The Fed versus the Banks: Who Will Blink First?
By Robert Higgs • Saturday December 27, 2008 2:25 PM PDT • 25 Comments
During recent months, the Fed has flooded the banking system with reserves, which the banks have chosen to accumulate as (legally) excess reserves, rather than using the funds to add to the volume of their outstanding loans and investments. The Fed’s recently adopted policy of paying a small rate of interest on bank reserves accounts for some of this accumulation, but the amount is so gigantic that it seems much more likely that the banks have greatly increased their assessment of the risk involved in lending and investing as usual and therefore have chosen the lower yielding but less risky alternative of accumulating more and more reserves. Since August, the amount of excess reserves has risen from $2 billion to $559 billion. A graph of this astonishing development shows an abrupt transition from a virtually horizontal line (approximately zero excess reserves for decades) to a virtually vertical line (a quick jump of $557 billion in three months).
So far, this explosive increase of reserves has had only a small effect on the growth of the money stock as measured by the conventional monetary aggregates, such as M2, although the rate of growth of the monetary aggregates is beginning to increase substantially, as shown in this graph.
Because the public’s demand for cash balances has also risen, the recent increases in the money stock have not given rise to increased prices in general. In fact, the major price indexes have fallen slightly in recent months, although the bulk of this decline has occurred because of the decline in the price of oil and related products since July. Price indexes for goods other than energy have declined very little – certainly not enough to justify the many expressions of fear of impending deflation (setting aside whether an actual deflation ought to be feared or not).
At matters now stand, by far the greater threat is rapid inflation, notwithstanding the ongoing recession. When the banks begin to feel more comfortable with expanding the volume of their conventional loans and investments, they will have more than $550 billion on hand to employ for that purpose. The multiplied effect of such a vast amount of lending, as newly created deposits make their way through the fractional-reserve banking system, portends a gargantuan increase in the money stock and hence a correspondingly enormous jump in the general price level. As the public responds to the acceleration of inflation by reducing its demand for cash balances, the increased velocity of monetary circulation will contribute to even more rapid price inflation.
So much potential new money is now impounded in the commercial banks’ holdings of excess reserves that it is difficult to see how the Fed will be able to stem the flood once the banks begin to transform those excess reserves into normal loans and investments. If the Fed attempts to sell enough government securities to soak up the growing money stock, it will drive down the prices of Treasury bonds and hence drive up their yield, increasing the government’s cost of borrowing to finance the huge budget deficits the government will be running because of its various bailout commitments and so-called stimulus programs. This scenario holds the potential for a complete monetary crackup.
I have never been inclined toward touting doomsday financial scenarios. I raise the possibility now only because, as I consider the situation portrayed in the graph of excess reserves linked above, I am unable to foresee how the Fed and the Treasury can navigate through these treacherous waters – waters that their own previous actions have whipped to a foam – without creating terrible financial and economic harm. If the dollar survives the ministrations of Bernanke, Paulson, Bush, and the Obama gang, its survival will be something of a miracle.
I notice that excess reserves went above a billion in 1933 and stayed there until the end of the war. (They did not get back to that level again until 1999). Is that consistent with you thesis of the Great Deferral. The fed was pumping money, but the banks were unwilling to lend, due to the uncertainty caused by the Roosevelt regime.
Ron McKenzie | Dec 28, 2008 | Reply
And so gold rises, as the sun reaches towards the Statue of Liberty, and dawn brings a new day that the Fed can screw with!
Kelly Lieberman | Dec 28, 2008 | Reply
Perhaps someone can explain this to me; Mr. Higgs lists as his first hyperlink above http://research.stlouisfed.org/fred2/series/EXCRESNS/viewdata which shows the monthly figures for excess reserves since 1929. This page shows (as an example) for 1929-01-01 an excess amount of 0.054 billion. One would normally assume that this means that as of the first day of January, 1929, the excess then existing was 0.054 billion. If this is the case, then how does one explain that on 2001-08-01 the excess was 1.203 billion, but by 2001-09-01 the excess had jumped to 19.015 billion? By 2001-10-01 it had returned to 1.326. Was there, in fact a fifteen fold increase during the month of August 2001, followed by a return to normal in October?
Jason Calley | Dec 28, 2008 | Reply
One additional wrinkle — as Paul Kasriel of the Northern Trust has pointed out, the Fed’s holding of Treasuries has slipped to roughly 20% of its assets. Therefore, the Fed does not have enough Treasuries to sell to sop up the excess reserves. This problem is why the Fed floated the idea a month ago of issuing its own debt.
Peter Collins | Dec 28, 2008 | Reply
I sent an email to the author inquiring about that same thing. I must ask: Does looking at that wildly anomalous figure in the graph, appearing as it does with a report date a mere TEN DAYS PRIOR to September 11th, give you a pronounced chill? It does to me.
jasper sneed | Dec 28, 2008 | Reply
Mr. Calley and Mr. Sneed,
The amount of excess reserves given for September 1, 2001, is so far out of line with the preceding and succeeding amounts that I am wondering whether it might be a typo. Perhaps the decimal point should be moved one place to the left? I am only guessing, but in my experience of working with data bases over the years, I have seen such an error more than once.
Robert Higgs
Robert Higgs | Dec 28, 2008 | Reply
I was under the impression the Fed pumped in a ton of reserves after 9/11, and then sucked them out again. The reported data could be monthly averages? Note the 12/108 figure is not available.
Grant | Dec 28, 2008 | Reply
I don’t know the degree to which simple arithmetic applies to a situation like this but it appears that the Fed has nearly doubled the size of its reserves and to me that would imply that within the next two years this will result in at least a doubling of the money supply and hence consumer prices.
Not exactly Weimar territory, but certainly catastrophic. And will the central bankers admit they made a mistake, or will they call for even more spending? Then things could get very ugly.
Joe S. | Dec 28, 2008 | Reply
Yes, I had wondered the same thing — have been hoping, actually, the explanation could be a simple one like a typo. Or perhaps that the 1st of the month dates shown alongside each entry are not necessarily precise — meaning the bizarre jump to 19 Billion might simply be in response TO the events of 9/11, rather than a prescient manoeuver anticipating it.
And yet, I am finding the same anomaly appearing in other historical reportings of the FRB, and in different configurations, such as this one:
http://www.federalreserve.gov/releases/h3/hist/h3hist2.htm
Also, in addition to the same 19 Billion appearing in the Excess Reserves column, there is also a sudden, near quadrupling of the Reserve Balance with Fractional Reserve Banks for that very same month. A decade earlier that figure had been trending in the 20 to 30 Billion range, but had gradually fallen to below 10 Billion by the year 2000, at which single digit level it more or less stayed right up until Sept 2001, when the report shows it jumped back up to about 25 Billion. It’s almost as though a cushion was very deliberately — and very suddenly — installed to accommodate potential bank runs. Whether this was done before or after 9/11 is, it seems to me, as profoundly significant as the true date, and true amount, of the reported 19 Billion anomaly.
jasper sneed | Dec 28, 2008 | Reply
Mr. Sneed,
ALL the EXCRESNS data points fall on the 1st of the month, but roughly 2/7th of those dates were weekend days. The day-of-month field in this report is probably not significant data. The second table you cited gives months only, no precise dates.
I too recall extraordinary interventions that month, so it would be wrong to discard it as a typo. The story here is dramatic and profound even as a response, the fact that the present numbers dwarf that little blip is testament to gravity of the present crisis.
Greg Jaxon | Dec 28, 2008 | Reply
A second aspect of the Sept 2001 swing in excess reserves is that it demonstrates how volatile the numbers on the Fed’s books can become in a crisis. A forecast of hyper-inflation cannot rest solely on so skittish a data series, although as Collins (citing Kasriel) noted above, this time could be different. Bernanke’s thesis – that “anchoring” of (price) inflation expectations is the crucial factor – will get a real test over the next year. To me, Bernanke’s ideas are a lesson in “The Anatomy of Illusions”. The watch we think has been smashed will emerge unscathed on Goldman Sachs’ wrist. Only later will we notice that our own watch was the one smashed.
Greg Jaxon | Dec 29, 2008 | Reply
To speak to the jump on 9/1/01...
The dates on FRED are not precise. Actually, these are averages for the month, I’m pretty sure. You can get weekly historical data here: http://www.federalreserve.gov/releases/h3/hist/h3hist4.txt
I’m pretty sure that these numbers are actual “snapshots” from the dates recorded, since this data is gathered weekly.
The weekly data shows what we expect: the first week of September reserves were “as usual”. Second and Third week showed about a doubling in total reserves (which means that excess reserves spiked to the $30billion range). Fourth week – normal.
If you average the weekly data, you get the monthly data reported on FRED (or at least something very close to it). Also, the monthly data reported on FRED matches that reported at federalreserve.gov, which doesn’t report a day of the month – suggesting that it might be an average for the month.
The Fed really needs to document their data better...
Lucas M. Engelhardt | Dec 29, 2008 | Reply
I’m just a layman here trying to understand this, so I have two questions:
1. How does the situation now with excess reserves in America compare with the situation in Japan with excess reserves in the late 1990s to early 2000s. Clearly, there were different variables, but in Japan excess reserves did *not* lead to inflation, did it? Or perhaps the yen was made artificially weak (an inflation of sorts)?
2. Are these *true* excess reserves? If one were to actually discount bad or false assets that are on the books of these banks, would there still be excess reserves? Is this a case where in a perfect central banker’s world, they would actually be able to match all deflation with an exact dose of inflation, there by keeping the system stable somehow?
Thank you for any answers.
Matt Dioguardi | Dec 29, 2008 | Reply
The Federal Reserve Bank can finance its assets not only through bank notes and demand deposits, but also with term deposits or other term borrowings. This is exactly what they have been considering doing.
The reason for considering this is because the asset quality and liquidity is going down. By this means, the Fed can repay as much of its demand debt as bank-note holders and demand depositors demand, while still holding the lower quality and less liquid assets and not running too low on highly liquid assets.
I think it is more likely that the govt Treasury will create a vehicle to hold these assets, and the Fed will go back to holding only high quality liquid assets.
Higgs is obviously assuming what he is trying to prove: that the quantity of money drives the purchasing power of money.
David Hillary | Dec 29, 2008 | Reply
David Hillary says, “Higgs is obviously assuming what he is trying to prove: that the quantity of money drives the purchasing power of money.”
I definitely don’t want to get involved in a conversation where I don’t know what I am talking about, but here I go. My only goal here is not to prove anything, but to understand what others’ are speaking of. The best way I see to do this is to make some claims myself and hope people show why they are false.
The assumption that quantify of money drives the purchasing power of money seems to have a lot of truth to this. We can establish this quickly with a thought experiment. Imagine everyone woke up tomorrow having twice as much money as they did the day before. Clearly this would devalue the currency. So it seems to me that the assumption that the quantity of money drives its purchasing power is a good one.
Of course, there might be other things that drive the purchasing power, but we need not get into that. Clearly the expansion of the money supply via bank lending has slowed. Right? Is that a controversial statement?
So there is deflation. However, the fed is expanding the money supply via depositing money in commercial banks (via the printing press). However, the banks aren’t lending out this money, instead they are hanging on to it.
I’m guessing that in a central banker’s utopian conception of the world, they could always create just enough new money via the printing press to offset money being lost to the money supply as a result of poor lending practices.
Robert Higgs initially claimed that this might create hyper-inflation. I certainly can see his point, and the numbers or quite staggering. However, I was only claiming that instead of this, we might just see banks clinging to excess reserves and refusing to lend. This was basically the situation in Japan. You had years where the economy didn’t grow and the banks were called zombie banks. Even up until very recently Japan’s economy has been in the doldrums.
So what will it be in America? I guess it depends just how intent Bernanke is on mad-doctoring the economy.
Matt Dioguardi | Jan 1, 2009 | Reply
I thought this was due to 9/11? I remember reading that Greenspan had injected extra fund at the anticipation of Y2K and likewise after 9/11.
Bill Prebble | Jan 6, 2009 | Reply
Prof. Higgs is not trying to prove that. It is already proven.See many sources, but George Reisman’s Capitalism comes to mind.
Bill Prebble | Jan 6, 2009 | Reply