20 years, this is mostly evidence from more than 20 years ago. But it's the evidence we have. And, you know, studies from other countries have been supportive. There's a big literature on this topic. A multiplier of 0, which would mean no effect, is outside the range of almost all empirical work and almost all modeling. It would not be a viable position. And I know Valerie's work; and Barro's paper, QJE (Quarterly Journal of Economics) paper, gets positive multipliers. Not large, but positive. And there's pretty much agreement now that a 0-lower-bound where monetary policy's hands are tied, that multiplier is higher. And that's what's Valerie's most recent work shows. So I think it all fits together. But there's nobody, nobody who says that it makes no difference at all if the government buys more. And that's a substantive thing that's happening in the economy. Of course it's going to influence the economy.

Russ Roberts: Well, we started this conversation by saying that the increase in Federal government spending, which went through the states, didn't have much of an impact because they didn't spend it.

Robert Hall: No, no, no. We have no clue how much impact it had because it didn't happen. So, you have to--you learn nothing about the multiplier if nothing happens to government purchases. Which is exactly what happened over the last decade.