July 30, 2010
Some observations regarding interest on reserves
One of the livelier discussions following Federal Reserve Chairman Ben Bernanke's testimony to Congress on monetary policy has revolved around the issue of the payment of interest on bank reserves. Here, for what it's worth, are a few reactions to questions raised by that discussion:
Is interest paid on reserves (IOR) a free lunch?
"… in September of 2008, the Fed decides to pay interest on reserves—including Excess Reserves. The banks can now make 25 times what they pay in interest, risk-free, just by holding onto money. The Fed is, essentially, leaving $100 bills on the sidewalk."
I'm not sure exactly where the "25 times" comes from, but it seems to me that the most obvious transaction would be to borrow in the overnight interbank lending market—the federal funds market—and then "lend" those funds to the Fed by placing them in the Fed's deposit facility. The differential between the return on those options is a good deal lower than a multiple of 25.
In fact, as many have noted before, the puzzle is why the gap between the funds rate and the deposit rate exists at all. As explained on the New York Fed's FAQ sheet:
"With the payment of interest on excess balances, market participants will have little incentive for arranging federal funds transactions at rates below the rate paid on excess. By helping set a floor on market rates in this way, payment of interest on excess balances will enhance the Desk's ability to keep the federal funds rate around the target for the federal funds rate."
It didn't quite work out that way, so clearly there is a limit to arbitrage. But if you really think that an 8 basis point spread between the effective funds rate and the deposit rate is a problem, the best approach would be, in my opinion, to address the institutional arrangements that are limiting arbitrage in the funds market. (Some of those features are discussed here and here.)
What is the opportunity cost of not lending?
That said, certainly the real issue about the IOR policy concerns the presumed incentive for banks to sit on excess reserves rather than putting those reserves into use by creating loans. This, from Bruce Bartlett, is fairly representative of the view that IOR is, at least in part, to blame for the slow pace of credit expansion in the United States:
"… As I pointed out in my column last week, banks have more than $1 trillion of excess reserves—money that the Fed has created that banks could lend immediately but are just sitting on. It's the economic equivalent of stuffing cash under one's mattress.
"Economists are divided on why banks are not lending, but increasingly are focusing on a Fed policy of paying interest on reserves—a policy that began, interestingly enough, on October 9, 2008, at almost exactly the moment when the financial crisis became acute."
OK, but the spread that really matters in the bank lending decision is surely the difference between the return on depositing excess reserves with the Fed versus the return on making loans. In fairness, it does appear that this spread dropped when the IOR was raised from its implicit prior setting of zero…
… but it's also pretty clear that this development largely reflects a general fall in market yields post-October 2008 as much is it does the increase in IOR rate:
And here's another thought: As of now, the IOR policy applies to all reserves, required or excess. Consider the textbook example of a bank that creates a loan. In the simple example, a bank creates a loan asset on its book by creating a checking account for a customer, which is the corresponding liability. It needs reserves to absorb this new liability, of course, so the process of creating a loan converts excess reserves into required reserves. But if the Fed pays the same rate on both required and excess reserves, the bank will have lost nothing in terms of what it collects from the Fed for its reserve deposits. In this simple case, the IOR plays no role in determining the opportunity cost of extending credit.
Of course, the funds created in making a loan may leave the originating bank. Though reserves don't leave the banking system as a whole, they may certainly flow away from an individual institution. So things may not be as nice and neat as my simple example. But at worst, that just brings the question back to the original point: Is the 25 basis point return paid by the central bank creating a significant incentive for banks to sit on reserves rather than lend them out to consumers or businesses? At least some observers are skeptical:
"Barclays Capital's Joseph Abate…noted much of the money that constitutes this giant pile of reserves is 'precautionary liquidity.' If banks didn't get interest from the Fed they would shift those funds into short-term, low-risk markets such as the repo, Treasury bill and agency discount note markets, where the funds are readily accessible in case of need. Put another way, Abate doesn't see this money getting tied up in bank loans or the other activities that would help increase credit, in turn boosting overall economic momentum."
Are there good reasons for paying interest on reserves?
Even if we concede that there is some gain from eliminating or cutting the IOR rate, what of the costs? Tim Duy, quoting the Wall Street Journal, makes note (as does Steve Williamson) of the following comment from the Chairman:
"… Lowering the interest rate it pays on excess reserve—now at 0.25%—could create trouble in money markets, he said.
" 'The rationale for not going all the way to zero has been that we want the short-term money markets, like the federal funds market, to continue to function in a reasonable way,' he said.
" 'Because if rates go to zero, there will be no incentive for buying and selling federal funds—overnight money in the banking system—and if that market shuts down … it'll be more difficult to manage short-term interest rates when the Federal Reserve begins to tighten policy at some point in the future.' "
Professor Duy interprets this as aversion to the possibility that "the failure to meet expectations would be the real cost to the Federal Reserve," but I would have taken the words for exactly what they seem to say—that the skills and infrastructure required to maintain a functioning federal funds rate might atrophy if cutting the rate to zero brings activity in the market to a trickle. And that observation is relevant because of the following, from the minutes of the April 27–28 meeting of the Federal Open Market Committee:
"Meeting participants agreed broadly on key objectives of a longer-run strategy for asset sales and redemptions. The strategy should be consistent with the achievement of the Committee's objectives of maximum employment and price stability. In addition, the strategy should normalize the size and composition of the balance sheet over time. Reducing the size of the balance sheet would decrease the associated reserve balances to amounts consistent with more normal operations of money markets and monetary policy."
"Normal" may not mean the exact status quo ante, but to the extent that federal funds targeting is a desirable part of the picture, it sure will be helpful if a federal funds market exists.
Even if you don't buy that argument—and the point is debatable—it is useful to recall that the IOR policy has long been promoted on efficiency grounds. There is this argument for example, from a New York Fed article published just as the IOR policy was introduced:
"… reserve balances are used to make interbank payments; thus, they serve as the final form of settlement for a vast array of transactions. The quantity of reserves needed for payment purposes typically far exceeds the quantity consistent with the central bank's desired interest rate. As a result, central banks must perform a balancing act, drastically increasing the supply of reserves during the day for payment purposes through the provision of daylight reserves (also called daylight credit) and then shrinking the supply back at the end of the day to be consistent with the desired market interest rate.
"… it is important to understand the tension between the daylight and overnight need for reserves and the potential problems that may arise. One concern is that central banks typically provide daylight reserves by lending directly to banks, which may expose the central bank to substantial credit risk. Such lending may also generate moral hazard problems and exacerbate the too-big-to-fail problem, whereby regulators would be reluctant to close a financially troubled bank."
Put more simply, one broad justification for an IOR policy is precisely that it induces banks to hold quantities of excess reserves that are large enough to mitigate the need for central banks to extend the credit necessary to keep the payments system running efficiently. And, of course, mitigating those needs also means mitigating the attendant risks.
That is not to say that these risks or efficiency costs unambiguously dominate other considerations—for a much deeper discussion I refer you to a recent piece by Tom Sargent. But they should not be lost in the conversation.
By Dave Altig, senior vice president and research director at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference Some observations regarding interest on reserves:
July 28, 2010
The money-inflation connection: It's baaaack!
Our St. Louis Fed colleague David Andolfatto declares it is time to bury the old saw that says when it comes to inflation, follow the money:
"One of the ideas that stuck in my head as an undergrad was the proposition that 'inflation is always an[d] everywhere a monetary phenomenon.' The idea is usually formalized by way of the Quantity Theory of Money (QTM)—or more precisely—the Quantity of Money Theory of the Price-Level. (QTM is not a theory of money, it is a theory of the price-level).
"In its simplest version, the QTM asserts that the equilibrium price-level is roughly proportional to the outstanding supply of money (however defined). As inflation is the rate of change in the price-level, the phenomenon of inflation is attributed primarily to excessive growth in the money supply (typically viewed as being controlled by the monetary or fiscal authority)."
Andolfatto goes on to note that the monetary base—the sum of currency in circulation and the banks' reserve balances held at the Federal Reserve (at that page, search "reserves")—more than doubled since fall 2008, while the rate of inflation fell.
That's certainly true, though most versions of the quantity theory applied to monetary policy discussions lean on broader measures of money—for no better reason than those measures help the theory fit the facts. Specifically, since the 1980s the phrase "inflation is everywhere and always a monetary phenomenon" has in effect meant "inflation is everywhere and always a monetary phenomenon when we measure money by M2."
And here's an interesting thing. If you look at the relationship between M2 growth and core inflation over the past decade and a half, it appears that the money-inflation nexus has been gaining in strength:
Another way to see this relationship is to look at the correlation between M2 growth and core inflation over rolling 10-year windows:
Could it be that the death of the quantity theory has been greatly exaggerated?
There are plenty of reasons to be cautious. For one thing, it is oft-noted that any connection between money and inflation could be purely coincidental. In fact, if you stare hard at the picture it does appear that changes in inflation often precede changes in money growth. One interpretation is that the same factors that push trend inflation around also result in responses by policymakers or private market participants that ultimately cause the money supply to move in a sympathetic direction.
But even if causation does run from money to prices, the case is not quite solved. The monetary base measure that the Andolfatto post emphasizes has a lot to recommend itself, not least being that it is the measure of money that central banks actually control. The stark disconnect between the growth in currency and bank reserves (the quantity of which is determined by the Fed) and M2 growth (the quantity of which is determined by the decisions of banks to expand their balance sheets) raises legitimate questions about how policymakers would exploit an M2-inflation connection in an environment when the monetary base–M2 connection—the so-called "money multiplier"—has changed so dramatically.
There could be lots of answers to that question. The relatively new Federal Reserve policy of paying interest on bank reserves is one possibility. Andolfatto's suggestion that all changes in money are not created equal might contain the germ of another explanation. For our part, we think the question is quite a bit more than academic.
TrackBack URL for this entry:
Listed below are links to blogs that reference The money-inflation connection: It's baaaack!:
July 21, 2010
Gauging the inflation expectations of business
Last Friday, the U.S. Bureau of Labor Statistics (BLS) reported that the consumer price index (CPI) declined in June for the third consecutive month. And although core inflation edged up a bit, the entire increase can be accounted for by the BLS's seasonal adjustment factor. In an environment of "business-not-as-usual" like today, data driven by seasonal adjustment are certainly suspect. So overall, the June CPI news seems largely in line with the downward inflation trend we've been seeing for a while.
Does recent disinflation imply deflation? Well, that wouldn't be the consensus coming out of the June 22–23, 2010, FOMC meeting minutes:
"A broad set of indicators suggested that underlying inflation remained subdued and was, on net, trending lower,… However, inflation expectations were seen by most participants as well anchored, which would tend to curb any tendency for actual inflation to decline."
A similar sentiment was expressed recently by European Central Bank (ECB) President Jean-Claude Trichet in describing the ECB's view on inflation expectations:
"Inflation expectations remain firmly anchored in line with our aim of keeping inflation rates below, but close to, 2% over the medium term."
Of course, how firmly something is anchored has meaning only relative to the forces working to move that anchor. Being well anchored against a five-knot drift isn't exactly the same as being well anchored against a 10-knot current. But assuming the idea here is that expectations are likely to hold against the usual range of events one might expect in an environment like ours, we can ask the question: How does one judge whether expectations are well anchored?
Presuming this analogy, one way we might gauge how anchored inflation expectations are is to monitor the behavior of inflation expectations relative to recent shocks. By this standard, expectations seem rock-solid. Virtually every measure of inflation expectations has held steady against the tug of widely fluctuating commodity prices, persistent retail disinflation, expansion of the central bank's balance sheet, large current and projected fiscal imbalances, and the general economic and financial volatility of the past few years.
But economists know very little about how expectations are formed and, therefore, we don't know what sorts of events are likely to pose the greatest threats to the expectations' anchor. In other words, we may not know when inflation expectations are likely to move until, well, they actually move.
In an attempt to get a more direct read of inflationary sentiment and to put more light on how inflation expectations are formed, the Federal Reserve Bank of Atlanta is looking into polling businesses about their inflation expectations. With help from the folks at Kennesaw State University (a very big hat-tip to Don Sabbarese and Dimitri Dodonova, who compile the Georgia and Southeast Purchasing Managers' Indexes) we asked a group of purchasing managers a handful of questions related to the inflation outlook. The poll was conducted during the week of July 7–July 13, and 32 respondents answered the call. Here's what we learned.
Over the next 12 months, this sample of purchasing managers expects unit costs to increase 1.7 percent, just a shade higher than the consensus CPI forecast of economists. The distribution of the poll responses is represented by the red bars in the chart below. About half of the respondents saw unit costs rising "somewhat" defined by the range of 2 percent to 4 percent, while about one-third of the respondents indicated they expect virtually no change in unit costs over the period.
But what probability do respondents attach to their expectations? It turns out that some respondents have great confidence in their expectation for unit cost changes—they assigned little chance that unit labor costs would do anything other than what they forecast. But most purchasing managers attached a significant likelihood to a large range of possible outcomes. We show the distribution of the average respondents' expectation for unit costs by the blue bars in the chart. So, keeping in mind that the mean expectation of the group was for unit costs to rise 1.7 percent, respondents on average assigned a 17 percent chance that unit costs could decline over the coming year, while they put an equally large likelihood of inflation at 5 percent or more (20 percent).
What does all this mean for the inflation outlook? Well, first, let us caution that a sample of this size doesn't lend itself to any strong conclusions, and these data will have to be carefully evaluated in light of other poll questions and against other benchmarks. Those important caveats aside, we can say that while the average purchasing manager in our poll is expecting price pressures that pretty closely correspond to the Federal Reserve's long-term inflation projection, this group attaches significant upside and downside risks to the inflation outlook.
Have any thoughts about how we proceed from here? We'd love to hear your ideas. The next poll will be sent to potential respondents in about three weeks.
By Mike Bryan, vice president and senior economist, and Laurel Graefe, senior economic research analyst, both at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference Gauging the inflation expectations of business:
July 16, 2010
A curious unemployment picture gets more curious
UPDATE: One of our eagle-eyed macroblog readers thought something was fishy-looking in the second chart of yesterday's (July 15) post. He was right—the chart was in error. This post is an updated, edited version with the erroneous chart replaced. There have also been some text revisions to better reflect the revised chart. The new text is bolded in this post.
At first blush, the second quarter statistics from the Job Openings and Labor Turnover Survey (commonly referred to as JOLTS and released Tuesday by the U.S. Bureau of Labor Statistics) suggest little has changed recently in U.S. labor markets:
"There were 3.2 million job openings on the last business day of May 2010, the U.S. Bureau of Labor Statistics reported today. The job openings rate was little changed over the month at 2.4 percent. The hires rate (3.4 percent) was little changed and the separations rate (3.1 percent) was unchanged."
Despite a slight step backward in May, the overall trend in job openings has been positive—Calculated Risk has the picture—but in a sense this fact has just deepened the puzzle of why the unemployment rate is so darn high. As we wrote in the first quarter issue of the Atlanta Fed's EconSouth:
"The disconnect between the supply of and demand for workers that is reflected in statistics such as the unemployment rate, the hiring rate, and the layoff rate can be dynamically expressed by the Beveridge curve. Named after British economist William Beveridge, the curve is a graphical representation of the relationship between unemployment (from the BLS's household survey) and job vacancies, reflected here through the JOLTS."
Since the second quarter of last year, the unemployment rate has far exceeded the level that would be predicted by the average correlation between unemployment and job vacancies over the past decade. Tuesday's report indicates that the anomaly only deepened in the first two months of the second quarter.
The dashed line in the chart above, which is estimated from the data from 2000–08, represents the predicted relationship between the number of unemployed persons in the United States and the number of job openings. That simple relationship would suggest that, given the average number of job openings in April and May, the unemployed would be expected to number about 10.4 million—not the nearly 15 million we actually saw.
Some analysts have suggested the unemployment benefits policies of the last couple of years may be responsible for abnormally high unemployment rates. Estimates generated by several researchers in the Federal Reserve—here and here, for example—suggest that extended unemployment benefits may have increased the unemployment rate by somewhere between 0.4 and 1.7 percentage points. But even if we accept those numbers and adjust the Beveridge curve by assuming that the number of unemployed would be correspondingly lower without the benefits policy, it's not clear that the puzzle is resolved:
If you tend to believe the higher end of the benefits-bias estimates, no puzzle emerges until the second quarter of 2010. And, of course, some estimates apparently deliver an even larger impact of the extended benefits policy. Let's call the question unsettled at this point.
The most tempting explanation for the seeming shift in the Beveridge curve relationship (to me, anyway) is a problem with the mismatch between skills required in the jobs that are available and skills possessed by the pool of workers available to take those jobs. The problem with this tempting explanation is that it is not so clear that the usual sort of structural shifts we might point to—for example, only nursing jobs being available to laid-off construction workers—are so obviously an explanation (an issue we explored in a previous macroblog post).
But these sorts of subplots may miss the truly big part of the story. I have noticed a recent spate of articles repeating a theme we hear anecdotally from many sources, in many industries. For example, this from a June USA Today article…
"…the [auto] industry is poised to add up to 15,000 this year and could need up to 100,000 new workers a year from 2011 through 2013.
"…Automakers need workers with more and different skills than in the past on the factory floor.… Among priorities: computer skills and the ability to work with less supervision than their predecessors. That likely means education beyond high school."
… or more recently, this one from the New York Times:
"Factory owners have been adding jobs slowly but steadily since the beginning of the year, giving a lift to the fragile economic recovery…
"Yet some of these employers complain that they cannot fill their openings.
"Plenty of people are applying for the jobs. The problem, the companies say, is a mismatch between the kind of skilled workers needed and the ranks of the unemployed."
Now I realize that a few anecdotes don't make facts, but I have been in more than a few conversations with businesspeople who have claimed that the productivity gains realized in the United States throughout the recession and early recovery reflect upgrades in business processes—bundled with a necessary upgrade in the skill set of the workers who will implement those processes. This dynamic suggests that the shift in required skills has been concentrated within individual industries and businesses, not across sectors or geographic areas that would be captured by our most straightforward measures of structural change.
The data necessary to test this proposition are not easy to come by. That challenge is unfortunate, because the return on figuring out what is beneath those Beveridge curve graphs is very high.
By Dave Altig, senior vice president and research director at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference A curious unemployment picture gets more curious:
July 09, 2010
How close to deflation are we? Perhaps just a little closer than you thought
Since last October, the consumer price index (CPI) has gone up an annualized 0.7 percent. On an ex-food and energy basis, the number is a little lower, at 0.5 percent. And the Cleveland Fed's trimmed-mean and median CPIs, at 0.7 percent and 0.2 percent, respectively, also put the recent trend in consumer prices in pretty low territory.
And this is before we take into account any potential mismeasurement, or "bias," in the construction of the CPI.
How big is the CPI's bias? Well, in 1996, the Social Security Administration commissioned a study on the accuracy of the CPI as a measure of the cost of living. This so-called "Boskin Commission Report" said the CPI was overstated by about 1.1 percentage points per year. The commission identified several sources of potential bias, but about half of the 1.1 percentage points resulted from new products and quality changes that were slow or otherwise imperfectly introduced into the price statistic.
Since that time, the Bureau of Labor Statistics has initiated a number of methodological changes that have reduced the CPI's mismeasurement. In a 2001 paper, Federal Reserve Board economists David Lebow and Jeremy Rudd put the CPI bias at only about 0.6 percentage points. And again, of this amount, the big share of the bias (about 0.4 percentage points) resulted from the imperfect accounting of new and improved goods.
Now, in an article (available to all in its working paper version) appearing in the latest issue of the American Economic Review, Christian Broda and David Weinstein say the earlier estimates of the new goods/quality bias may be a bit understated. The authors examine prices from the AC Nielsen Homescan database and conclude that between 1996 and 2003, new and improved goods biased the CPI, on average, by about 0.8 percentage points per year. If this estimate is accurate, consumer price increases since last October would actually be around zero, or even slightly negative, once we account for the mismeasurement of the CPI caused by new and improved goods.
But (oh, you just knew there was going to be a "but" in here, right?) the authors also point out that, because new goods are introduced procyclically, this bias tends to be larger during expansions and smaller during recessions. In other words, given the severity of the recession and the modest pace of the recovery, there may not be a whole lot of innovation going on right now in consumer goods. This is a bad thing for consumers, of course, but it would be a good thing for the accuracy of the CPI.
By Mike Bryan, vice president and senior economist at the Atlanta Fed
TrackBack URL for this entry:
Listed below are links to blogs that reference How close to deflation are we? Perhaps just a little closer than you thought: