CANNOT UPLOAD CHARTS AND IMAGES
II Rules, Discretionary Authorities and Slow Productivity Growth. The Bureau of Labor Statistics (BLS) of the Department of Labor provides the quarterly report on productivity and costs. The operational definition of productivity used by the BLS is (http://www.bls.gov/news.release/pdf/prod2.pdf 1): “Labor productivity, or output per hour, is calculated by dividing an index of real output by an index of hours worked of all persons, including employees, proprietors, and unpaid family workers.” The BLS has revised the estimates for productivity and unit costs. Table II-1 provides the third estimate for IIQ2017 and revision of the estimates for IQ2017 and IVQ2016 together with data for nonfarm business sector productivity and unit labor costs in seasonally adjusted annual equivalent (SAAE) rate and the percentage change from the same quarter a year earlier. Reflecting increase in output at 4.0 percent and increase at 2.5 percent in hours worked, nonfarm business sector labor productivity increased at the SAAE rate of 1.5 percent in IIQ2017, as shown in column 2 “IIQ2017 SAEE.” The increase of labor productivity from IIQ2016 to IIQ2017 was 1.3 percent, reflecting increases in output of 2.8 percent and of hours worked of 1.5 percent, as shown in column 3 “IIQ2017 YoY.” Hours worked increased from 1.0 percent in IVQ2016 at SAAE to 1.6 percent in IQ2017 and increased to 2.5 percent in IIQ2017 while output growth decreased from 2.2 percent in IVQ2016 at SAAE to 1.8 percent in IQ2017, increasing to 4.0 percent in IIQ2017. The BLS defines unit labor costs as (http://www.bls.gov/news.release/pdf/prod2.pdf 1): “BLS calculates unit labor costs as the ratio of hourly compensation to labor productivity. Increases in hourly compensation tend to increase unit labor costs and increases in output per hour tend to reduce them.” Unit labor costs increased at the SAAE rate of 0.2 percent in IIQ2017 and decreased 0.2 percent in IIQ2017 relative to IIQ2016. Hourly compensation increased at the SAAE rate of 1.8 percent in IIQ2017, which deflating by the estimated consumer price increase SAAE rate in IIQ2017 results in increase of real hourly compensation at 2.1 percent. Real hourly compensation decreased 0.8 percent in IIQ2017 relative to IIQ2016.
Table II-1, US, Nonfarm Business Sector Productivity and Costs %
| II 2017 SAAE | II 2017 YOY | IQ | IQ | IVQ 2016 SAAE | IVQ 2016 YoY | |
| Productivity | 1.5 | 1.3 | 0.1 | 1.2 | 1.3 | 0.8 |
| Output | 4.0 | 2.8 | 1.8 | 2.5 | 2.2 | 2.1 |
| Hours | 2.5 | 1.5 | 1.6 | 1.3 | 1.0 | 1.2 |
| Hourly | 1.8 | 1.1 | 4.9 | 1.9 | -4.5 | -0.3 |
| Real Hourly Comp. | 2.1 | -0.8 | 1.7 | -0.7 | -7.3 | -2.1 |
| Unit Labor Costs | 0.2 | -0.2 | 4.8 | 0.7 | -5.7 | -1.2 |
| Unit Nonlabor Payments | 1.0 | 3.3 | -3.9 | 3.1 | 13.9 | 5.6 |
| Implicit Price Deflator | 0.6 | 1.3 | 0.9 | 1.7 | 2.3 | 1.7 |
Notes: SAAE: seasonally adjusted annual equivalent; Comp.: compensation; YoY: Quarter on Same Quarter Year Earlier
The analysis by Kydland (http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2004/kydland-bio.html) and Prescott (http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2004/prescott-bio.html) (1977, 447-80, equation 5) uses the “expectation augmented” Phillips curve with the natural rate of unemployment of Friedman (1968) and Phelps (1968), which in the notation of Barro and Gordon (1983, 592, equation 1) is:
Ut = Unt – α(πt – πe) α > 0 (1)
Where Ut is the rate of unemployment at current time t, Unt is the natural rate of unemployment, πt is the current rate of inflation and πe is the expected rate of inflation by economic agents based on current information. Equation (1) expresses unemployment net of the natural rate of unemployment as a decreasing function of the gap between actual and expected rates of inflation. The system is completed by a social objective function, W, depending on inflation, π, and unemployment, U:
W = W(πt, Ut) (2)
The policymaker maximizes the preferences of the public, (2), subject to the constraint of the tradeoff of inflation and unemployment, (1). The total differential of W set equal to zero provides an indifference map in the Cartesian plane with ordered pairs (πt, Ut - Un) such that the consistent equilibrium is found at the tangency of an indifference curve and the Phillips curve in (1). The indifference curves are concave to the origin. The consistent policy is not optimal. Policymakers without discretionary powers following a rule of price stability would attain equilibrium with unemployment not higher than with the consistent policy. The optimal outcome is obtained by the rule of price stability, or zero inflation, and no more unemployment than under the consistent policy with nonzero inflation and the same unemployment. Taylor (1998LB) attributes the sustained boom of the US economy after the stagflation of the 1970s to following a monetary policy rule instead of discretion (see Taylor 1993, 1999). Professor John B. Taylor (2014Jul15, 2014Jun26) building on advanced research (Taylor 2007, 2008Nov, 2009, 2012FP, 2012Mar27, 2012Mar28, 2012JMCB, 2015, 2012 Oct 25; 2013Oct28, 2014 Jan01, 2014Jan3, 2014Jun26, 2014Jul15, 2015, 2016Dec7, 2016Dec20 http://www.johnbtaylor.com/) finds that a monetary policy rule would function best in promoting an environment of low inflation and strong economic growth with stability of financial markets. There is strong case for using rules instead of discretionary authorities in monetary policy (http://cmpassocregulationblog.blogspot.com/2017/01/rules-versus-discretionary-authorities.html and earlier http://cmpassocregulationblog.blogspot.com/2012/06/rules-versus-discretionary-authorities.html). It is not uncommon for effects of regulation differing from those intended by policy. Professors Edward C. Prescott and Lee E. Ohanian (2014Feb), writing on “US productivity growth has taken a dive,” on Feb 3, 2014, published in the Wall Street Journal (http://online.wsj.com/news/articles/SB10001424052702303942404579362462611843696?KEYWORDS=Prescott), argue that impressive productivity growth over the long-term constructed US prosperity and wellbeing. Prescott and Ohanian (2014Feb) measure US productivity growth at 2.5 percent per year since 1948. Average US productivity growth has been only 1.1 percent since 2011. Prescott and Ohanian (2014Feb) argue that living standards in the US increased at 28 percent in a decade but with current slow growth of productivity will only increase 12 percent by 2024. There may be collateral effects on productivity growth from policy design similar to those in Kydland and Prescott (1977). Professor Edward P. Lazear (2017Feb27), writing in the Wall Street Journal, on Feb 27, 2017 (https://www.wsj.com/articles/how-trump-can-hit-3-growthmaybe-1488239746), finds that productivity growth was 7 percent between 2009 and 2016 at annual equivalent 1 percent. Lazear measures productivity growth at 2.3 percent per year from 2001 to 2008. The Bureau of Labor Statistics important report on productivity and costs released on Aug 9, 2017 (http://www.bls.gov/lpc/) supports the argument of decline of productivity growth in the US analyzed by Prescott and Ohanian (2014Feb) and Lazear (2017Feb27). Table II-2 provides the annual percentage changes of productivity, real hourly compensation and unit labor costs for the entire economic cycle from 2007 to 2016. The estimates incorporate the yearly revision of the US national accounts (http://www.bea.gov/national/an1.htm#2016annualupdate). The data confirm the argument of Prescott and Ohanian (2014Feb) and Lazear (2017Feb27): productivity increased cumulatively 3.5 percent from 2011 to 2016 at the average annual rate of 0.6 percent. The situation is direr by excluding growth of 0.9 percent in 2012, which leaves an average of 0.5 percent for 2011-2016. Average productivity growth for the entire economic cycle from 2007 to 2016 is only 1.2 percent. The argument by Prescott and Ohanian (2014Feb) is proper in choosing the tail of the business cycle because the increase in productivity in 2009 of 3.1 percent and 3.3 percent in 2010 consisted of reducing labor hours.
Table II-2, US, Revised Nonfarm Business Sector Productivity and Costs Annual Average, ∆% Annual Average
| 2016 ∆% | 2015 ∆% | 2014 ∆% | 2013 ∆% | 2012 ∆% | 2011 ∆% | |
| Productivity | -0.1 | 1.3 | 1.0 | 0.3 | 0.9 | 0.1 |
| Real Hourly Compensation | -0.2 | 3.0 | 1.3 | -0.3 | 0.6 | -1.0 |
| Unit Labor Costs | 1.1 | 1.8 | 1.9 | 0.9 | 1.7 | 2.1 |
| 2010 ∆% | 2009 ∆% | 2008 ∆% | 2007∆% | |
| Productivity | 3.3 | 3.1 | 0.8 | 1.6 |
| Real Hourly Compensation | 0.3 | 1.4 | -1.0 | 1.4 |
| Unit Labor Costs | -1.3 | -2.0 | 2.0 | 2.7 |
Source: US Bureau of Labor Statistics
Productivity jumped in the recovery after the recession from Mar IQ2001 to Nov IVQ2001 (http://www.nber.org/cycles.html). Table II-3 provides quarter on quarter and annual percentage changes in nonfarm business output per hour, or productivity, from 1999 to 2017. The annual average jumped from 2.7 percent in 2001 to 4.4 percent in 2002. Nonfarm business productivity increased at the SAAE rate of 9.3 percent in the first quarter after the recession in IQ2002. Productivity increases decline later in the expansion period. Productivity increases were mediocre during the recession from Dec IVQ2007 to Jun IIIQ2009 (http://www.nber.org/cycles.html) and increased during the first phase of expansion from IIQ2009 to IQ2010, trended lower and collapsed in 2011 and 2012 with sporadic jumps and declines. Productivity increased at 4.4 percent in IVQ2013 and contracted at 3.4 percent in IQ2014. Productivity increased at 2.4 percent in IIQ2014 and at 4.4 percent in IIIQ2014. Productivity contracted at 1.8 percent in IVQ2014 and increased at 2.7 percent in IQ2015. Productivity grew at 1.3 percent in IIQ2015 and increased at 1.3 percent in IIIQ2015. Productivity contracted at 2.6 percent in IVQ2015 and contracted at 1.2 percent in IQ2016. Productivity increased at 0.8 percent in IIQ2016 and expanded at 2.5 percent in IIIQ2016. Productivity grew at 1.3 percent in IVQ2016 and increased at 0.1 percent in IQ2017. Productivity increased at 1.5 percent in IIQ2017.
| Year | Qtr1 | Qtr2 | Qtr3 | Qtr4 | Annual |
| 1999 | 4.5 | 1.2 | 3.6 | 6.6 | 3.7 |
| 2000 | -1.9 | 8.2 | -0.2 | 4.1 | 3.0 |
| 2001 | -1.6 | 7.0 | 2.1 | 5.2 | 2.7 |
| 2002 | 9.3 | 0.3 | 3.1 | -0.7 | 4.4 |
| 2003 | 4.2 | 5.5 | 9.0 | 3.9 | 3.7 |
| 2004 | -0.1 | 3.8 | 1.4 | 1.3 | 3.1 |
| 2005 | 4.5 | -0.4 | 3.0 | 0.1 | 2.1 |
| 2006 | 2.4 | -0.3 | -1.8 | 3.0 | 0.9 |
| 2007 | 0.4 | 2.5 | 4.9 | 1.7 | 1.6 |
| 2008 | -3.8 | 4.0 | 1.0 | -2.5 | 0.8 |
| 2009 | 3.1 | 7.9 | 5.9 | 4.9 | 3.1 |
| 2010 | 2.1 | 1.4 | 2.0 | 1.6 | 3.3 |
| 2011 | -3.3 | 1.3 | -0.7 | 2.8 | 0.1 |
| 2012 | 0.6 | 2.3 | -0.7 | -1.8 | 0.9 |
| 2013 | 0.9 | -0.7 | 1.6 | 4.4 | 0.3 |
| 2014 | -3.4 | 2.4 | 4.4 | -1.8 | 1.0 |
| 2015 | 2.7 | 1.3 | 1.3 | -2.6 | 1.3 |
| 2016 | -1.2 | 0.8 | 2.5 | 1.3 | -0.1 |
| 2017 | 0.1 | 1.5 |
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Chart II-1 of the Bureau of Labor Statistics (BLS) provides SAAE rates of nonfarm business productivity from 1999 to 2017. There is a clear pattern in both episodes of economic cycles in 2001 and 2007 of rapid expansion of productivity in the transition from contraction to expansion followed by more subdued productivity expansion. Part of the explanation is the reduction in labor utilization resulting from adjustment of business to the sudden shock of collapse of revenue. Productivity rose briefly in the expansion after 2009 but then collapsed and moved to negative change with some positive changes recently at lower rates. Contractions in the cycle from 2007 to 2016 have been more frequent and sharper.
Chart II-1, US, Nonfarm Business Output per Hour, Percent Change from Prior Quarter at Annual Rate, 1999-2017
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Percentage changes from prior quarter at SAAE rates and annual average percentage changes of nonfarm business unit labor costs are provided in Table II-4. Unit labor costs fell during the contractions with continuing negative percentage changes in the early phases of the recovery. Weak labor markets partly explain the decline in unit labor costs. As the economy moves toward full employment, labor markets tighten with increase in unit labor costs. The expansion beginning in IIIQ2009 has been characterized by high unemployment and underemployment. Table II-4 shows continuing subdued increases in unit labor costs in 2011 but with increase at 8.9 percent in IQ2012 followed by decrease at 0.1 percent in IIQ2012, increase at 1.1 percent in IIIQ2012 and increase at 13.2 percent in IVQ2012. Unit labor costs decreased at 9.7 percent in IQ2013 and increased at 6.5 percent in IIQ2013. Unit labor costs decreased at 0.5 percent in IIIQ2013 and decreased at 1.9 percent in IVQ2013. Unit labor costs increased at 10.5 percent in IQ2014 and at minus 4.7 percent in IIQ2014. Unit labor costs decreased at 1.2 percent in IIIQ2014 and increased at 7.1 percent in IVQ2014. Unit labor costs increased at 0.6 percent in IQ2015 and increased at 2.3 percent in IIQ2015. Unit labor costs decreased at 0.2 percent in IIIQ2015 and increased at 7.2 percent in IVQ2015. Unit labor costs decreased at 2.7 percent in IQ2016 and increased at 3.9 percent in IIQ2016. Unit labor costs increased at 0.1 percent in IIIQ2016 and decreased at 5.7 percent in IVQ2016. Unit labor costs increased at 4.8 percent in IQ2017 and increased at 0.2 percent in IIQ2017.
Table II-4, US, Nonfarm Business Unit Labor Costs, Percent Change from Prior Quarter at Annual Rate 1999-2017
| Year | Qtr1 | Qtr2 | Qtr3 | Qtr4 | Annual |
| 1999 | 2.9 | 0.3 | 0.0 | 1.6 | 0.9 |
| 2000 | 17.4 | -6.8 | 8.2 | -1.7 | 4.0 |
| 2001 | 11.4 | -5.4 | -1.7 | -1.4 | 1.6 |
| 2002 | -6.6 | 3.3 | -1.1 | 1.7 | -2.0 |
| 2003 | -1.5 | 1.6 | -2.6 | 1.5 | 0.1 |
| 2004 | -0.5 | 3.9 | 5.6 | 0.5 | 1.4 |
| 2005 | -1.3 | 2.6 | 2.0 | 2.3 | 1.6 |
| 2006 | 6.1 | 0.5 | 2.3 | 4.0 | 3.0 |
| 2007 | 9.8 | -2.7 | -3.2 | 2.6 | 2.7 |
| 2008 | 8.3 | -3.6 | 2.4 | 7.1 | 2.0 |
| 2009 | -12.3 | 2.1 | -3.0 | -2.3 | -2.0 |
| 2010 | -4.8 | 3.2 | -0.2 | 0.2 | -1.3 |
| 2011 | 11.0 | -3.5 | 3.3 | -7.7 | 2.1 |
| 2012 | 8.9 | -0.1 | 1.1 | 13.2 | 1.7 |
| 2013 | -9.7 | 6.5 | -0.5 | -1.9 | 0.9 |
| 2014 | 10.5 | -4.7 | -1.2 | 7.1 | 1.9 |
| 2015 | 0.6 | 2.3 | -0.2 | 7.2 | 1.8 |
| 2016 | -2.7 | 3.9 | 0.1 | -5.7 | 1.1 |
| 2017 | 4.8 | 0.2 |
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Chart II-2 provides change of unit labor costs at SAAE from 1999 to 2017. There are multiple oscillations recently with negative changes alternating with positive changes.
Chart II-2, US, Nonfarm Business Unit Labor Costs, Percent Change from Prior Quarter at Annual Rate 1999-2017
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Table II-5 provides percentage change from prior quarter at annual rates for nonfarm business real hourly worker compensation. The expansion after the contraction of 2001 was followed by strong recovery of real hourly compensation. Real hourly compensation increased at the rate of 2.9 percent in IQ2011 but fell at annual rates of 6.5 percent in IIQ2011 and 6.8 percent in IVQ2011. Real hourly compensation increased at 7.1 percent in IQ2012, increasing at 1.4 percent in IIQ2012, declining at 1.4 percent in IIIQ2012 and increasing at 8.3 percent in IVQ2012. Real hourly compensation fell at 1.0 percent in 2011 and increased at 0.6 percent in 2012. Real hourly compensation fell at 10.4 percent in IQ2013 and increased at 6.3 percent in IIQ2013, falling at 1.1 percent in IIIQ2013. Real hourly compensation increased at 0.9 percent in IVQ2013 and at 4.1 percent in IQ2014. Real hourly compensation decreased at 4.2 percent in IIQ2014. Real hourly compensation increased at 2.1 percent in IIIQ2014. The annual rate of increase of real hourly compensation for 2013 is minus 0.3 percent. Real hourly compensation increased at 5.8 percent in IVQ2014. The annual rate of increase of real hourly compensation in 2014 is 1.3 percent. Real hourly compensation increased at 6.0 percent in IQ2015 and increased at 1.2 percent in IIQ2015. Real hourly compensation decreased at 0.4 percent in IIIQ2015 and increased at 4.1 percent in IVQ2015. Real hourly compensation increased at 3.0 percent in 2015. Real hourly compensation decreased at 3.9 percent in IQ2016 and increased at 2.4 percent in IIQ2016. Real hourly compensation increased at 0.7 percent in IIIQ2016 and decreased at 7.3 percent in IVQ2016. Real hourly compensation decreased 0.2 percent in 2016. Real hourly compensation increased at 1.7 percent in IQ2017 and increased at 2.1 percent in IIQ2017.
Table II-5, Nonfarm Business Real Hourly Compensation, Percent Change from Prior Quarter at Annual Rate 1999-2016
| Year | Qtr1 | Qtr2 | Qtr3 | Qtr4 | Annual |
| 1999 | 6.0 | -1.5 | 0.5 | 5.1 | 2.5 |
| 2000 | 10.6 | -2.2 | 4.1 | -0.5 | 3.5 |
| 2001 | 5.4 | -1.7 | -0.7 | 4.1 | 1.4 |
| 2002 | 0.7 | 0.3 | -0.2 | -1.3 | 0.7 |
| 2003 | -1.5 | 7.8 | 3.0 | 3.9 | 1.5 |
| 2004 | -3.8 | 4.5 | 4.5 | -2.5 | 1.8 |
| 2005 | 1.1 | -0.6 | -1.0 | -1.3 | 0.3 |
| 2006 | 6.4 | -3.5 | -3.1 | 8.8 | 0.6 |
| 2007 | 6.0 | -4.7 | -1.0 | -0.5 | 1.4 |
| 2008 | -0.3 | -4.7 | -2.6 | 14.6 | -1.0 |
| 2009 | -7.1 | 7.8 | -0.7 | -0.6 | 1.4 |
| 2010 | -3.4 | 4.8 | 0.6 | -1.4 | 0.3 |
| 2011 | 2.9 | -6.5 | -0.1 | -6.8 | -1.0 |
| 2012 | 7.1 | 1.4 | -1.4 | 8.3 | 0.6 |
| 2013 | -10.4 | 6.3 | -1.1 | 0.9 | -0.3 |
| 2014 | 4.1 | -4.2 | 2.1 | 5.8 | 1.3 |
| 2015 | 6.0 | 1.2 | -0.4 | 4.1 | 3.0 |
| 2016 | -3.9 | 2.4 | 0.7 | -7.3 | -0.2 |
| 2017 | 1.7 | 2.1 |
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Chart II-3 provides percentage change from prior quarter at annual rate of nonfarm business real hourly compensation. There have been multiple negative percentage quarterly changes in the current cycle since IVQ2007.
Chart II-3, US, Nonfarm Business Real Hourly Compensation, Percent Change from Prior Quarter at Annual Rate 1999-2017
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Chart II-4 provides percentage change of nonfarm business output per hour in a quarter relative to the same quarter a year earlier. As in most series of real output, productivity increased sharply in 2010 but the momentum was lost after 2011 as with the rest of the real economy.
Chart II-4, US, Nonfarm Business Output per Hour, Percent Change from Same Quarter a Year Earlier 1999-2017
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Chart II-5 provides percentage changes of nonfarm business unit labor costs relative to the same quarter a year earlier. Softening of labor markets caused relatively high yearly percentage changes in the recession of 2001 repeated in the recession in 2009. Recovery was strong in 2010 but then weakened.
Chart II-5, US, Nonfarm Business Unit Labor Costs, Percent Change from Same Quarter a Year Earlier 1999-2017
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Chart II-6 provides percentage changes in a quarter relative to the same quarter a year earlier for nonfarm business real hourly compensation. Labor compensation eroded sharply during the recession with brief recovery in 2010 and another fall until recently.
Chart II-6, US, Nonfarm Business Real Hourly Compensation, Percent Change from Same Quarter a Year Earlier 1999-2017
2005=100
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
In the analysis of Hansen (1939, 3) of secular stagnation, economic progress consists of growth of real income per person driven by growth of productivity. The “constituent elements” of economic progress are “(a) inventions, (b) the discovery and development of new territory and new resources, and (c) the growth of population” (Hansen 1939, 3). Secular stagnation originates in decline of population growth and discouragement of inventions. According to Hansen (1939, 2), US population grew by 16 million in the 1920s but grew by one half or about 8 million in the 1930s with forecasts at the time of Hansen’s writing in 1938 of growth of around 5.3 million in the 1940s. Hansen (1939, 2) characterized demography in the US as “a drastic decline in the rate of population growth.” Hansen’s plea was to adapt economic policy to stagnation of population in ensuring full employment. In the analysis of Hansen (1939, 8), population caused half of the growth of US GDP per year. Growth of output per person in the US and Europe was caused by “changes in techniques and to the exploitation of new natural resources.” In this analysis, population caused 60 percent of the growth of capital formation in the US. Declining population growth would reduce growth of capital formation. Residential construction provided an important share of growth of capital formation. Hansen (1939, 12) argues that market power of imperfect competition discourages innovation with prolonged use of obsolete capital equipment. Trade unions would oppose labor-savings innovations. The combination of stagnating and aging population with reduced innovation caused secular stagnation. Hansen (1939, 12) concludes that there is role for public investments to compensate for lack of dynamism of private investment but with tough tax/debt issues.
The current application of Hansen’s (1938, 1939, 1941) proposition argues that secular stagnation occurs because full employment equilibrium can be attained only with negative real interest rates between minus 2 and minus 3 percent. Professor Lawrence H. Summers (2013Nov8) finds that “a set of older ideas that went under the phrase secular stagnation are not profoundly important in understanding Japan’s experience in the 1990s and may not be without relevance to America’s experience today” (emphasis added). Summers (2013Nov8) argues there could be an explanation in “that the short-term real interest rate that was consistent with full employment had fallen to -2% or -3% sometime in the middle of the last decade. Then, even with artificial stimulus to demand coming from all this financial imprudence, you wouldn’t see any excess demand. And even with a relative resumption of normal credit conditions, you’d have a lot of difficulty getting back to full employment.” The US economy could be in a situation where negative real rates of interest with fed funds rates close to zero as determined by the Federal Open Market Committee (FOMC) do not move the economy to full employment or full utilization of productive resources. Summers (2013Oct8) finds need of new thinking on “how we manage an economy in which the zero nominal interest rates is a chronic and systemic inhibitor of economy activity holding our economies back to their potential.”
Former US Treasury Secretary Robert Rubin (2014Jan8) finds three major risks in prolonged unconventional monetary policy of zero interest rates and quantitative easing: (1) incentive of delaying action by political leaders; (2) “financial moral hazard” in inducing excessive exposures pursuing higher yields of risker credit classes; and (3) major risks in exiting unconventional policy. Rubin (2014Jan8) proposes reduction of deficits by structural reforms that could promote recovery by improving confidence of business attained with sound fiscal discipline.
Professor John B. Taylor (2014Jan01, 2014Jan3) provides clear thought on the lack of relevance of Hansen’s contention of secular stagnation to current economic conditions. The application of secular stagnation argues that the economy of the US has attained full-employment equilibrium since around 2000 only with negative real rates of interest of minus 2 to minus 3 percent. At low levels of inflation, the so-called full-employment equilibrium of negative interest rates of minus 2 to minus 3 percent cannot be attained and the economy stagnates. Taylor (2014Jan01) analyzes multiple contradictions with current reality in this application of the theory of secular stagnation:
- Secular stagnation would predict idle capacity, in particular in residential investment when fed fund rates were fixed at 1 percent from Jun 2003 to Jun 2004. Taylor (2014Jan01) finds unemployment at 4.4 percent with house prices jumping 7 percent from 2002 to 2003 and 14 percent from 2004 to 2005 before dropping from 2006 to 2007. GDP prices doubled from 1.7 percent to 3.4 percent when interest rates were low from 2003 to 2005.
- Taylor (2014Jan01, 2014Jan3) finds another contradiction in the application of secular stagnation based on low interest rates because of savings glut and lack of investment opportunities. Taylor (2009) shows that there was no savings glut. The savings rate of the US in the past decade is significantly lower than in the 1980s.
- Taylor (2014Jan01, 2014Jan3) finds another contradiction in the low ratio of investment to GDP currently and reduced investment and hiring by US business firms.
- Taylor (2014Jan01, 2014Jan3) argues that the financial crisis and global recession were caused by weak implementation of existing regulation and departure from rules-based policies.
- Taylor (2014Jan01, 2014Jan3) argues that the recovery from the global recession was constrained by a change in the regime of regulation and fiscal/monetary policies.
The analysis by Kydland (http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2004/kydland-bio.html) and Prescott (http://www.nobelprize.org/nobel_prizes/economic-sciences/laureates/2004/prescott-bio.html) (1977, 447-80, equation 5) uses the “expectation augmented” Phillips curve with the natural rate of unemployment of Friedman (1968) and Phelps (1968), which in the notation of Barro and Gordon (1983, 592, equation 1) is:
Ut = Unt – α(πt – πe) α > 0 (1)
Where Ut is the rate of unemployment at current time t, Unt is the natural rate of unemployment, πt is the current rate of inflation and πe is the expected rate of inflation by economic agents based on current information. Equation (1) expresses unemployment net of the natural rate of unemployment as a decreasing function of the gap between actual and expected rates of inflation. The system is completed by a social objective function, W, depending on inflation, π, and unemployment, U:
W = W(πt, Ut) (2)
The policymaker maximizes the preferences of the public, (2), subject to the constraint of the tradeoff of inflation and unemployment, (1). The total differential of W set equal to zero provides an indifference map in the Cartesian plane with ordered pairs (πt, Ut - Un) such that the consistent equilibrium is found at the tangency of an indifference curve and the Phillips curve in (1). The indifference curves are concave to the origin. The consistent policy is not optimal. Policymakers without discretionary powers following a rule of price stability would attain equilibrium with unemployment not higher than with the consistent policy. The optimal outcome is obtained by the rule of price stability, or zero inflation, and no more unemployment than under the consistent policy with nonzero inflation and the same unemployment. Taylor (1998LB) attributes the sustained boom of the US economy after the stagflation of the 1970s to following a monetary policy rule instead of discretion (see Taylor 1993, 1999). Professor John B. Taylor (2014Jul15, 2014Jun26) building on advanced research (Taylor 2007, 2008Nov, 2009, 2012FP, 2012Mar27, 2012Mar28, 2012JMCB, 2015, 2012 Oct 25; 2013Oct28, 2014 Jan01, 2014Jan3, 2014Jun26, 2014Jul15, 2015, 2016Dec7, 2016Dec20 http://www.johnbtaylor.com/) finds that a monetary policy rule would function best in promoting an environment of low inflation and strong economic growth with stability of financial markets. There is strong case for using rules instead of discretionary authorities in monetary policy (http://cmpassocregulationblog.blogspot.com/2017/01/rules-versus-discretionary-authorities.html and earlier http://cmpassocregulationblog.blogspot.com/2012/06/rules-versus-discretionary-authorities.html). It is not uncommon for effects of regulation differing from those intended by policy. Professors Edward C. Prescott and Lee E. Ohanian (2014Feb), writing on “US productivity growth has taken a dive,” on Feb 3, 2014, published in the Wall Street Journal (http://online.wsj.com/news/articles/SB10001424052702303942404579362462611843696?KEYWORDS=Prescott), argue that impressive productivity growth over the long-term constructed US prosperity and wellbeing. Prescott and Ohanian (2014Feb) measure US productivity growth at 2.5 percent per year since 1948. Average US productivity growth has been only 1.1 percent since 2011. Prescott and Ohanian (2014Feb) argue that living standards in the US increased at 28 percent in a decade but with current slow growth of productivity will only increase 12 percent by 2024. There may be collateral effects on productivity growth from policy design similar to those in Kydland and Prescott (1977). Professor Edward P. Lazear (2017Feb27), writing in the Wall Street Journal, on Feb 27, 2017 (https://www.wsj.com/articles/how-trump-can-hit-3-growthmaybe-1488239746), finds that productivity growth was 7 percent between 2009 and 20016 at annual equivalent 1 percent. Lazear measures productivity growth at 2.3 percent per year from 2001 to 2008. The Bureau of Labor Statistics important report on productivity and costs released on Mar 8, 2017 (http://www.bls.gov/lpc/) supports the argument of decline of productivity growth in the US analyzed by Prescott and Ohanian (2014Feb) and Lazear (2017Feb27). Table II-2 provides the annual percentage changes of productivity, real hourly compensation and unit labor costs for the entire economic cycle from 2007 to 2017. The estimates incorporate the yearly revision of the US national accounts (https://www.bea.gov/national/an1.htm#2017annualupdate). The data confirm the argument of Prescott and Ohanian (2014Feb) and Lazear (2017Feb27): productivity increased cumulatively 3.5 percent from 2011 to 2016 at the average annual rate of 0.6 percent. The situation is direr by excluding growth of 0.9 percent in 2012, which leaves an average of 0.5 percent for 2011-2016. Average productivity growth for the entire economic cycle from 2007 to 2016 is only 1.2 percent. The argument by Prescott and Ohanian (2014Feb) is proper in choosing the tail of the business cycle because the increase in productivity in 2009 of 3.1 percent and 3.3 percent in 2010 consisted of reducing labor hours.
In revealing research, Edward P. Lazear and James R. Spletzer (2012JHJul22) use the wealth of data in the valuable database and resources of the Bureau of Labor Statistics (http://www.bls.gov/data/) in providing clear thought on the nature of the current labor market of the United States. The critical issue of analysis and policy currently is whether unemployment is structural or cyclical. Structural unemployment could occur because of (1) industrial and demographic shifts and (2) mismatches of skills and job vacancies in industries and locations. Consider the aggregate unemployment rate, Y, expressed in terms of share si of a demographic group in an industry i and unemployment rate yi of that demographic group (Lazear and Spletzer 2012JHJul22, 5-6):
Y = ∑isiyi (1)
This equation can be decomposed for analysis as (Lazear and Spletzer 2012JHJul22, 6):
∆Y = ∑i∆siy*i + ∑i∆yis*i (2)
The first term in (2) captures changes in the demographic and industrial composition of the economy ∆si multiplied by the average rate of unemployment y*i , or structural factors. The second term in (2) captures changes in the unemployment rate specific to a group, or ∆yi, multiplied by the average share of the group s*i, or cyclical factors. There are also mismatches in skills and locations relative to available job vacancies. A simple observation by Lazear and Spletzer (2012JHJul22) casts intuitive doubt on structural factors: the rate of unemployment jumped from 4.4 percent in the spring of 2007 to 10 percent in October 2009. By nature, structural factors should be permanent or occur over relative long periods. The revealing result of the exhaustive research of Lazear and Spletzer (2012JHJul22) is:
“The analysis in this paper and in others that we review do not provide any compelling evidence that there have been changes in the structure of the labor market that are capable of explaining the pattern of persistently high unemployment rates. The evidence points to primarily cyclic factors.”
The theory of secular stagnation cannot explain sudden collapse of the US economy and labor markets. The theory of secular stagnation departs from an aggregate production function in which output grows with the use of labor, capital and technology (see Pelaez and Pelaez, Globalization and the State, Vol. I (2008a), 11-6). Simon Kuznets (1971) analyzes modern economic growth in his Lecture in Memory of Alfred Nobel:
“The major breakthroughs in the advance of human knowledge, those that constituted dominant sources of sustained growth over long periods and spread to a substantial part of the world, may be termed epochal innovations. And the changing course of economic history can perhaps be subdivided into economic epochs, each identified by the epochal innovation with the distinctive characteristics of growth that it generated. Without considering the feasibility of identifying and dating such economic epochs, we may proceed on the working assumption that modern economic growth represents such a distinct epoch - growth dating back to the late eighteenth century and limited (except in significant partial effects) to economically developed countries. These countries, so classified because they have managed to take adequate advantage of the potential of modern technology, include most of Europe, the overseas offshoots of Western Europe, and Japan—barely one quarter of world population.”
Chart II-7 provides nonfarm-business labor productivity, measured by output per hour, from 1947 to 2017. The rate of productivity increase continued in the early part of the 2000s but then softened and fell during the global recession. The interruption of productivity increases occurred exclusively in the current business cycle. Lazear and Spletzer (2012JHJul22) find “primarily cyclic” factors in explaining the frustration of currently depressed labor markets in the United States. Stagnation of productivity is another cyclic event and not secular trend. The theory and application of secular stagnation to current US economic conditions is void of reality.
Chart II-7, US, Nonfarm Business Labor Productivity, Output per Hour, 1947-2017, Index 2009=100
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Table II-6 expands Table II-2 providing more complete measurements of the Productivity and Cost research of the Bureau of Labor Statistics. The proper emphasis of Prescott and Ohanian (2014Feb) is on the low productivity increases from 2011 to 2016. Labor productivity increased 3.3 percent in 2010 and 3.1 percent in 2009. There is much stronger yet not sustained performance in 2010 with productivity growing 3.3 percent because of growth of output of 3.2 percent with decline of hours worked of 0.1 percent. Productivity growth of 3.1 percent in 2009 consists of decline of output by 4.3 percent while hours worked collapsed 7.2 percent, which is not a desirable route to progress. The expansion phase of the economic cycle concentrated in one year, 2010, with underperformance in the remainder of the expansion from 2011 to 2016 of productivity growth at average 0.6 percent per year.
Table II-6, US, Productivity and Costs, Annual Percentage Changes 2007-2016
| 2016 | 2015 | 2014 | 2013 | 2012 | |
| Productivity | -0.1 | 1.3 | 1.0 | 0.3 | 0.9 |
| Output | 1.5 | 3.4 | 3.3 | 2.0 | 3.1 |
| Hours Worked | 1.6 | 2.1 | 2.2 | 1.7 | 2.2 |
| Employment | 1.8 | 2.2 | 2.0 | 1.8 | 2.0 |
| Average Weekly Hours Worked | -0.3 | -0.1 | 0.2 | -0.1 | 0.2 |
| Unit Labor Costs | 1.1 | 1.8 | 1.9 | 0.9 | 1.7 |
| Hourly Compensation | 1.0 | 3.1 | 2.9 | 1.2 | 2.6 |
| Consumer Price Inflation | 1.6 | 0.1 | 3.2 | 1.5 | 2.1 |
| Real Hourly Compensation | -0.2 | 3.0 | 1.3 | -0.3 | 0.6 |
| Non-labor Payments | 2.9 | 3.1 | 4.9 | 4.4 | 5.3 |
| Output per Job | -0.4 | 1.2 | 1.2 | 0.2 | 1.1 |
| 2011 | 2010 | 2009 | 2008 | 2007 | |
| Productivity | 0.1 | 3.3 | 3.1 | 0.8 | 1.6 |
| Output | 2.2 | 3.2 | -4.3 | -1.3 | 2.3 |
| Hours Worked | 2.1 | -0.1 | -7.2 | -2.1 | 0.7 |
| Employment | 1.6 | -1.2 | -5.7 | -1.5 | 0.9 |
| Average Weekly Hours Worked | 0.5 | 1.1 | -1.6 | -0.6 | -0.2 |
| Unit Labor Costs | 2.1 | -1.3 | -2.0 | 2.0 | 2.7 |
| Hourly Compensation | 2.2 | 1.9 | 1.0 | 2.8 | 4.3 |
| Consumer Price Inflation | 3.2 | 1.6 | -0.4 | 3.8 | 2.8 |
| Real Hourly Compensation | -1.0 | 0.3 | 1.4 | -1.0 | 1.4 |
| Non-labor Payments | 3.7 | 7.5 | 0.0 | -0.4 | 3.4 |
| Output per Job | 0.6 | 4.4 | 1.5 | 0.2 | 1.4 |
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Productivity growth can bring about prosperity while productivity regression can jeopardize progress. Cobet and Wilson (2002) provide estimates of output per hour and unit labor costs in national currency and US dollars for the US, Japan and Germany from 1950 to 2000 (see Pelaez and Pelaez, The Global Recession Risk (2007), 137-44). The average yearly rate of productivity change from 1950 to 2000 was 2.9 percent in the US, 6.3 percent for Japan and 4.7 percent for Germany while unit labor costs in USD increased at 2.6 percent in the US, 4.7 percent in Japan and 4.3 percent in Germany. From 1995 to 2000, output per hour increased at the average yearly rate of 4.6 percent in the US, 3.9 percent in Japan and 2.6 percent in Germany while unit labor costs in USD fell at minus 0.7 percent in the US, 4.3 percent in Japan and 7.5 percent in Germany. There was increase in productivity growth in Japan and France within the G7 in the second half of the 1990s but significantly lower than the acceleration of 1.3 percentage points per year in the US. Table II-7 provides average growth rates of indicators in the research of productivity and growth of the US Bureau of Labor Statistics. There is dramatic decline of productivity growth from 2.1 percent per year on average from 1947 to 2016 to 1.2 percent per year on average in the whole cycle from 2007 to 2016. Productivity increased at the average rate of 2.3 percent from 1947 to 2007. There is profound drop in the average rate of output growth from 3.4 percent on average from 1947 to 2016 to 1.4 percent from 2007 to 2016. Output grew at 3.7 percent per year on average from 1947 to 2007. Long-term economic performance in the United States consisted of trend growth of GDP at 3 percent per year and of per capita GDP at 2 percent per year as measured for 1870 to 2010 by Robert E Lucas (2011May). The economy returned to trend growth after adverse events such as wars and recessions. The key characteristic of adversities such as recessions was much higher rates of growth in expansion periods that permitted the economy to recover output, income and employment losses that occurred during the contractions. Over the business cycle, the economy compensated the losses of contractions with higher growth in expansions to maintain trend growth of GDP of 3 percent and of GDP per capita of 2 percent. The US maintained growth at 3.0 percent on average over entire cycles with expansions at higher rates compensating for contractions. US economic growth has been at only 2.2 percent on average in the cyclical expansion in the 32 quarters from IIIQ2009 to IIQ2017. Boskin (2010Sep) measures that the US economy grew at 6.2 percent in the first four quarters and 4.5 percent in the first 12 quarters after the trough in the second quarter of 1975; and at 7.7 percent in the first four quarters and 5.8 percent in the first 12 quarters after the trough in the first quarter of 1983 (Professor Michael J. Boskin, Summer of Discontent, Wall Street Journal, Sep 2, 2010 http://professional.wsj.com/article/SB10001424052748703882304575465462926649950.html). There are new calculations using the revision of US GDP and personal income data since 1929 by the Bureau of Economic Analysis (BEA) (http://bea.gov/iTable/index_nipa.cfm) and the second estimate of GDP for IIQ2017 (https://www.bea.gov/newsreleases/national/gdp/gdpnewsrelease.htm). The average of 7.7 percent in the first four quarters of major cyclical expansions is in contrast with the rate of growth in the first four quarters of the expansion from IIIQ2009 to IIQ2010 of only 2.7 percent obtained by dividing GDP of $14,745.9 billion in IIQ2010 by GDP of $14,355.6 billion in IIQ2009 {[($14,745.9/$14,355.6) -1]100 = 2.7%], or accumulating the quarter on quarter growth rates (https://cmpassocregulationblog.blogspot.com/2017/09/mediocre-cyclical-united-states.html and earlier https://cmpassocregulationblog.blogspot.com/2017/08/data-dependent-monetary-policy-with.html). The expansion from IQ1983 to IVQ1985 was at the average annual growth rate of 5.9 percent, 5.4 percent from IQ1983 to IIIQ1986, 5.2 percent from IQ1983 to IVQ1986, 5.0 percent from IQ1983 to IQ1987, 5.0 percent from IQ1983 to IIQ1987, 4.9 percent from IQ1983 to IIIQ1987, 5.0 percent from IQ1983 to IVQ1987, 4.9 percent from IQ1983 to IIQ1988, 4.8 percent from IQ1983 to IIIQ1988, 4.8 percent from IQ1983 to IVQ1988, 4.8 percent from IQ1983 to IQ1989, 4.7 percent from IQ1983 to IIQ1989, 4.7 percent from IQ1983 to IIIQ1989, 4.5 percent from IQ1983 to IVQ1989. 4.5 percent from IQ1983 to IQ1990, 4.4 percent from IQ1983 to IIQ1990, 4.3 percent from IQ1983 to IIIQ1990, 4.0 percent from IQ1983 to IVQ1990 and at 7.8 percent from IQ1983 to IVQ1983 (https://cmpassocregulationblog.blogspot.com/2017/09/mediocre-cyclical-united-states.html and earlier https://cmpassocregulationblog.blogspot.com/2017/08/data-dependent-monetary-policy-with.html). The National Bureau of Economic Research (NBER) dates a contraction of the US from IQ1990 (Jul) to IQ1991 (Mar) (http://www.nber.org/cycles.html). The expansion lasted until another contraction beginning in IQ2001 (Mar). US GDP contracted 1.3 percent from the pre-recession peak of $8983.9 billion of chained 2009 dollars in IIIQ1990 to the trough of $8865.6 billion in IQ1991 (http://www.bea.gov/iTable/index_nipa.cfm). The US maintained growth at 3.0 percent on average over entire cycles with expansions at higher rates compensating for contractions. Growth at trend in the entire cycle from IVQ2007 to IIQ2017 would have accumulated to 32.4 percent. GDP in IIQ2017 would be $19,849.1 billion (in constant dollars of 2009) if the US had grown at trend, which is higher by $2819.1 billion than actual $17,030.0 billion. There are about two trillion dollars of GDP less than at trend, explaining the 22.2 million unemployed or underemployed equivalent to actual unemployment/underemployment of 13.1 percent of the effective labor force (Section I and earlier https://cmpassocregulationblog.blogspot.com/2017/08/data-dependent-monetary-policy-with.html). US GDP in IIQ2017 is 14.2 percent lower than at trend. US GDP grew from $14,991.8 billion in IVQ2007 in constant dollars to $17,030.0 billion in IIQ2017 or 13.6 percent at the average annual equivalent rate of 1.4 percent. Professor John H. Cochrane (2014Jul2) estimates US GDP at more than 10 percent below trend. Cochrane (2016May02) measures GDP growth in the US at average 3.5 percent per year from 1950 to 2000 and only at 1.76 percent per year from 2000 to 2015 with only at 2.0 percent annual equivalent in the current expansion. Cochrane (2016May02) proposes drastic changes in regulation and legal obstacles to private economic activity. The US missed the opportunity to grow at higher rates during the expansion and it is difficult to catch up because growth rates in the final periods of expansions tend to decline. The US missed the opportunity for recovery of output and employment always afforded in the first four quarters of expansion from recessions. Zero interest rates and quantitative easing were not required or present in successful cyclical expansions and in secular economic growth at 3.0 percent per year and 2.0 percent per capita as measured by Lucas (2011May). There is cyclical uncommonly slow growth in the US instead of allegations of secular stagnation. There is similar behavior in manufacturing. There is classic research on analyzing deviations of output from trend (see for example Schumpeter 1939, Hicks 1950, Lucas 1975, Sargent and Sims 1977). The long-term trend is growth of manufacturing at average 3.1 percent per year from Jul 1919 to Jul 2017. Growth at 3.1 percent per year would raise the NSA index of manufacturing output from 108.2393 in Dec 2007 to 145.0407 in Jul 2017. The actual index NSA in Jul 2017 is 102.1551, which is 29.6 percent below trend. Manufacturing output grew at average 2.0 percent between Dec 1986 and Jul 2017. Using trend growth of 2.0 percent per year, the index would increase to 130.8588 in Jul 2017. The output of manufacturing at 102.1551 in Jul 2017 is 21.9 percent below trend under this alternative calculation.
Table II-7, US, Productivity and Costs, Average Annual Percentage Changes 2007-2016 and 1947-2016
| Average Annual Percentage Rate 2007-2016 | Average Annual Percentage Rate 1947-2007 | Average Annual Percentage Rate 1947-2016 | |
| Productivity | 1.2 | 2.3 | 2.1 |
| Output | 1.4 | 3.7 | 3.4 |
| Hours | 0.2 | 1.4 | 1.2 |
| Employment | 0.3 | 1.6 | 1.5 |
| Average Weekly Hours | -0.7* | -14.4* | -15.0* |
| Hourly Compensation | 2.1 | 5.4 | 5.0 |
| Consumer Price Inflation | 1.6 | 3.8 | 3.5 |
| Real Hourly Compensation | 0.4 | 1.7 | 1.5 |
| Unit Labor Costs | 0.9 | 3.0 | 2.8 |
| Unit Non-Labor Payments | 2.0 | 3.5 | 3.3 |
| Output per Job | 1.1 | 2.0 | 1.9 |
* Percentage Change
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Unit labor costs increased sharply during the Great Inflation from the late 1960s to 1981 as shown by sharper slope in Chart II-8. Unit labor costs continued to increase but at a lower rate because of cyclic factors and not because of imaginary secular stagnation.
Chart II-8, US, Nonfarm Business, Unit Labor Costs, 1947-2017, Index 2009=100
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
Real hourly compensation increased at relatively high rates after 1947 to the early 1970s but reached a plateau that lasted until the early 1990s, as shown in Chart II-9. There were rapid increases until the global recession. Cyclic factors and not alleged secular stagnation explain the interruption of increases in real hourly compensation.
Chart II-9, US, Nonfarm Business, Real Hourly Compensation, 1947-2017, Index 2009=100
Source: US Bureau of Labor Statistics http://www.bls.gov/lpc/
No comments:
Post a Comment