• Shortcuts : 'n' next unread feed - 'p' previous unread feed • Styles : 1 2

» Publishers, Monetize your RSS feeds with FeedShow:  More infos  (Show/Hide Ads)


Date: Friday, 11 Apr 2014 05:02

Recent indications are that, in the face of declining inflation, ECB President Draghi is considering embarking upon quantitative easing. Despite the technical difficulties accompanying such a measure [0] [1] I believe additional stimulative measures are nonetheless called for.

This figure from the recently released World Economic Outlook (p.55) demonstrates that while the overall euro area inflation rate is still (barely) positive, the minimum inflation has clearly dipped into the negative region.

ea_infl

Figure 2.3.3 from IMF, World Economic Outlook (April 2014), p. 55.

Not only is deflation occurring in the periphery countries, from one perspective, the need for higher short term inflation extends even to countries not experiencing deflation. Figure 1 depicts the euro area price level, as measured by the Harmonized Index of Consumer Prices (HICP), as well as the levels for several individual countries within the euro area, all normalized to 1999M01=0.

hicp_pix

Figure 1: HICP for euro area (bold blue), Germany (red), Ireland (green), Greece (teal), Spain (purple), France (olive), Italy (dark blue), Portugal (pink), 2% growth line (bold black), and 4% growth line (bold gray). Source: ECB, and author’s calculations.

Giannoni and Harmen argue that in the case of the United States, the Fed, by virtue of a long term 2% inflation target, has effectively implemented a two percent price level target. In some sense, the same appears true for the ECB and the euro area. Figure 1 indicates that the euro area wide HICP has followed the 2% price line (disturbingly, the overall HICP has dipped below the 2% line).

At this juncture, the distinction between the US as a monetary and fiscal union with high interstate factor mobility, the euro area as a monetary union with relatively low inter-country factor mobility, becomes important. While inflation is negative in the periphery countries, the deviation from the trend line is negative for the core (and large) euro area countries of Germany and France. The German deviation is about 5% in log terms. While the French deviation is smaller in absolute value, it contrasts with the pre-crisis value of essentially nil. If nominal debt had been accumulated with the expectation of the two percent trend in the price level, the very fact that the price level is lagging implies higher than expected debt burdens and hence more binding collateral constraints.

The IMF concludes:

Macroeconomic policies should stay accommodative. In the euro area, additional demand support is necessary. More monetary easing is needed both to increase the prospects that the ECB’s price stability objective of keeping inflation below, but close to, 2 percent will be achieved and to support demand. These measures could include further rate cuts and longer-term targeted bank funding (possibly to small and medium-sized enterprises). …

Unencumbered by institutional constraints, I would argue for even more forceful measures, aiming for a higher inflation target. In fact, by a 4% price level target (roughly equivalent to a 4% long run inflation target, as mooted by Blanchard in 2010 [2]), the periphery countries (aka GIIPS) should be aiming for a higher trend inflation rate.

In addition to reducing the likelihood of encountering the liquidity trap [3], and heightening the credibility of future inflation during periods of deflation [3], a steeper price level target allows for better relative price adjustment. This last point is critical given the conditions of the euro area, which include downward price rigidities (specifically, it is easier to have prices in country A rise by 5% and those in country B stay constant, than a 3% rise and 2% decline, respectively).

In other words, we need higher inflation now!

Author: "Menzie Chinn" Tags: "Uncategorized"
Send by mail Print  Save  Delicious 
Date: Thursday, 10 Apr 2014 02:58

No catch-up in sight for Wisconsin with Minnesota [1] (or the Nation, for that matter)

In a previous post, I noted that the most recent revisions had widened the gap between Wisconsin and other economies. Recently released forecasts from the Philadelphia Fed indicate the gaps will persist.

wisccoin2a

Figure 1: Log coincident indices for Minnesota (blue), Wisconsin (red), Kansas (teal) and for the United States (black), all normalized to 2011M01=0. Source: Philadelphia Fed, and author’s calculations.

As noted previously, the MN-WI gap has been revised to larger in the latest dataset, and the newest set of leading indicators indicate no diminishment of the gap.

Also notable is the fact that Kansas is in something of a slump, at least according to the most recent reading on current economic activity and the leading indices (February reading -0.1%). [Some discussion of the efficacy of Kansas's fiscal policy measures here -- added 4/10 7:47AM].

Author: "Menzie Chinn" Tags: "Wisconsin"
Send by mail Print  Save  Delicious 
Date: Tuesday, 08 Apr 2014 17:36

From the IMF:

…Although downside risks have diminished overall, lower-than-expected inflation poses risks for advanced economies, there is increased financial volatility in emerging market economies, and increases in the cost of capital will likely dampen investment and weigh on growth. Advanced economy policymakers need to avoid a premature withdrawal of monetary accommodation. Emerging market economy policymakers must adopt measures to changing fundamentals, facilitate external adjustment, further monetary policy tightening, and carry out structural reforms.

See the Chris Giles/FT assessment of the forecast.

The analytical chapters (released last week) address low interest ratesand emerging market challenges.

The entire report is here.

Author: "Menzie Chinn" Tags: "Uncategorized"
Send by mail Print  Save  Delicious 
Date: Tuesday, 08 Apr 2014 04:15

The Wisconsin Manufacturers and Commerce group has a recent piece, written by Scott Manley (VP), highlighting negative impacts of a minimum wage increase. I have attempted to track down the numbers cited. It has been an illuminating experience.

From the article:

A recent analysis by the Employment Policies Institute (EPI) found increasing the minimum wage to $10.10 per hour would kill as many as 27,937 jobs in Wisconsin. That estimate is consistent with a projection by the Congressional Budget Office (CBO) that predicted 500,000 lost jobs nationwide if the federal minimum wage was increased to $10.10.

It is instructive to see how this 27,937 figure is obtained. The estimate is found on this website sponsored by Employment Policies Institute. The supporting documentation, written by Even and Macpherson, reports the calculations using an elasticity of 0.3 for the high end. The website also reports estimated job loss, for a low-end elasticity of 0.1 (that low end elasticity, with associated job loss of 9,312 is not cited by Mr. Manley). The midpoint of the range is 18,625 (which might differ slightly from the estimate obtained using a 0.2 elasticity.)

Is the use of the high-end 0.3 elasticity justified? I would suggest not, as discussed in this post. It’s much higher than most recent estimates, and higher than the more precisely estimated coefficient values, as recounted in Doucouliagos and Stanley (2009).

Moreover, I cannot understand how the Wisconsin figure of 27,937 lost jobs is “consistent” with the CBO midpoint estimate of 500,000 lost jobs nationwide. 27,397 is 5.6 percent of 500,000; however, as of January 2014, Wisconsin nonfarm payroll employment constituted only 2.1 percent of national employment. It is conceivable that the composition of Wisconsin employment — in particular the minimum wage sensitive component — is so different from the national composition that the 27,397 figure is consistent with the CBO nationwide estimate. However, I must confess a such a large compositional effect seems implausible to me.

Here is an additional assertion from the Manley piece:

According to the U.S. Bureau of Labor Statistics, only 1.1 percent of workers over age 25 earn the minimum wage. That’s because typical minimum wage earners in America are teenagers living with their parents in middle class families. They are not living in poverty nor are they earning a wage that is responsible for sustaining a family.

This figure can be verified for one’s self by inspecting BLS, Characteristics of Minimum Wage Workers: 2013 (Mar. 2014), Table 1, page 4. Of course, this figure tells one how many are now at or below the current minimum wage. It doesn’t tell one how many would be affected by an increase to $10.10. According to a Economic Policy Institute study, using the CPS, 15.4 percent of all Wisconsin workers would be affected directly by the increase (by directly, I mean the constraint would bind; spillover, or indirect, effects on wages close to the new minimum wage are not incorporated). Here is the relevant table.

wisc_minwage

Table from David Cooper, Economic Policy Institute Briefing Paper #371, December 19, 2013

Would it only be youngsters affected, as implied by the WMC op-ed? According to the table, of the workers older than 29, 142,000 would be directly affected (constituting 5.4 percent of the total Wisconsin workforce). In other words, it’s not just teenagers in relatively affluent households that are affected. Figure 1 illustrates the age distribution:

wiagedist

Figure 1: Age distribution of directly affected Wisconsin workers. Source: David Cooper, Economic Policy Institute Briefing Paper #371, December 19, 2013

In fact, of the 404,000 directly affected by the $10.10 minimum wage, 104,000 individuals — with household income below $20,000 — would be affected. 104,000 is 25.7 percent of the total directly affected by a minimum wage increase to $10.10, and is 4 percent of the total workforce.

Finally:

The empirical evidence also suggests raising the minimum wage simply does not impact poverty levels.

This is an interesting statement, and is made on the basis of a Employment Policies Institute funded study written by Saiba and Burkhauser. This assertion is seemingly contradicted by this recent analysis that highlights the fact that earlier studies of the impact of the minimum wage on poverty might have been outpaced by recent developments.

In this study I show that the target efficiency of the minimum wage improved between 1999 and 2013. In 1999-2001, 15.3% of the minimum wage benefits went to workers in poor families. By 2011-2013 this figure had risen to 18.8%. Nearly two-thirds of the improvement in target efficiency occurred during pre-recession years (1999-2001 to 2005-2007), and the balance of the improvement occurred since the onset of the Great Recession. The improvement in target efficiency during pre-recession years was entirely due to an increase in the share of minimum wage workers in poor families. Decreased income among near-poor minimum wage workers drove the majority of the increase in the share of minimum wage workers in poverty. Reduced teen employment, increased teen wages (relative to the minimum wage), and increased employment among poor low-skilled 20-29 year-olds also contributed.

The WMC conclusion also contrasts strongly with the CBO reading of the literature, and the corresponding estimate that 900,000 individuals would be moved above the poverty line by a $10.10 minimum wage (see Table 1, page 2).

It is at this juncture it is useful to remember what the Employment Policies Institute is. As noted in the NY Times, it is a nonprofit organization which bills a for-profit organization for services (and shares office space with that for-profit organization). The IRS form 990 for the Employment Policies Institute makes for illuminating reading.

Bottom line: Beware of unfootnoted articles! And beware the estimates of the Employment Policies Institute!

Update, 4/8 2PM Pacific: Arindrajat Dube points me this paper:

I use data from the March Current Population Survey between 1990 and 2012 to evaluate the
effect of minimum wages on the distribution of family incomes for non-elderly individuals. I find robust evidence that higher minimum wages moderately reduce the share of individuals with incomes below 50, 75 and 100 percent of the federal poverty line. The elasticity of the poverty rate with respect to the minimum wage ranges between -0.12 and -0.37 across specifications with alternative forms of time-varying controls and lagged effects; most of these estimates are statistically significant at conventional levels. For my preferred (most saturated) specification, the poverty rate elasticity is -0.24, and rises in magnitude to -0.36 when accounting for lags.

I also use recentered influence function regressions to estimate unconditional quantile partial effects of minimum wages on family incomes. The estimated minimum wage elasticities are sizable for the bottom quantiles of the equivalized family income distribution. The clearest effects are found at the 10th and 15th quantiles, where estimates from most specifications are statistically significant; minimum wage elasticities for these two family income quantiles range between 0.10 and 0.43 depending on control sets and lags. I also show that the canonical two-way fixed effects model—used most often in the literature—insufficiently accounts for the spatial heterogeneity in minimum wage policies, and fails a number of key falsification tests. Accounting for time-varying regional effects, and state-specific recession effects both suggest a greater impact of the policy on family incomes and poverty, while the addition of state-specific trends does not appear to substantially alter the estimates. I also provide a quantitative summary of the literature, bringing together nearly all existing elasticities of the poverty rate with respect to minimum wages from 12 different papers. The range of the estimates in this paper is broadly consistent with most existing evidence, including for some key subgroups, but previous studies often suffer from limitations including insufficiently long sample periods and inadequate controls for state-level heterogeneity, which tend to produce imprecise and erratic results.

Author: "Menzie Chinn" Tags: "labor market"
Send by mail Print  Save  Delicious 
Date: Monday, 07 Apr 2014 14:51

Congratulations to James Gualtieri, winner of the world famous Econbrowser NCAA tournament challenge.

Although the final game won’t be played until tonight, we already know the winner of the challenge, since none of the prognosticators in our competition picked either of the teams playing in the championship game. Gualtieri did pick Wisconsin to be in the Final Four, which I knew was a good call as soon as I saw them playing in the tournament. Like many of us, though, he also thought Duke would do better.

If things didn’t go as you wanted, don’t worry. There’s always next year!

Author: "James_Hamilton" Tags: "here and there"
Send by mail Print  Save  Delicious 
Date: Sunday, 06 Apr 2014 22:22

Last week ECB President Mario Draghi revealed that the European Central Bank has been considering large-scale asset purchases as a tool to prevent European inflation from falling too far below the ECB’s target rate of 2%. What is the evidence for the effectiveness of these policies, and are there any risks?

Historically, when the U.S. economy went into recession, the Federal Reserve would lower interest rates through its control of the fed funds rate, an interest rate on overnight loans between banks. Lower interest rates helped stimulate demand for items such as housing and autos and raise asset prices more generally, all of which would help spur economic recovery. But that tool is not available to the Fed in today’s environment in which the fed funds rate is essentially already zero.

Source: FRED.

Source: FRED.

The Fed has adopted less conventional policies in the current environment by buying assets in huge volumes (large-scale asset purchases) and by indicating its intention to keep the fed funds rate low even after the economy was well into recovery (forward guidance). Although the trillions of dollars involved in LSAP may seem dramatic, the reserves that the Fed created to pay for these operations are for the most part still just sitting idle on banks’ balance sheets at the end of each day. Since those reserves earn interest when left with the Fed overnight, I think the best way to think of them is as overnight loans from banks to the Fed. If all the Fed did with these operations was buy short-term Treasury bills, essentially LSAP would just swap one short-term liability of the U.S. government (Treasury bills) for another very similar short-term liability of the U.S. government (reserves held in accounts with the Fed). Such a swap would be unlikely to change anything that matters regardless of how many trillions of dollars were involved.

For this reason, the Fed has been buying not Treasury bills but instead longer-maturity Treasury bonds and mortgage-backed securities. Taking a huge volume of these assets out of private hands and replacing them with overnight government liabilities could perhaps affect the yields on long-term Treasuries and MBS. An early analysis by Modigliani and Sutch (1966) found little evidence of an effect on interest rates of “Operation Twist” in 1961 when the Treasury tried to issue more short-term and fewer long-term securities, though a more detailed analysis of that episode by Swanson (2011) concluded that Operation Twist did have a measurable effect on yields. A number of other studies have looked at the variation over time in the maturity structure of U.S. Treasury debt and have found statistically significant correlations between the maturity structure and relative yields; see for example Kuttner (2006), Gagnon, et. al. (2011), Greenwood and Vayanos (2013), Doh (2010), and Hamilton and Wu (2012) (ungated version here).

Another approach is to look at what happened to yields on days when the Fed announced changes in LSAP that caught the markets by surprise. For example, when the Fed first announced the scope of its intended large-scale asset purchases on March 18, 2009, the yield on 10-year Treasuries fell 51 basis points. Last May, when the Fed indicated it was considering slowing the pace of asset purchases, yields went back up. Analyses of events like these provide fairly convincing evidence that market yields react to Fed announcements of changes in its LSAP or forward guidance.

A new paper by Rogers, Scotti and Wright (2014) reviews and updates evidence from the experience of the U.S., U.K., Europe, and Japan, and concludes, like most of the studies summarized in the table above, that LSAP and forward guidance are potentially effective policies for easing financial conditions when the short-term interest rate is already near zero.

Another new study by Hayashi and Koeda provides further evidence looking at the data in a completely different way. They note that during episodes in which Japan’s short rate was near zero, an increase in the level of excess reserves in month t tended to be followed by a higher level of output in month t+1.

There is thus considerable evidence that these unconventional monetary policy measures have some modest potential to help stimulate a struggling economy. Less clear is what risks may be associated with unconventional monetary policy. If we think of LSAP as primarily a shortening of the maturity structure of outstanding government debt, replacing long-term Treasury bonds with overnight liabilities of the central bank, it’s clear that one would not want to expect too much from such measures. Assuming though there are some benefits for a struggling economy of converting government debt to shorter-term liabilities, why might the Treasury be reluctant to shorten its maturity structure? The answer is fairly obvious. Although there might be plenty of demand for short-term obligations of the U.S. government right now, will that still be the case one or two years down the road? If the Treasury was forced into a situation of borrowing in a future setting in which the huge daily auction of Treasuries turned out to be undersubscribed, the result could be a sharp spike in interest rates and substantial financial instability.

How are things different when the government’s short-term liabilities are in the form of overnight deposits with the Federal Reserve? One important point to note is that although an individual bank can always get rid of reserves it doesn’t want by buying some alternative asset, in doing so those reserves don’t disappear from the system, but are instead simply passed on to another bank from whose customer the first bank purchased the asset. Some bank will end up holding the reserves whether or not anybody actually wants them. In this sense, the Fed can always be sure that someone will lend to them overnight, whereas the Treasury cannot.

If demand for reserves is less than the supply, it’s not the supply of reserves that has to change, but instead the change has to come in terms of the price on alternative assets. For example, if assets denominated in foreign currency look more attractive than low-yielding deposits with the Fed, the dollar would have to decline sufficiently until one dollar would buy so little in the way of foreign assets that it no longer is attractive to try to do so. How far would that be? To adapt President Draghi’s famous phrase, “whatever it takes.” If banks are holding trillions more in reserves than they want, I could imagine this adjustment in items such as the exchange rate and commodity prices being rapid and chaotic.

Although the mechanics of a flight from unwanted reserves would work differently from those for an undersubscribed Treasury option, it seems to me that the economic fundamentals are the same. When faced with such a situation, there would be nothing the Treasury or the Fed could do to avoid a spike in interest rates and a potentially very disruptive financial situation if lenders are no longer willing to continue to roll over what has effectively become a huge volume of overnight government debt.

It is of course inappropriate to be paralyzed by fear of such a scenario at the present moment, when interest rates and inflation are so low; any such problems will not arise until well into the future. But I also do not think it is inappropriate to dismiss these considerations entirely, particularly if we take the view (as I do) that the benefits of additional LSAP are also likely to be fairly modest. In the case of Europe, the ECB is missing its inflation target by a sufficiently large margin that further action seems to be warranted. But in the case of the United States, I think the current course signaled by the Federal Reserve– that growth of its balance sheet will end by the end of this year– is the correct one.

Author: "James_Hamilton" Tags: "Federal Reserve"
Send by mail Print  Save  Delicious 
Date: Saturday, 05 Apr 2014 03:40

After government spending fell, industrial production, GDP, and consumption decline, and unemployment rises.

Reader Patrick R. Sullivan tries to argue that the US economy boomed immediately after the reduction in government spending at the end of World War II.

I find it useful when asserting something to have some facts at hand. Here are data showing what happened as government expenditures rose, in 1941, and as they fell, in 1945.

ww2

Figure 1: Government consumption and investment spending, at annual rates, in billions of dollars (blue), and unemployment rate (bold red), and annual averages from Lebergott (1957) (red). NBER defined recession dates shaded gray. Source: NBER Macrohistory database via FRED, and Lebergott (1957), and NBER.

Bottom line: Historical data indicate when government spending rose tenfold, unemployment fell from 12% to 1%; when spending fell to about 1/3 of peak, then unemployment rose to 4%.

Update, 4/5 8:25AM Pacific: Mr. Sullivan rules unemployment (as well as GDP) as meaningless during this period. Specifically, for unemployment, he writes that it’s a meaningless statistics since so many individuals are in the military. So, I will repeat this Figure from the previous post that sparked the debate — plotting GDP, employment (not unemployment), and industrial production.

really_a_boom_in_1945

Figure 2: Log nonfarm payroll employment (blue), log industrial production (red) and log real GDP measured in Ch.2009$ (bold green), all normalized to 1944M12=0. Real GDP interpolated using quadratic-match-average. Tan shaded area is 1945 onward. Source: BLS and Federal Reserve via FRED, and Measuring Worth, and author’s calculations.

I fully expect a exegesis from Mr. Sullivan on how down is actually up for industrial production and employment.

Update, 4/5 9:25AM Pacific: Now one could define a “boom” by reference to private activity. Here is consumption expenditures, deflated by the wholesale price index.

ww2consumption

Figure 3: Real consumption expenditures (blue), and seasonally adjusted using ARIMA X-12 over 1939-64 by MDC (red), both at annual rates, in billions of 1957-59$. The deflator is the wholesale price index, 1957-59=100. Source: NBER Macrohistory database via FRED,NBER, and author’s calculations.

Note that unadjusted consumption declines after 1945Q4, and seasonally adjusted consumption peaks at 1946Q2.

Author: "Menzie Chinn" Tags: "economic indicators"
Send by mail Print  Save  Delicious 
Date: Friday, 04 Apr 2014 17:39

BLS reports nonfarm payroll employment up by 192 thousand, and prior months revised up.

nfp_mar14

Figure 1: Nonfarm payroll employment from January release (blue), from February release (red), from March release (black), in thousands, seasonally adjusted. Bloomberg consensus in gray (based upon 206 thousand increase, using March series observation on February employment). Source: BLS, Bloomberg, author’s calculations.

For more on the release, see Portlock/WSJ RTE Leubsdorff/WSJ RTE, Furman/CEA.

Author: "Menzie Chinn" Tags: "employment"
Send by mail Print  Save  Delicious 
Date: Thursday, 03 Apr 2014 03:31

(The can opener reference can be understood by clicking here)

The budget proposed by Representative Ryan touts the pro-growth impacts of deficit reduction ($5.1 trillion over ten years, according to Table S-2). It is instructive to actually read the documents that Representative Ryan’s budget cites (as it has in the past — read about the Heritage CDA previous assessments [1] [2], as well as Representative Ryan’s previous attempt to use CBO documents to lend a patina of respectability to his projections).

From the House budget proposal:

In a report published in February of 2013, CBO concluded that reducing budget deficits, thereby bending the curve on debt levels, would be a net positive for economic growth. According to that analysis, a large deficit-reduction package of $4 trillion, which this budget resolution actually exceeds, would increase real economic output by 1.7 percent in 2023. Their analysis concludes that deficit reduction creates long-term economic benefits because it increases the pool of national savings and boosts investment, thereby raising economic growth and job creation. The greater economic output that stems from a large deficit-reduction package would have a sizeable impact on the federal budget. For instance, higher output would lead to greater revenues through the increase in taxable incomes. Lower interest rates and a reduction in the stock of debt would lead to lower government spending on net interest expenses. CBO finds that this dynamic would reduce budget deficits by a net $186 billion over ten years, including $82 billion in the tenth year alone.

Two observations, regarding the CBO “discussion” of the Ryan budget:


  • The “assume a can opener” component is here again — the paths of revenues and spending are assumed to hold, and the plausibility of these were not assessed by CBO.
  • Even with the aid of dynamic scoring, there is an “interesting” omission regarding time frame of positive impacts and the degree of uncertainty.

Regarding the first point, here is the quote from the CBO updated discussion.

The projections do not represent a cost estimate for legislation or an analysis of the effects of any specific policies. In particular, CBO has not considered whether the specified paths are consistent with the policy proposals or budget numbers that Chairman Ryan released on April 1, 2014, as part of his proposed budget resolution.

On the second point, the astute observer will note that these dynamic effects are emphasized only when Representative Ryan writes about the long term (2025 onward), as displayed on Page 94, Table S-6 of the House budget. Short run effects are in Table S-2. There is a very good reason for the de-emphasis of short term effects. One discerns this reason when one actually reads the CBO document cited in the House budget, and inspects Figure 2, on page 7.

deficit_reduction_impact

Figure 2 from CBO, Macroeconomic Effects of Alternative Budgetary Paths, (Feb. 2013).

In other words, for a ten-year $4 trillion deficit reduction, in the subsequent four years, the net impact on GDP is negative. Figure 2 pertains to a $4 trillion plan implemented in 2014. The corresponding impacts for the current incarnation of the current House budget could be approximated by shifting forward the curves by one year. According to the CBO discussion of the Ryan proposal (Figure 4, page 9), if the Ryan budget is implemented, then by 2016, per capita GNP will be nearly a percentage point lower (-0.8% to be specific). Of course, by 2025, using midpoint estimates, output per capita would be higher. However, there is wide uncertainty regarding the impact, as indicated in Table 2: the range is from as little as 0.8 ppts to 2.6 ppts higher than baseline.

Stepping aside from the macroeconomic implications, it is of interest to observe the choices Representative Ryan has made in his cuts; these include $125 billion to SNAP (partly by implementation of work requirements, see page 63). According to CBPP, ten-year cuts to Medicaid and SNAP would constitute $0.9 trillion. This apparently this is his view of the “roadmap out of poverty”.

The CBO discussion of the Ryan budget did not incorporate behavioral responses due to changes in policies (e.g. workhouses implementation of work requirements associated with SNAP), but it also did not incorporate changes in government investment productivity that contributes to future output. For a case wherein supply-side aspects were incorporated into the dynamic-scoring process, see here.

Update, 9:15PM Pacific: Keith Hennessey compares the Ryan budget and the Obama budget, with the caveat:

In this post I’m just going to compare the short-term deficit and debt effects of the two proposals. While I’d like to use comparable numbers, CBO has not yet rescored President Obama’s proposal because the President released his budget six weeks late. So for now I’ll compare Ryan’s numbers to Obama’s. That is suboptimal but the best we can do for now, and I am confident it doesn’t change the overall picture. Let’s start with deficits.

The way this is stated, one gets the impression that the Ryan budget has been scored. I don’t know if this is what Mr. Hennessey intended, but of course, regardless of intent, it is a misleading statement, because the Ryan plan has not been scored either. CBO took at face value Representative Ryan’s spending and tax revenue numbers, and tabulated the macro feedback effects. Hence, it would not be right to compare a CBO-scored Obama budget to Ryan’s numbers, and Mr. Hennessey knows it (or at least should).

Update, 4/4, 11AM Pacific: From the chorus of the data-challenged, Ed Hanson asks:

Can you think of a time or occurrence in US history when The Fed Government drastically reduced its budget and borrowing and yet the economy boomed?

And Patrick R. Sullivan answers: “September 1945″, appealing to the NBER peak/trough dates. Well, a relevant question is what constitutes a “boom”, and these same people who argue that the current expansion is not a boom are happy to tabulate the post post-September 1945 period as a boom. All I can say is that they either must (1) not believe in reading data, or (2) have a radically different interpretation than I do of what a “boom” is. Here is my depiction of data around 1945; I will let the readers decide if post-1945 through 1946 was a “boom” (I’m willing to say it’s a recovery). Just ignore the 15%-odd decline in GDP, and the 20%-odd decline in industrial production (log terms).

really_a_boom_in_1945

Figure 1: Log nonfarm payroll employment (blue), log industrial production (red) and log real GDP measured in Ch.2009$ (bold green), all normalized to 1944M12=0. Real GDP interpolated using quadratic-match-average. Tan shaded area is 1945 onward. Source: BLS and Federal Reserve via FRED, and Measuring Worth, and author’s calculations.

If you say September 1945 through 1946 was a boom, well, you’d better be willing to say 2009M06 ownard was a “boom”.

Author: "Menzie Chinn" Tags: "budget"
Send by mail Print  Save  Delicious 
Date: Tuesday, 01 Apr 2014 19:03

Today we are fortunate to have a guest contribution by Ron Alquist, Policy Adviser at the Bank of Canada, and Olivier Coibion, Assistant Professor at the University of Texas, Austin. The views expressed here are those of the authors and should not be interpreted as representing the views of the Bank of Canada or any other institution with which the authors are affiliated.



From droughts in the American Midwest to labor strikes in the mines of South America to geopolitical instability in the Middle East, there are many potential sources of exogenous commodity-price fluctuations. Such changes in commodity prices can, in principle, have substantial macroeconomic effects and many have argued that shocks to commodity prices have indeed contributed significantly to global business cycles, such as those observed in the 1970s (Hamilton 1983; and Blinder and Rudd 2012) or more recently at the onset of the Great Recession (Hamilton 2009). But because commodity prices also respond to changes in global business cycle conditions, decomposing changes in commodity prices into those reflecting endogenous responses to global business cycles and those stemming from shocks originating in commodity markets has proven challenging. In a recent paper, we propose a new empirical strategy based on the theoretical predictions of a model of the comovement in non-energy commodity prices to identify the sources of historical commodity-price changes and their global macroeconomic implications (Alquist and Coibion 2014). The main conclusion of the paper is that commodity-related shocks have contributed modestly to global business cycle fluctuations, a finding in line with Kilian (2009) for oil markets.

Our approach to understanding the drivers of the comovement in commodity prices uses a theoretical model in which the prices of commodities are determined by two sets of forces. First, there are the forces that affect commodity prices directly, by which we mean forces which alter the supply or demand for commodities even for a fixed level of global economic activity. For example, we classify a technological improvement in the production of commodities as such a force, because it would increase the supply of commodities even in the absence of any subsequent effects on global economic activity. Of course, to the extent that these forces change commodity prices, they ultimately alter the level of global economic activity and feed back into commodity prices through general-equilibrium effects. But the key to identifying these forces is that they affect commodity prices even absent any endogenous response of global activity. By contrast, the second set of forces are those that affect commodity prices only through the changes that they induce in the level of global economic activity, i.e. “indirectly”. Changes in government spending, variation in the desired markups for the production of consumer goods, or improvements in the technology used to produce final goods are all examples of such forces. The composition of all such forces is summarized by the indirect factor. These two drivers are common to all commodity prices. The model also permits there to be idiosyncratic forces specific to individual commodities.

The key insight of classifying the common forces into these two categories is that they have different implications for the comovement in commodity prices. Specifically, the set of forces that affect commodity prices indirectly – that is, the indirect factor – through their effect on production all induce the same relative comovement in commodity prices, because their effects on commodity markets are summarized by their implications for global economic activity. Thus, for instance, an increase in the demand for commodities related to the global business cycle increases the prices of all commodities. By contrast, the common movement related to the forces that affect the supply of and demand for commodities directly – that is, the direct factor – induce different relative commodity-price movements. There is no guarantee that such changes induce similar comovement in commodity prices. Using the comovement of commodity prices therefore provides novel a way of identifying the channels underlying commodity price changes.

We apply these insights to a new data set of forty non-energy commodity prices to examine the role of these two factors in explaining historical commodity-price fluctuations. Figure 1 below depicts a historical decomposition of the annual percentage change in average commodity prices and relates them to the indirect and direct factors. The common driver that explains most of the variation in commodity prices is the one related to the indirect factor. Around 60 to 70% of the historical commodity-price movements are associated with the common driver that is related to aggregate non-commodity shocks rather than to direct shocks to commodity markets. During the commodity boom of 1973-74, for example, indirect shocks to commodity markets accounted for over two-thirds of the increase in commodity prices, with the remainder reflecting direct commodity-related shocks. Similarly, the fall in commodity prices during the Volcker era of the early 1980s is attributed almost entirely to a decrease in the indirect factor.

alquist_coibion_commod

Figure 1: The Contribution of “Indirect” and “Direct” Factors to Historical Commodity-Price Changes. Notes: The figure plots the annual change in commodity prices (average across forty commodities): solid black line, the contribution coming from “indirect” global factors (IAC factor): blue shaded area, and the contribution from direct commodity factors (DAC factor): red shaded area. US recessions are identified using vertical grey shaded areas.

The second commodity boom of the 1970s, however, suggests a more nuanced interpretation. While the increase in commodity prices in 1976 reflected rising levels of global economic activity, the indirect factor contributed much less to rising commodity prices during the second half of 1978 and was actually pushing commodity prices lower for most of 1979. Despite this downward pressure from non-commodity shocks, the direct factor pushed real commodity prices higher during 1979 and did not weaken until early 1980. Thus, while the bulk of the second commodity boom of the 1970s can be interpreted as an endogenous response of commodity prices to non-commodity shocks, commodity-related shocks played an important role in extending the period of rising commodity prices into early 1980.

The decomposition of commodity prices since the early 2000s also presents a mixed interpretation. While a substantial share of the increase in commodity prices since 2003 is accounted for by the indirect factor, the direct factor accounts for much of the surge in prices during 2004 and about 30% of the increase from early 2006 to late 2007. The majority of the subsequent decrease in commodity prices between October of 2008 and March of 2009 is also attributable to the direct commodity factor, while the indirect factor accounts for most of the continuing decline after March 2009. Out of the total decline in commodity prices between October of 2008 and October of 2009, over half (56%) was due to direct commodity shocks. By contrast, the resurgence in commodity prices since the end of 2009 primarily reflects non-commodity shocks as measured by the indirect factor.

Our approach also has a more practical application: forecasting. Because we rely only on commodity prices traded on global markets, our decomposition can be applied in real-time to forecast a wide range of commodity prices within a common framework. We show that a simple model that includes the indirect factor and the price of an individual commodity generates out-of-sample improvements in forecast accuracy relative to the no-change forecast — that is, the forecast that assumes the price of the commodity does not change. The improvements in forecast accuracy are the largest at the 1- to 6-month horizons. What is more, the model helps to forecast the behavior of broader commodity-price indices, such as those produced by the IMF and the World Bank. The model also works well at forecasting the price of oil at short horizons. The improvements in oil forecasting accuracy are similar in size to those based on models of the global oil market (see, e.g., Baumeister and Kilian 2012; and Alquist et al. 2013), but our model does not require data on quantities produced, which are often available only with a delay. Our approach thus provides a coherent and tractable framework for forecasting commodity prices while also providing an economic interpretation for the forecasts.
In sum, our paper offers a new framework for identifying the sources and implications of commodity-price comovement and its relationship to global macroeconomic conditions. The framework provides an economic interpretation of the common factors driving commodity prices and offers a new perspective on the historical behavior of a broad cross-section of internationally traded commodities since the early 1970s. It can also be used to generate more accurate real-time forecasts for a wide range of commodity prices and commodity-price indices.


References


  • Alquist, Ron and Olivier Coibion, 2014. “Commodity-Price Comovement and Global Economic Activity,” NBER Working Paper 20003.
  • Alquist, Ron, Lutz Kilian, and Robert Vigfusson, 2013. “Forecasting the Price of Oil,” forthcoming in: G. Elliott and A. Timmermann (eds.), Handbook of Economic Forecasting, 2, Amsterdam: North-Holland.
  • Baumeister, Christiane and Lutz Kilian, 2012. “Real-Time Forecasts of the Real Price of Oil,” Journal of Business and Economic Statistics, 30(2), 326-336.
  • Blinder, Alan S. and Jeremy B. Rudd, 2012. “The Supply Shock Explanation of the Great Stagflation Revisited,” NBER Chapters, in: The Great Inflation: The Rebirth of Modern Central Banking National Bureau of Economic Research, Inc.
  • Hamilton, James D., 1983. “Oil and the Macroeconomy since World War II,” Journal of Political Economy 91(2), 228-248.
  • Hamilton, James D. 2009. “Causes and Consequences of the Oil Shock of 2007-2008,” Brookings Papers on Economic Activity 2009(2): 215-259.
  • Kilian, Lutz. 2009. “Not All Oil Price Shocks Are Alike: Disentangling Demand and Supply Shocks in the Crude Oil Market,” American Economic Review, 99(3), June 2009, 1053-1069.

This post written by Ron Alquist and Olivier Coibion.

Author: "Menzie Chinn" Tags: "commodities"
Send by mail Print  Save  Delicious 
Date: Tuesday, 01 Apr 2014 19:01

Philadelphia Fed revised coincident indices released today show the cumulative Minnesota-Wisconsin gap (since January 2011) has grown to 2.3% (as of February, in log terms).

Moreover, in my 3/20 post on the Midwestern laggard (Wisconsin), I observed that the cumulative gap between Minnesota and Wisconsin was 1.8% as of January. The revised series released today indicate the cumulative gap between Minnesota and Wisconsin for January is now over 2.2% (vs. the previous 1.8%, both in log terms). Hence, Wisconsin’s underperformance was even larger than previously estimated.

feb14coinpix

Figure 1: Log coincident indices for Minnesota (blue), Wisconsin (red), California (green) and for the United States (black), all normalized to 2011M01=0. Source: Philadelphia Fed, and author’s calculations.

Data regarding Wisconsin’s lagging performance on nonfarm payroll employment, in this post.

Update, 4:40PM Pacific: dilbert dogbert suggests adding in Kansas. Lo and behold, Kansas’s (under)performance currently “rivals” that of Wisconsin.

feb14coinpix1

Figure 2: Log coincident indices for Minnesota (blue), Wisconsin (red), California (green), Kansas (teal), and for the United States (black), all normalized to 2011M01=0. Source: Philadelphia Fed, and author’s calculations.

Dilbert dogbert refers the reader to this CBPP article:

Tax cuts enacted in Kansas in 2012 were among the largest ever enacted by any state, and have since been held up by tax-cut proponents in other states as a model worth replicating. In truth, Kansas is a cautionary tale, not a model. As other states recover from the recent recession and turn toward the future, Kansas’ huge tax cuts have left that state’s schools and other public services stuck in the recession, and declining further — a serious threat to the state’s long-term economic vitality. Meanwhile, promises of immediate economic improvement have utterly failed to materialize.

Author: "Menzie Chinn" Tags: "Wisconsin"
Send by mail Print  Save  Delicious 
Date: Tuesday, 01 Apr 2014 01:22

Senator Barraso (Republican-Wyoming) follows in the fine tradition of Allen West in charging a conspiracy so vast…

From FoxNews, Barraso describes the Administration’s statement on enrollments [1] thusly:

“I don’t think it means anything,” he told Fox News Sunday. “They are cooking the books on this.”

The Administration’s estimates are depicted in this figure

ACA_enroll

Figure from Andrews, Park and Tse, “Ten Key Questions on Health Care Enrollment,” NY Times (27 March 2014).

The tabulation by Charles Gaba at ACASignup.net estimates 6.9 million enrollees by March 31 (h/t Paul Krugman).

See also Paul Krugman‘s discussion of the strange failure of the media to highlight the achievement of the “target”. Nonetheless, while we are likely well on the way to exceeding the 6 million target, there are several targets that we do not yet know whether we have hit [3] (although I am not sure that it is required that actual 40% youth enrollment is necessary for viability).

Several observers have raised the issue of paid vs. unpaid enrollments; Secretary Sebelius has indicated 80-85% of enrollments are paid. ACASignups provides an estimate at 85%, as well.

For a broader measurement of coverage under all aspects of the ACA, see this graph:

aca_chart_140330b (1)

Figure 2:from ACASignups.

For other examples of data paranoia, see here and here.

Author: "Menzie Chinn" Tags: "health care"
Send by mail Print  Save  Delicious 
Date: Sunday, 30 Mar 2014 21:23

Oil produced from tight formations in the United States inaccessible before the days of horizontal fracturing is now accounting for 4.3% of total global crude oil production, according to new estimates released by the EIA last week.

Source: EIA.

Source: EIA.

Most of the tight oil production is coming from the Bakken in North Dakota and Eagle Ford in Texas, though 340,000 b/d is now also being produced in Canada and 120,000 b/d in Russia.

Source: EIA.

Source: EIA.

But this does not mean a flood of new oil is about to come on market. If it were not for the new tight oil, total world field production would be lower today than it was in 2005. And the new stuff is expensive. Chevron CEO John Watson last month said “Essentially, for a company like mine and many others, $100 a barrel is becoming the new $20 in our business”. And even with these elevated prices, Royal Dutch Shell; recently gave up on its large investments in Texan tight oil after concluding they couldn’t make a profit.

New production from tight formations drove the price of natural gas below the point at which it was profitable to produce. We’re still in the process of correcting that, and I expect natural gas prices to continue to rise from here. That same cycle of undershooting the sustainable price could replay in oil markets.

We’re used to thinking of a technological advance as something that enables us to produce better products at a lower price. Accessing tight oil formations using fracking is an important technological advance. But it’s clearly a much inferior source of energy compared to the days when we could just drill a few hundred feet into the earth and the oil would come gushing out. And anyone interpreting the recent trends as signaling that we’re about to return to that kind of a world is misreading the true meaning of what is happening.

Author: "James_Hamilton" Tags: "energy"
Send by mail Print  Save  Delicious 
Date: Friday, 28 Mar 2014 20:49

Why bother with econometrics when revealed truth will do?

Reader Rick Stryker comments on my post on employment effects of the minimum wage:

When you look at the research, the elasticities of employment of young people with respect to the minimum wage are typically between -0.1 to -0.3. … Let’s take -0.2 as the central estimate [of the elasticity for young people with respect to the minimum wage]

If we compare to the CBO study, we see they get an estimate of 500K with an upper bound estimate of 1 million, which is a lot smaller than my back of the envelope. The reason for that is that CBO makes many downward adjustments to the published estimates. For example, they use an elasticity of -0.1 as their central estimate, based on their judgmental reading of which papers they believe are more accurate. They also claim there is publication bias and so adjust downward. And they also make a number of downward adjustments by trying to estimate the number of people affected, etc.


I don’t believe these non-transparent adjustments are very credible. …

Let’s take a look at where this supposed conventional wisdom lies in the context of the extant estimates, as reported by Hristos Doucouliagos and Tom D. Stanley, “Publication Selection Bias in Minimum-Wage Research? A Meta-Regression Analysis,” British Journal of Industrial Relations 47.2 (2009): 406-428. [ungated working paper version]. I illustrate where Rick Stryker’s preferred estimate lies in this familiar funnel graph, which plots 1/standard errors (1/se) of an estimate against the point estimate. Observations higher on the graph are more “precise”, as measured by the standard error of the estimate.

funnel_minwageelast1

Figure 2 from Doucouliagos, Hristos, and Tom D. Stanley. “Publication Selection Bias in Minimum-Wage Research? A Meta-Regression Analysis.” British Journal of Industrial Relations 47.2 (2009): 406-428. [ungated working paper version], with red line drawn in at elasticity = -0.2.

I have drawn the red line at the reader’s preferred estimate of -0.2. Now it is true that these estimates in the funnel graph pertain to elasticities for differing groups — not all for young workers. Still, just looking at all these estimates, I would be tempted to focus on an elasticity of around -0.05, rather than -0.2 — but that’s just me. (I invite readers to vote by commenting, with explanation.)

By the way, if one is interested in the methodology of meta-studies, one of the authors of this paper, T.D. Stanley, has an article in the Journal of Economic Perspectives exactly on this subject: Stanley, Tom D. “Wheat from chaff: Meta-analysis as quantitative literature review.” Journal of economic perspectives (2001): 131-150.

So, for me, I think the assessment of small or zero impacts is the most reasonable (and I am willing to entertain positive impacts, especially in the short run when income effects are more plausible).

Update, 3/30 1:25PM Pacific: I strikes me that people might want to know what the result of the meta-analysis is for the point estimate. From the article:

The one difference between the best-set and the all-set MRA results is that the latter suggests the existence of a very small, but statistically significant, negative minimum-wage effect. A 10 per cent increase in the minimum wage reduces employment by about 0.10 per cent (see column 4 of Table 3). But even if this adverse employment effect were true, it would be of no practical relevance. An elasticity of -0.01 has no meaningful policy implications. If correct, the minimum wage could be doubled and cause only a 1 per cent decrease in teenage employment.

The best-set estimates range from -0.002 to -0.024, none of which are statistically significant (see Table 2).

Author: "Menzie Chinn" Tags: "labor market"
Send by mail Print  Save  Delicious 
Date: Thursday, 27 Mar 2014 22:10

With updates on the econometric debates on effects, and efficacy in targeting low income groups (3/30)


Minimal employment number impacts and minimal inflation impacts. But I am sure the resistance to having a greater share of income going to labor will continue.

From Goldman Sachs, “What to Expect from a Minimum Wage Hike” (3/25, not online), a survey of studies relevant to the debate over employment effects:

GS_Summary_MinWage_mar14

Source: Michael Cahill and David Mericle, “What to expect from a minimum wage hike,” GS Daily (3/25/2014).

Confirming the summaries of the literature contained in CBO and CEA (discussed in this post), most estimates are for quantitatively small impacts on employment, even when the estimates are statistically significant. It’s important to further recall that in the CBO assessment, the distribution of estimates spans positive impacts on employment (for some simple analytics of why this can occur in the short run, see this post; people averse to analytics should steer clear). Regarding the CBO midpoint estimate, the authors remark:

In our view, the CBO estimate is likely a bit toward the upper end of reasonable estimates, for two reasons. First, as Exhibit 3 shows, a large number of economic studies have found no statistically significant effect. Second, demand effects are likely to be particularly pronounced under current conditions in which considerable slack remains in the economy with the funds rate already near zero. As a result, raising the incomes of low-wage workers, who are likely to spend a larger share of their income, should provide a larger-than-usual offsetting boost.

That is, for the same reason fiscal policy has a greater impact at the ZLB and when slack exists, [1] boosts to labor income from an increase in the minimum wage can have a meaningful impact.

I’m teaching econometrics this semester, and it’s interesting to see how some of the prominent studies in the minimum wage literature dovetail with the more recent (not new any more) approaches to controlling for endogeneity — in particular the use of quasi-natural experiments, as in the Card and Krueger (AER, 1994) study of the New Jersey minimum wage increase. In that case, they used a differences-in-differences approach, examining how the gap between NJ-PA employment growth changed after the advent of the minimum wage. Card and Krueger found that there was a small employment increase in employment growth when NJ raised its minimum wage (borderline significant in many cases, and significant in others).

As the Note remarks upon, a number of states have recently implemented increases in their minimum wage rates; these changes constitute a series of quasi-experiments. Here is their assessment of the early outcomes:

… January’s state-level payrolls data failed to show a negative impact of state-level hikes. Relative to recent averages, the group of states that had hikes at the start of 2014 in fact performed better than states without hikes. While this is only one month’s data, it suggests that the negative impact of a higher federal minimum wage–if any–would likely be small relative to normal volatility.

The authors also conduct an event analysis of minimum wage increases on inflation on post-1990 data. They find no evidence of a discernable impact on PCE inflation. Their best estimate is 0.3 percent elevation in the price level by the end of the three year implementation of the minimum wage increase to $10.10.

The impact on employment is minimal, while wages will rise for many, thereby inducing an increase in the low-wage labor share. If we are concerned about affecting inequality, rather than mouthing platitudes, then the minimum wage seems a reasonable place to start.

Digression: Since I’m teaching the Card Krueger paper in my econometrics course, I have been reading the rebuttals and replies. In the Card-Krueger reply to the Neumark and Wascher paper, they discuss the latter’s use of an “interesting” dataset — with markedly different attributes from other data sets in use — originally compiled by the “Employment Policies Institute”. I find it remarkable how influential this think tank has been — see discussion here; link to IRS form 990 for interesting reading) (it gives “cozy” industry ties a new meaning). One of the most recent takes on the employment impact, accounting for spillover effects, is in Dube, Lester and Reich (REStat, 2010).

Update, 6:40PM Pacific: Neumark provides a critique of the DLR and other papers, in an Employment Policies Institute document (January 2013).

The Employment Policies Institute has an interesting blogpost observing that some of the 600 or so economists who signed the petition in favor of raising the minimum wage are “Specifically, at least 40 of the signers are specialists in or have done considerable work in the economics of Marxism or Socialism, or are affiliated with the “radical” study of economics.” (Full disclosure: I signed the petition, and I once took a course on Socialism from the famous leftist(!!!) Adam Ulam, so I must be on their list as well).

Update, 10:20PM Pacific: Professor Reich directs me to this paper which addresses critiques leveled by Neumark et al. (see Appendix B).

Update, 3/30 2:15PM Pacific: Recent work documenting the increasing target efficiency of the minimum wage, in this paper:

In this study I show that the target efficiency of the minimum wage improved between 1999 and 2013. In 1999-2001, 15.3% of the minimum wage benefits went to workers in poor families. By 2011-2013 this figure had risen to 18.8%. Nearly two-thirds of the improvement in target efficiency occurred during pre-recession years (1999-2001 to 2005-2007), and the balance of the improvement occurred since the onset of the Great Recession. The improvement in target efficiency during pre-recession years was entirely due to an increase in the share of minimum wage workers in poor families. Decreased income among near-poor minimum wage workers drove the majority of the increase in the share of minimum wage workers in poverty. Reduced teen employment, increased teen wages (relative to the minimum wage), and increased employment among poor low-skilled 20-29 year-olds also contributed.

Author: "Menzie Chinn" Tags: "labor market"
Send by mail Print  Save  Delicious 
Date: Tuesday, 25 Mar 2014 16:05

From Reuters:

Russia is at risk of recession as investors pull money out of the country, with growth likely to evaporate if capital outflows reach $100 billion, the head of its largest bank, state-owned Sberbank, said on Monday.

Capital has fled Russia, with stocks and the rouble sliding following Moscow’s seizure of Crimea from Ukraine and western sanctions against Russian individuals. Analysts at Goldman Sachs recently predicted capital outflows for this year could reach $130 billion, or double 2013 levels.

And from K. Hille and R. McGregor in the FT, assessments from Russian government officials:

The Russian government is braced for the country’s capital outflows to soar to $70bn in the first three months of the year as investors seek cover from the fallout of President Vladimir Putin’s Ukrainian land grab.


…That would exceed the $63bn that flowed out of the country in the whole of last year and is higher than the $50bn figure mooted by Mr Putin’s economic adviser Alexei Kudrin 10 days ago.

It appears that Russia is more vulnerable to sanctions, and the threat of sanctions, than some observers believed (see discussion here). Figure 1 places in context the developments in the external balances, including net capital inflows, through 2013.

russiapix4

Figure 1 from Y. Lissovolik, A. Zaigrin, “Russia: macro implications of increased geopolitical risk,” Emerging Markets Monthly (Deutsche Bank, 13 March 2014) [not online].

The figures in the graph are not annualized, so net outflows of $100 billion in Q1 would dwarf those recorded in earlier quarters. There is a strong correlation between capital inflows and fixed investment (and the investment contribution to GDP growth in 2013Q4 was already zero). Figure 7 illustrates this correlation.

russiacapgrowth

Figure 7 from Y. Lissovolik, A. Zaigrin, “Russia: macro implications of increased geopolitical risk,” Emerging Markets Monthly (Deutsche Bank, 13 March 2014) [not online].

From my own perspective, I was surprised that growth forecasts had not already been marked down to zero and below. After all, the policy rate has already been raised by 1.5 ppts to 7% as of the beginning of March. [1] A standard Mundell-Fleming analysis in the absence of balance-sheet/foreign exchange concerns suggests this measure is contractionary; with foreign currency denominated debt, the interest rate defense might make sense as the lesser of two evils, but doesn’t take away from the fact that raising the policy rate is contractionary. Admittedly, with y/y inflation at 6.2%, the increase makes real interest rates less negative, rather than more positive. In addition, with an underdeveloped and fragmented financial system, the link from the policy rate to lending rates is probably weak.

It’s of interest that these developments are occurring against a backdrop of already priced in monetary tightening in the United States. Should economic conditions induce a revision toward more rapid tightening, or general risk appetite decline, pressure on Russian external balances — and foreign exchange reserves — will only increase.

Update, 7:45AM 3/26: The World Bank has released a new assessment of the Russian Economy.

Given the higher risk environment — since political uncertainties around the Crimea crisis in early March 2014 led to an increase in market volatility, the World Bank developed two alternative scenarios for Russia’s 2014-2015 growth outlook.


  • The low-risk scenario assumes a limited and short-lived impact of the Crimea crisis and projects growth to slow to 1.1 percent in 2014 and slightly picking up to 1.3 percent in 2015.
  • The high-risk scenario assumes a more severe shock to economic and investment activities if the geopolitical situation worsens and projects a contraction in output of 1.8 percent for 2014.

Author: "Menzie Chinn" Tags: "international"
Send by mail Print  Save  Delicious 
Date: Sunday, 23 Mar 2014 14:01

Here are some graphs of economic data that illustrate some interesting trends.

Atif Mian and Amir Sufi note that U.S. median family income grew with productivity in the forty years following World War II but has since fallen behind.

productivity_median_income_mar14

And Martin Neil Baily and Barry Bosworth note that while U.S. manufacturing output has grown at the same pace as the rest of the economy, U.S. manufacturing employment has not.

manufacturing_mar14

In terms of monetary policy, the Federal Reserve Bank of Atlanta is now regularly reporting the Wu-Xia shadow fed funds rate. The latest estimate indicates that a return to higher interest rates was farther away than ever as of the end of last month. For a description of what this series tells us, see my discussion here.

 

On world oil markets, the Wall Street Journal observes that Iraqi oil production is at its highest level in 30 years.

Iraqi oil production in millions of barrels per day.  Source: Wall Street Journal.

Iraqi oil production in millions of barrels per day. Source: Wall Street Journal.

But Peak Oil Barrel notes that’s more than outweighed by recent turmoil in Libya.

 

Author: "James_Hamilton" Tags: "economic indicators, employment, energy,..."
Send by mail Print  Save  Delicious 
Date: Friday, 21 Mar 2014 21:03

How vulnerable to sanctions — official and market-driven — is Russia?

The Short-Run Economic Fallout of Russian Intervention

In the wake of unilateral moves by the Russian Federation, the US has imposed sanctions on Russia. [1] [2] [3] These measures have added pressure on Russian markets above and beyond those imposed by the international financial markets: the stock market is down, the currency has weakened, and sovereign bond yields are up. (Russian officials concede impact [4]

russiapix1

Figure 1: MICEX index, April 2013-April 2014. Source: Bloomberg.

russiapix2

Figure 2: USDRUB, April 2013-April 2014. Up is a depreciation of the Russian rouble. Source: Bloomberg.

russiapix3

Figure 3: Ten year Russian sovereign yield, 2007-2014. Source: TradingEconomics.

Elevated risk showed up in capital outflows, even as of the end of 2013.

russiapix4

Figure 1 from Y. Lissovolik, A. Zaigrin, “Russia: macro implications of increased geopolitical risk,” Emerging Markets Monthly (Deutsche Bank, 13 March 2014) [not online].

The Impact of Sanctions

The question, then, is whether economic sanctions will have the desired effect. Much of the conventional wisdom regarding sanction efficacy can be found in the work of Hufbauer, Schott, Elliott, and Oegg, who draw their conclusions from an analysis of a large set of sanction episodes. Successful sanctions (about a third of cases) imposed costs in excess of 2 percent of GDP, while unsuccessful sanctions imposed costs of about a percent of GDP. Much depends on what is the “desired effect”; I’ll return to this point at the end.

In most — or at least many — previous episodes, the sanctions worked more through the trade channel. Most of the impact seen thus far in the Russian episode has been along the financial dimension; it remains to be seen whether the financial impact will be greater, or perhaps more focused, on relevant target groups. From a memo by Steve Englander/Citi yesterday (“International finance as continuation of war by other means”):

The Russia/Ukraine crisis may be the first major political conflict that is played out in international financial markets. The difference between this and standard imposition of sanctions is that both sides have some options that can inflict damage on the other side. Normally sanctions involve a big group of countries on one side and a vulnerable target on the other so the vulnerability is very asymmetric. Small countries can occasionally expropriate big country assets but in most cases that ostracizes the small country from the international financial community.

Obviously, the impact of the financial sanctions depends on the credibility of the threat of imposition. For Europe, where natural gas supplies and bank loan exposure loom large, the threat is less credible. From G. Moec, M. Stringa “Who is exposed to Russia?” Deutsche Bank, 20 March 2014:

In absolute terms, French banks are – by far – the most exposed, with USD 51bn claims (loans and securities) over Russia in Q3 2013.


However, these claims must be compared with total bank assets in each country to get a sense of the systemic impact that an asset freeze or deterioration in the creditworthiness of Russian assets could have. Austria then comes out with the highest ratio (1.4%), followed by the Netherlands and Italy.
Still, at 0.5% of total bank assets, we would regard the French ratio as significant, especially in a period of pressure on capital ratios.

The United States, however, has less direct bank exposure, and doesn’t depend heavily on Russian exports, energy or otherwise. In that respect, the threat of additional US sanctions are credible.

In addition, while Russia seems fairly resilient along some dimensions — high foreign exchange reserves to short term external debt, a relatively small budget deficit (see IMF Article IV report from October here; and the Economist’s capital freeze index) — there are some clear vulnerabilities. First and foremost is the slowing growth rate.

Prior to the Crimean crisis, the IMF was already forecasting modest growth in 2014 (see this post). The IMF has not released new estimates to my knowledge, but like others, they are probably marking down growth estimates due to reduced capital inflows and heightened uncertainty. Citi has reduced its 2014 forecast from 2.6% to 1% (Weisenthal/BI)

russiagdpgrowth

Figure 4: Russian q/q annualized GDP growth, in 2000 prices. Source: OECD via FRED, and author’s calculations.

Other macroeconomic indicators are suggesting a slowdown.

russiapix5

Figure 5: Russia: Key Economic Indicators. Source: “Russia,” Emerging Markets Monthly (Deutsche Bank, 13 March 2014) [not online].

Second, the Russian government remains highly dependent on oil export revenue. In the absence of oil revenues, the government deficit is on the order of 10% of GDP. A big drop in oil prices would be extremely painful.

brentoil

Figure 6: Price of crude oil,Brent (blue), and futures as of 3/31/2014 (red). Source: EIA via FRED and Barchart.com.

While there is little the US could do to affect world oil prices now by way of production measures (see Jim’s assessment here), any economic slowdown which decreases demand for (price inelastic) oil would surely put pressure on Russia. And oil conservation measures in the US would have an impact, although I am sure that all the folks arguing for more vigorous diplomatic and military measures (e.g., [5]) would not favor policies such as a Strategic Petroleum Reserve (SPRO) release, or a higher gasoline tax.

Hence, the past might not be a completely adequate guide to assessing the sensitivity to sanctions (and spillover effects), given the fragile nature of growth, and the localization of induced pain to groups important to the ruling elite.

Defining Success

What is the desired effect of the sanctions? If it is to restrain Russian proclivities for intervention, the costs that can be imposed might be sufficient. In fact the Institute for International Finance predicts a severe recession in Russia — a 5% cumulative drop in GDP — should there be a secession of Eastern Ukraine. [6] If success entails eliciting a Russian withdrawal from the Crimea, that might be a tougher task — as argued by Dan Drezner.

Author: "Menzie Chinn" Tags: "international"
Send by mail Print  Save  Delicious 
Date: Thursday, 20 Mar 2014 21:00

Wisconsin continues to lag the other regional economies

The Wisconsin Department of Revenue today released its Winter Wisconsin Economic Outlook. The forecast indicates a shortfall relative to Governor Walker’s promise to create 250,000 new private sector jobs.

wimar14_1

Figure 1: Nonfarm private payroll employment for Wisconsin (blue), quadratic match interpolation of quarterly forecast (red), and linear trend for Walker’s target of 250,000 new jobs (black). Source: BLS and Wisconsin Economic Outlook (March 2014).

In order to hit the target, 12,275 jobs per month would need to be created each month for the next year; the mean job creation rate since 2011M02 has been 2,892, with standard deviation of 3,817. Achievement of the Governor’s goal would therefore seem “unlikely”.

How about Wisconsin’s relative performance? As noted in an earlier post, Wisconsin private nonfarm payroll employment is flat going from December to January (the earlier post used state data, but the BLS released series are numerically indistinguishable). Of course, one should not put too much weight on one month’s numbers. It’s interesting to observe that while the 3 month annualized growth rate of Wisconsin private employment (Governor Walker’s preferred measure) is 1.5%, Minnesota’s is 2.4%.

Figure 2 depicts overall nonfarm payroll employment (which is more favorable to Wisconsin of late).

wimar14_2

Figure 2: Log nonfarm payroll employment for Wisconsin (red), Minnesota (blue) and the US, all seasonally adjusted, 2011M01=0. Source: BLS, and author’s calculations.

Minnesota’s cumulative growth rate since 2011M01 relative to Wisconsin is 1.6% as of January 2014. Wisconsin’s performance relative to Minnesota is not unique. Using a broader measure of economic activity, the Philadelphia Fed’s coincident indices, one finds that Wisconsin is outpaced by all her neighbors. (see also [1])

wimar14_3

Figure 3: Log coincident indices for MN (blue), WI (red), IL (green), IA (purple), IN (teal), MI (olive), and US (black). Source: Philadelphia Fed, and author’s calculations.

Indiana and Michigan outpace the Nation, but these states experienced a much deeper drop during the Great Recession.

The cumulative growth gap since 2011M01 between Minnesota and Wisconsin is 1.8% (log terms); using the Philadelphia Fed leading indices as of the December release, the gap will shrink to 1.3% by June 2014 (as pointed out by some observors). However, with upwardly revised employment figures for Minnesota (more so than for Wisconsin), my guess is that the January release of the leading indices (slated for March 24) will imply faster Minnesota growth, so that the gap does not shrink as much as suggested by the current forecast (if at all). Furthermore, using the December forecasts, the US-WI gap will widen from 1.7% to 2.5%. Hence, Wisconsin also lags the Nation.

Author: "Menzie Chinn" Tags: "Wisconsin"
Send by mail Print  Save  Delicious 
Date: Tuesday, 18 Mar 2014 13:14

Total vehicle miles driven in the United States have not re-attained pre-recession peaks.

Upon inspecting Figure 1 in this post, reader Patrick R. Sullivan decries the current level of auto production. His reasoning?

Looking at Figure 1, I’d say that sales of cars are pathetic. Only back to the level of 2005. And that after years of far below normal sales figures, which should have resulted in pent-up demand. This is nothing to brag about for this economy.

In a similar vein, Bruce Hall writes:

Just curious how much the present demand still represents “pent-up demand.” With domestic production running at 300-350K per month and then dropping like a rock for three years, it would seem that just getting back to the old “normal” still represents “behind the curve.” There are plenty of anecdotal accounts of small fleets being run into the ground because owners are afraid to spend money on replacement vehicles.


I’d suggest there is still a story behind the chart that is not being told.

I agree that there is more aspects to consider. First, I think it is useful, when considering what is an appropriate level of production, to take into account what is the end-use of the product. In the case of motor vehicles, it’s miles driven. And here, the trend has been sideways — after a drop in 2009.

vmd_percap

Figure 1: Total motor vehicle miles driven, in millions, 12 month moving average (blue, left scale), and vehicle miles driven per person, in thousands (red, right scale). NBER defined recession dates shaded gray. Source: Federal Highway Administration and BLS via FRED, and author’s calculations.

What is more interesting to me is the decline in per capita miles driven. This variable has dropped 9% (log terms) since 2005M08 (vehicle miles have fallen 2.1% since the miles peak of 2007M09).

Part of the reason for the reversed trend in per capita miles driven is surely elevated gasoline prices; however, this seems unlikely to be the only factor.

traffic_gas

Figure 2: Log relative price of gasoline to core CPI (teal, left scale), and log vehicle miles driven per person, in thousands (red, right scale). NBER defined recession dates shaded gray. Source: Federal Highway Administration and BLS via FRED, and author’s calculations.

Calculated Risk observes:

[G]asoline prices are just part of the story. The lack of growth in miles driven over the last 6 years is probably also due to the lingering effects of the great recession (high unemployment rate and lack of wage growth), the aging of the overall population (over 55 drivers drive fewer miles) and changing driving habits of young drivers.

Update, 3/21 4PM Pacific: More from SSTI, here.

Author: "Menzie Chinn" Tags: "Uncategorized"
Send by mail Print  Save  Delicious 
Next page
» You can also retrieve older items : Read
» © All content and copyrights belong to their respective authors.«
» © FeedShow - Online RSS Feeds Reader