Investment and Unemployment over Longer Periods of Time

• Like Paul Krugman, Justin Wolfers also wrote yesterday about my blog post of January 14 on the correlation between investment and unemployment. Wolfers argues that the relationship did not exist in earlier years. He is wrong.


• His argument is based on the observation that the scatter of points for the 1990-2010 period, shown in one of my graphs, shifts up and to the right—higher unemployment for a given level of investment—if you include the 1970s and 1980s. The scatter of points shift back down and to the left if you go back further. This shift up in unemployment in the 1970s and 1980s was due in part to the well-documented longer term increase in the natural rate of unemployment in the 1970s and 1980s, which many macroeconomists have researched and written about, but which Wolfers does not mention. When you recognize that such longer-term historical trends exist, you can see that there is a strong correlation between investment and unemployment that goes back before 1990.


• The time-series plot below goes back to 1960. You can see frequent ups and downs in the investment ratio and most of these are related to the downs and ups in the unemployment rate. The exceptions are mainly due to the secular, longer-term rise and fall in the unemployment rate.


• If you go beyond the chart and use some statistics, you can confirm these observations and be more pricise. For example, a simple regression of the change in the unemployment rate on the change in the investment ratio has a large and statistically significant coefficient of -.7 with a t-value of -10 over the 50 period.


• As always in economics there are many factors at work, especially when you consider longer periods. But a longer history confirms the strong correlation I wrote about.

Investment and Unemployment: A Reply

• Paul Krugman wrote a post yesterday afternoon and another one last evening on a January 14 post of mine. In the Janaury post I pointed out the strong correlation between total fixed investment as a share of GDP and the unemployment rate during the past two decades; total fixed investment equals business fixed investment plus residential investment. In his afternoon post, he argued that it’s misleading to look at total fixed investment because most of the recent swing has been in residential investment. “It’s mostly the housing bust!” he argued, continuing that “The rest”—the business fixed investment part—“is just politically motivated mythology.”


• But the correlation I pointed out is not just due to housing. There is a close relationship during the past two decades between business fixed investment and unemployment. In fact it’s closer than for residential investment and unemployment. Here are the time series charts for business fixed investment and residential investment. You can see the ups and downs in business fixed investment as a share of GDP and the corresponding downs and ups in the unemployment rate, which result in the negative correlation; there is one big swing in housing. If you want to compare correlation coefficients, for business fixed investment the correlation is -.84 and for housing it is -.68. These are facts, not mythology, or whatever other name you want to use.


• Curiously, in his post of last evening, Krugman includes a scatter diagram between business fixed investment and unemployment, in which he admits to a strong correlation after all. He then brings attention to six points on the lower right of the diagram (he reverses the axes from the diagram I first used), identifies them with the Obama Administration, and notes that they are off an apparent pattern: unemployment is even higher than the high level that would be expected from the low level of business fixed investment. But housing is now omitted from the argument. Remember the “housing bust!” from the afternoon post? Clearly the housing bust has added to the unemployment rate. It is business fixed investment plus residential investment, as I originally argued.

A Good Exit Strategy Proposed by Philadelphia Fed President Plosser

• Last Friday in New York, Charles Plosser, President of the Philadelphia Fed, proposed an exit strategy for the Fed. It’s the first explicit exit strategy to be put forth by a member of the FOMC, so it deserves careful consideration and discussion.


• Previous statements about exits by Fed officials simply listed the tools that could be used in an exit strategy, but did not actually put forth an exit strategy. In contrast, President Plosser describes a specific strategy.


• Two things are very attractive about the strategy. First, it aims to return monetary policy to one in which the federal funds rate is determined by the supply and demand for reserves. This of course requires that the Fed bring down the enormous supply of reserve balances on its balance sheet to a level closer to a quantity demanded by banks at a positive interest rate. Reserves were $26 billion on September 10, 2008 when the funds rate was 2 percent just before the panic, which gives an order of magnitude of where reserves should go. Plosser assumes $50 billion, which seems reasonable to me. But reserve balances will be around $1,500 billion by the time QEII is over, so it’s a long way down.


• The second attractive feature of the exit strategy is that the path of reduction in reserve balances is tied to future movements of the federal funds rate. It is thus much like an exit rule, or a contingency plan, which both preserves flexibility and creates predictability. The exit rule would reduce reserves by $125 billion for each 25 basis point increase in the funds rate plus another $50 billion at each FOMC meeting. After 10 meetings $1,450 billion would be removed (the contingency plan starts at the second meeting) bringing reserves to $50 billion.


• This chart, drawn from data in his speech, shows how the strategy would work if the Fed increased the federal funds rate by 25 basis points for ten consecutive meetings. Of course an advantage of this strategy is that the pace of reserve drawdown is tied to the funds rate—if the funds rate rises more quickly reserves will come down more quickly.


• The strategy is very close to one I recommended a year ago in testimony, "An Exit Rule for Monetary Policy," prepared for a Congressional hearing at the House Financial Services Committee. I suggested $100 billion per 25 basis points with no constant amount, but that was before QEII when reserve balances were much lower. I argued that such a strategy would reduce risks in the markets. With increased predictability about policy, banks could better manage their own balance sheets and the price discovery process would be much smoother as funds traders and other market participants could better anticipate what the Fed would do, a view which was supported by Peter Fisher (head of the New York Fed trading desk in the 1990s), who I consulted about the idea at the time.


• Of course the New York Fed trading desk should have some discretion about how to execute the FOMC’s directive if such a strategy is adopted. I hope it is.

Misunderstanding Prescriptive Versus Descriptive Monetary Policy Rules

• Monetary policy rules can be used both for prescriptive and descriptive purposes, but it’s important to be clear about which purpose one has in mind. A policy rule estimated over a period which included the Great Inflation of the 1970s, for example, might be a good description of policy during that period, but it would be a terrible prescription for policy today. Similarly, a policy rule estimated over a period which included the 2003-2005 period, when rates are unusually low, would not be a good prescription, in my view, for policy today.


• Misunderstandings about whether a particular rule is meant to be descriptive or prescriptive can thus lead to policy mistakes. Consider this passage from the March 18, 2011 issue of JP Morgan’s Global Data Watch (p 16): “The original Taylor rule was descriptive and meant to match how Fed policy was set in the 1987-1992 period. Subsequently, some researchers found that variants of Taylor rules were optimal in certain economic models, and so could also be prescriptive. However, those classes of models were usually quite restrictive, such as assuming policy was not at the zero bound, or that the central bank only tried to influence the short end of the curve. In fact, very few optimal policy models even incorporate multi-period interest rates."


• But the Taylor rule was not meant to be descriptive as I made clear in my original paper. Rather it was very explicitly meant to be prescriptive. I derived it by experimenting with different types of rules in stochastic simulations of different monetary models, including my multi-country model at Stanford, and by studying the results of other people’s simulations. This pinned down the left-hand side variable and the right-hand side variables, and led to simple functional forms and coefficients.


• The models used to derive this rule were not restrictive in the sense used in the quoted paragraph. My multi-country model has a term structure of interest rates, with long rates affecting aggregate demand. Exchange rates are part of the monetary transmission mechanism through their effect on exports and imports. The zero bound on the interest rate was taken into account in the stochastic simulations. Later, simpler, three-equation, textbook models were used by other researchers to show that the Taylor rule was optimal using formal dynamic optimization methods (reverse engineering). These simpler models helped students understand how policy rules work.


• I referred to the 1987-92 period in my original paper, but the rule was not an estimated rule over that period or any other period. After deriving the rule I noted that it was close to actual policy during in 1987-92 and used that result to work through several case studies of how such a rule would have worked in practice. A pure estimation strategy would not have focused on that period, and even if it did, it would have included more variables on the right hand side (such as money growth or the exchange rate). An example of an estimated rule in existence then was that of Ray Fair of Yale.


• When teaching Economics 1, I stress the difference between positive (descriptive) economics and normative (prescriptive) economics. This discussion of policy rules illustrates the importance of understanding the difference in practice. You can’t justify QE2 by saying that the interest rate is negative with the prescriptive policy rule I proposed, because the implied rate is not negative, it’s close to 1 percent.

Blogging Blocked in Beijing

One of the things I like most about blogging is that I can post from anywhere in the world—Tokyo, Milan, Washington—not just from my home or office at Stanford.

Well not exactly. This past week I could not post from Beijing where I was visiting. I normally use Google’s Blogger platform to post blogs, but when I attempted to post from Beijing I could not get on Blogger. I soon found that I could not even get on my blog, nor on Greg Mankiw’s blog, nor any Blogspot blog. When I asked some students what the problem was, they told me that all Google platforms were blocked in China, and so were Facebook and Twitter platforms. As I later discovered in this Wall Street Journal article, the blockage is evidently part of an effort to prevent Internet traffic related to the so called “Jasmine Revolution.”

So I had to wait until I returned to Stanford to start blogging again, as I did yesterday. But it’s sad to know that students in China who might be interested can’t read, for example, about how the U.S. and the Chinese stimulus packages differed, or about many more important things.

Why the Stimulus Failed to Boost Infrastructure in the US: A Comparison With China

The data are now clear. Despite its large size, the 2009 U.S. stimulus package failed to increase government infrastructure spending or other government purchases as its promoters had claimed it would. The large federal stimulus grants sent to state and local governments for infrastructure spending were mainly used to reduce borrowing and thus did not result in an increase in purchases. Here is a summary of my research with John Cogan. The explanation is that local governments in effect acted as many American households did: when they received the stimulus money, they saved it rather than purchased goods and services. This is what permanent income theory would predict. It is also what previous empirical studies of the 1970s stimulus packages found.

To better understand this explanation one can look at other countries, and in particular at China’s recent stimulus package. This week I went to China and explored the question.

Local governments in China apparently did increase infrastructure spending in 2009 following the stimulus package. Why didn’t these governments simply reduce borrowing as did U.S local governments? Professor Chong-en Bai of Tsinghua University gave me the best answer using simple economic reasoning: the local governments appeared to behave more like liquidity constrained households than permanent income households. In China, local governments do not have much access to capital markets. They get their funding mainly from the central government, including loans from the central bank, and of course only for projects that are approved by the central government. At any point in time local governments are submitting new infrastructure projects for approval; some are being rejected and some are being accepted. If the central government wants to increase infrastructure spending by the local governments all it has to do is lower the acceptance criterion, instruct the central bank to provide the funds, and the volume of projects increases. This is apparently how the stimulus worked in China.

Note that the mechanism is essentially built into the structure of the economy, with characteristics similar to an automatic stabilizer in which the criteria are raised and lowered administratively according to the state of the business cycle. Of course, if the criteria before the stimulus were appropriate, then the projects accepted during the stimulus were of more dubious quality.

This hypothesis has yet to be tested, however. The type of data published by the U.S. Bureau of Economic Analysis—which John Cogan and I used to study the stimulus in the United States—are not publicly available in China, so alternative empirical tests will have to be devised.

Straightjackets and “We Rule” Shirts

David Wessel writes about the "straitjacket" of a one-size-fits-all monetary policy in his Wall Street Journal column this week. He shows why a single interest rate set by the European Central Bank is not necessarily appropriate for all the countries in the Euro zone: it can push some countries off the Taylor rule which can lead to financial excesses and other problems. In the period leading up to the financial crisis, the Euro interest rate was way too low for Ireland, Greece, and Spain. So it’s not surprising they ran into financial difficulties. Germany in contrast was right on the rule. The chart below (a favorite of mine produced in March 2008 by the OECD) illustrates the same point: the larger was the deviation from the Taylor rule, the larger was the housing boom. The chart also illustrates that the single rate would not have been such a problem if it were closer to the Taylor rule during that period: all the other countries were to the right of Germany.

Additional evidence that you win with rules and lose with discretion was provided this week in a video produced by Stanford economics graduate students for the annual skit party. Simply wearing “we rule” shirts transforms losers into winners in the annual little big game with Berkeley.

Evaluating TARP

Today’s TARP hearing at Senate Banking follows a slew of recent reports. The Congressional Oversight Panel (COP) issued its final report yesterday. Economists Simon Johnson, Allan Meltzer, Joe Stiglitz, and Luigi Zingales submitted testimony to COP two weeks ago. The Special Inspector General for TARP (SIGTARP) issued a comprehensive review in January. Three members of COP published an oped in today’s Wall Street Journal.

A common theme is the high cost of the TARP. I‘m not talking about whether the government lost or made money, which is not a good measure of effectiveness, but rather the costs to the economy (stability, growth, employment, etc). Since November 2008 I have been writing about the costs of the chaotic rollout of the TARP which in my view worsened the crisis and exacerbated the panic. (Here is my written testimony for today’s hearing.) In his recent book former FDIC chairman Bill Isaac concluded that “any objective analysis would conclude that the TARP legislation did nothing to stabilize the financial system that could not have been done without it. Moreover, the negative aspects of the TARP legislation far outweighed any possible benefit.” In his recent testimony Joe Stiglitz said that “TARP has not only been a dismal failure…but the way the program was managed has, I believe, contributed to the economy’s problems.”

Of course others are more positive about the stabilizing effect of TARP. Timothy Massad, current acting assistant secretary of Treasury, argues that the TARP prevented a more severe panic, citing as empirical evidence a paper by Alan Blinder and Mark Zandi. However, Blinder and Zandi explain that they don’t do a separate evaluation of the TARP: “We make no attempt to decompose the financial-policy effects into portions attributable to TARP, to the Fed’s quantitative easing policies, etc,” they say, so this is not really empirical evidence. COP is also positive about the short run impact though less positive than the Treasury

Though some disagree about the net costs of TARP in the short run, few disagree that the longer-run costs are substantial. In January the SIGTARP listed these costs:
· “damage to Government credibility that has plagued the program,”
· “failure of programs designed to help Main Street rather than Wall Street,”
· “moral hazard and potentially disastrous consequences associated with the continued existence of financial institutions that are ‘too big to fail’”
The COP final report listed these costs:
· “continuing distortions in the market”
· “public anger toward policymakers,”
· “a lack of full transparency and accountability.”
At the COP hearing, Stiglitz, Meltzer, Johnson, and Zingales (who rarely all agree) were unanimous in their view that the TARP actions have created an incentive for financial institutions and their creditors to take high risks due to the expectation of being bailed out, favoring big players and leaving the economy vulnerable to financial crisis. They also agreed that the Dodd-Frank legislation did not solve “too big to fail.”

To these costs I would add that the TARP established an unfortunate precedent of heavy-handed government intervention in the operations of businesses. The government forced some financial institutions to take TARP funds, even those that said they did not want them, by threatening actions from regulators. The government used the TARP for purposes other than originally stated in Congressional hearings, including the bailing out of automobile companies.

TARP is not popular with most Americans. Economic evaluations now rolling in support their view, but rather than pointing fingers, it is time to absorb the lessons and take actions to remove the legacy costs and try to end government bailouts as we know them.

A Credible First Step Toward a New Budget Strategy?

Last week the Senate voted down the House proposal (HR1) to slow the growth of spending in the current fiscal year 2011. Now alternative plans are being laid to slow the growth of spending in 2011 by roughly the same amount as HR1, possibly through a series of shorter continuing resolutions prorated at $2 billion per week. Simultaneously the House is developing a budget resolution for 2012 with the aim of reducing spending growth and the deficit in later years. Hence, the 2011 budget actions should be viewed as a first step of a longer term budget strategy. This first step is crucial. It establishes the credibility of the whole strategy.

Would a step any smaller than HR1 be credible? The following chart tries to present the basic facts in a simple way. It shows federal outlays from FY2000 to FY2011 as a share of GDP with and without HR1. The data are from CBO.

First note the remarkable explosion in government outlays in the past few years. In 2007 outlays were 19.6 percent of GDP. In 2011 outlays will be 24.7 percent of GDP.

Second, note the very modest effect HR1 would have on total outlays as a share of GDP in 2011: outlays would be 24.6 percent rather than 24.7 percent of GDP. (According to CBO, HR1 would reduce outlays by $19 billion or .12 percent of GDP in 2011).

So HR1 is a small first step, especially compared to the binge of the past two or three years. It is hard to see how a first step could be any smaller and still represent a credible start on a new strategy. Anything much less would not even be visible on the chart. For a budget strategy to be credible it has to start with real actions, not just promises of future actions. An actual step at least this large is essential for restoring credibility and increasing economic growth. Reasonable people can disagree about particular programs, but they should be able to agree that a step of this size is the bare minimum needed to get on with the important task of a budget strategy.

Costs of the New Government Activism

In “Activism,” a paper soon to be published by International Finance and already posted at the Council on Foreign Relations, Alan Greenspan delves into the consequences of the recent surge of what he describes as “government activism, as represented by the 2009 US$814 billion programme of fiscal stimulus, housing and motor vehicle subsidies and innumerable regulatory interventions.” Of course, this recent period of extraordinary government interventions has been commented on before. Gillian Tett in a Financial Times article called it the Ad Hoc Age. I called it the Great Deviation because it represents a major deviation from less interventionist or rules-based policies in the 1980s and 1990s. Noting that the current surge is one turn in a longer term cycle, Amity Shlaes argues in a Bloomberg column that we should speed up the cycle and get back to less intervention sooner.

But Alan Greenspan goes further by concentrating on the task of explaining and empirically estimating the costs of this intervention on the economy. “Much intervention turns out to hobble markets rather than enhancing them,” he explains, adding that “unpredictable discretionary government intervention scrambles the prospective underlying supply–demand balance.” More specifically he argues that government-induced uncertainty causes businesses to increase their liquid investments (bank deposits, government bonds, etc.) at the expense of illiquid investments (business fixed investment in structures and equipment). To test the theory he looks at the ratio of business fixed investment to the flow of internal funds which non-financial corporate businesses have available each quarter. That ratio is at historically low levels now which he attributes in large part to the increased activism.

Crowding Out Investment

Of course it’s hard to find a good quantitative measure of activism. Greenspan focuses on the cyclically adjusted budget deficit as a ratio to GDP. Because this measure does not include the “innumerable regulatory interventions” nor the monetary policy interventions not discussed in the paper, it is a downward biased metric of the recent surge in interventions. Nevertheless, he finds a large and significant negative effect of this variable: a one percentage point increase in the deficit as a share of GDP, due say to Keynesian stimulus programs, reduces the ratio of business fixed investment to internal funds by 3.3 percent after controlling for capacity in the nonfinancial corporate sector. (If you look at the regressions in the paper you will find logarithmic transformations of these ratios but they are very close to linear over the relevant range). This crowding out is one reason why he argues that “The recent pervasive macro-stimulus programs exhibit the practical shortfalls of massive intervention.”

The rest of the decline in investment he attributes to other aspects of the activism, concluding that “a minimum of half the post-crisis shortfall in capital investment, and possibly as much as three quarters, can be explained by the shock of vastly greater government-created uncertainties embedded in the competitive, regulatory and financial environments”

Relevant to the Recent Debate over the House Budget Proposals

The study is also relevant to the discussion over whether the new proposals by the House of Representatives to reduce government spending will ‘crowd in” private investment and thereby stimulate employment. Greenspan’s regression results (eg. Exhibit #5 in the paper) imply that reducing the cyclically-adjusted deficit through reduced spending will crowd in private investment as firms allocate a larger fraction of their cash flow to new investment. The estimated effect on investment is virtually immediate occurring within one quarter. I found it interesting that the size of the investment effect (as a share of GDP) is very close to the model simulations reported by Cogan, Cwik, Taylor and Wieland (see Figure 3 of that paper) using a modern “new Keynesian model.” The positive impacts on investment will be even larger as they are viewed as part of a credible longer term plan to reduce the deficit.

Lessons Learned from Ben Bernanke's Policy Rule Discussion at the Senate

At yesterday's hearing before the Senate Banking Committee, Fed Chairman Ben Bernanke talked about monetary policy rules in response to a series of questions by Senator Pat Toomey. First, the Chairman stated that the Taylor Rule calls for interest rates “way below zero” and that this justifies methods such as quantitative easing. This is puzzling because I have reported for months that the Taylor Rule (see 1993 paper) does not call for an interest rate below zero. Second, when Senator Toomey then asked if Taylor believed the Taylor Rule called for rates below zero, Chairman Bernanke didn’t answer directly, but instead claimed that in 1999 I preferred a different rule to the one I published in 1993; he then said that the 1999 rule gives a much different rate. Senator Toomey then pressed on and specifically said the Taylor Rule called for rates higher than we have now, at which point Chairman Bernanke changed tack and argued that there were other policy rules that call for below-zero interest rates. Here is the relevant part of the transcript.

MR. BERNANKE: ... The Taylor Rule suggests that we should be, in some sense, way below zero in our interest rate, and therefore we need some method other than just normal interest rate changes to --
SEN. TOOMEY: Do you know if Mr. Taylor believes that?
MR. BERNANKE: Well, there are different versions of the Taylor Rule, and there's no particular reason to pick the one that he picked in 1993. In fact, he preferred a different one in 1999 which, if you use that one, gives you a much different answer.
SEN. TOOMEY: My understanding is that his view of his own rule is that it would call for a higher Fed funds rate than what we have now.
MR. BERNANKE: There are, again, many ways of looking at that rule, and I think that ones that look at history, ones that are justified by modeling analysis, many of them suggest that we should be well below zero. And I just would disagree that that's the only way to look at it.
But anyway, so I think there are some -- there is some basis for doing that.

There are several issues raised by this back-and-forth exchange.

Most important, at least from my perspective, is that contrary to what was claimed in the hearing, I did not say that I preferred a different policy rule in 1999 rather than the rule I originally published in 1993. I am not sure where this idea of my preferring another rule came from; I went back and looked at the academic papers I published in 1999 (here is a list); the paper on this list that Chairman Bernanke may have been referring to is A Historical Analysis of Monetary Policy Rules where I looked at two different policy rules during different periods of U.S. history. However, as I said in that paper (page 325), one rule was the “policy rule I suggested” and the other one was what “others have suggested.” The “others” were people at the Federal Reserve so for completeness I included that rule in the historical comparison. I did not propose or prefer an alternative rule in that 1999 paper, and it is hard to see how one could interpret the paper that way. This is not just a matter of academic niceties and citations; it is important to correct the record because the “others have suggested” rule has a much larger coefficient on the GDP gap and is therefore more likely to generate negative interest rates and be used to rationalize discretionary actions such as quantitative easing.

Second, the exchange between Chairman Bernanke and Senator Toomey suggests that the Fed is unclear about what monetary policy strategy it is using for the interest rate. Is it the Taylor Rule, as in the first response? Is it the rule incorrectly attributed to me in 1999, as in the second response? Is it some estimated rule, as in the third response? Or is it something else? It would be useful to know what the strategy is. Greater transparency about the strategy would add greatly to predictability and would help markets understand whether quantitative easing will be extended or when the interest rate will break out of the 0-.25 percent range.

For example if the strategy was reasonably well described by the Taylor Rule the interest rate would equal about 1.5 times the inflation rate plus .5 times the GDP gap plus 1. The most recent quarterly data (through the 4th quarter of 2010, released by Bureau of Economic Analysis on February 25, 2011) show that the inflation rate is about 1.4 percent (change in GDP deflator over the last four quarters). According to the average of the most recent survey by the Federal Reserve Bank of San Francisco, (January 28, 2011, Williams-Weidner) the GDP gap is about 4.4 percent. This implies an interest rate of 1.5 X1.4 + .5X(-4.4) + 1 = 2.1 + -2.2 +1 = 0.9 percent, or about 1 percent, which suggests that the Fed should be raising the rate sometime soon, perhaps before the end of this year. But if it is one of the other rules mentioned by the Chairman we might have to wait longer.

Third, the exchange shows how it would be quite feasible and useful to restore some of the reporting and accountability requirements which were removed from the Federal Reserve Act in 2000. As I have proposed, with such requirements the Fed would establish and report to Congress its strategy or policy rule for monetary decision making. If it later deviated from that strategy it would have to provide an explanation to Congress in writing and at a public congressional hearing. With such a reporting requirement, the testimony at such a hearing would be like this exchange between Chairman Bernanke and Senator Toomey with the very important exception that the Congress and the American people would have some idea of what the Fed's basic strategy was.