<HTML>
<HEAD>
<TITLE>Productivity Puzzle</TITLE>
<META NAME="description" CONTENT="A key question for the economy's long-run performance is whether the recent acceleration in labor productivity growth is here to stay or whether it will be fleeting. New economy theorists argue that the recent acceleration in productivity is indeed indicative of a number of structural changes to the economy-accelerated technological change, ongoing globalization, deregulation-that have raised underlying productivity gains.">
<META NAME="keywords" CONTENT="productivity,gdp,labor,expansion,new economy,year 2000,skills,growth">
<META NAME="author" CONTENT="Kenneth Markus and Keith Antonyshyn">
<SCRIPT LANGUAGE="JavaScript">
<!-- hide script from old browsers
function today(thisDay,format)
{
if (thisDay.getDay() == 0)
dayOfTheWeek = "Sunday";
else if (thisDay.getDay() == 1)
dayOfTheWeek = "Monday";
else if (thisDay.getDay() == 2)
dayOfTheWeek = "Tuesday";
else if (thisDay.getDay() == 3)
dayOfTheWeek = "Wednesday";
else if (thisDay.getDay() == 4)
dayOfTheWeek = "Thursday";
else if (thisDay.getDay() == 5)
dayOfTheWeek = "Friday";
else
dayOfTheWeek = "Saturday";
if (thisDay.getMonth() == 0)
thisMonth = "January";
else if (thisDay.getMonth() == 1)
thisMonth = "February";
else if (thisDay.getMonth() == 2)
thisMonth = "March";
else if (thisDay.getMonth() == 3)
thisMonth = "April";
else if (thisDay.getMonth() == 4)
thisMonth = "May";
else if (thisDay.getMonth() == 5)
thisMonth = "June";
else if (thisDay.getMonth() == 6)
thisMonth = "July";
else if (thisDay.getMonth() == 7)
thisMonth = "August";
else if (thisDay.getMonth() == 8)
thisMonth = "September";
else if (thisDay.getMonth() == 9)
thisMonth = "October";
else if (thisDay.getMonth() == 10)
thisMonth = "November";
else
thisMonth = "December";
if (format == 0)
{
formatStyle = "<FONT FACE='Verdana, Arial' SIZE=2><B>"
endBold = "</B>"
}else{
formatStyle = "<FONT FACE='TimesNewRoman' SIZE=2>"
endBold = ""
}
dateValue = document.write(formatStyle + thisMonth + " " + thisDay.getDate()+", "+(1900+thisDay.getYear())+ endBold + "</FONT>")
return dateValue
}
if (document.images)
{
toolbox_on = new Image()
toolbox_on.src = '/icons/navbar/toolbox_on.gif'
thoughts_on = new Image()
thoughts_on.src = '/icons/navbar/thoughts_on.gif'
regions_on = new Image()
regions_on.src = '/icons/navbar/regions_on.gif'
forecasts_on = new Image()
forecasts_on.src = '/icons/navbar/forecasts_on.gif'
industries_on = new Image()
industries_on.src = '/icons/navbar/industries_on.gif'
economy_on = new Image()
economy_on.src = '/icons/navbar/economy_on.gif'
}
function imgOn(name)
{
if (document.images)
{
document.images[name].src = '/icons/navbar/' + name + '_on.gif'
}
}
function imgOff(name,section)
{
if (document.images)
{
document.images[name].src = '/icons/navbar/' + name + '.gif'
if (section != '') {imgOn(section)}
}
}
function getAd()
{
//var arrBanner = new Array("carbanner.gif","lunchbanner.gif","deadline.gif","dismal.gif")
//var intBanner = 4*Math.random()
//if (intBanner < 1) {intBanner = 0} //Necessary because in Netscape 4, if < 1 returns NaN
//intBanner = parseInt(intBanner)%4
//var adVal = "<A HREF='http://www.dismal.com' TARGET='_top'>"
//adVal += "<IMG SRC='/banners/" + arrBanner[intBanner] + "' BORDER='0'></A>"
var now = new Date()
var my_random = now.getTime()
var adSpace = "ErrataPages"
var path
if (window.location.href) {path = window.location.href}
else{document.location}
if (path.indexOf("economy") > 0 || path.indexOf("dict_") > 0) { adSpace = "EconomyPages" }
if (path.indexOf("regions") > 0 || path.indexOf("state") > 0 || path.indexOf("metro") > 0 ) { adSpace = "RegionsPages" }
if (path.indexOf("thoughts") > 0) { adSpace = "ThoughtsPages" }
if (path.indexOf("home") > 0 || path.indexOf("new_home.asp") > 0 ) { adSpace = "DismalHome" }
var adVal = "<A HREF='http://adex3.flycast.com/server/socket/127.0.0.1:2800/click/TheDismalScientist/" + adSpace + "/1" + my_random + "' TARGET=_top>"
adVal += "<IMG SRC='http://adex3.flycast.com/server/socket/127.0.0.1:2800/ad/TheDismalScientist/" + adSpace + "/1" + my_random + "' BORDER=0 WIDTH=468 HEIGHT=60></A>"
document.write(adVal)
}
// end hiding script from old browsers-->
</SCRIPT>
</HEAD>
<BODY BGCOLOR="#FFFFFF" LINK="#0E0F5C" TEXT="#000000" VLINK="#445B8B">
<CENTER>
<TABLE WIDTH=610 BORDER=0 CELLSPACING=0 CELLPADDING=0>
<TR>
<TD ALIGN="left" VALIGN="center"><A NAME="top"><A HREF="/home.stm" TARGET="_top" ALT="Dismal Scientist"><IMG SRC="/icons/navbar/dismal_scientist.gif" HEIGHT=60 WIDTH=142 BORDER=0></A></TD>
<TD ALIGN=right><IMG SRC="/icons/clear.gif" HEIGHT=1 WIDTH=468>
<SCRIPT LANGUAGE="JavaScript">
<!--
getAd()
//-->
</SCRIPT><BR>
<NOSCRIPT>
<A HREF="http://ads.ad-venture.com/cgi-bin/accipiter/adclick.exe/SITE=051/CAT=008/PLC=001/TYP=001">
<IMG SRC="http://ads.ad-venture.com/cgi-bin/accipiter/adserver.exe/SITE=051/CAT=008/PLC=001/TYP=001" HEIGHT="60" WIDTH="468" BORDER="0"></A>
</NOSCRIPT>
</TD></TR>
<TR>
<TD COLSPAN=2><IMG SRC="/icons/clear.gif" HEIGHT=5 WIDTH=610></TD></TR>
<TR>
<TD WIDTH=142 ALIGN="left" VALIGN="center"><IMG SRC="/icons/clear.gif" HEIGHT=1 WIDTH=142>
<SCRIPT LANGUAGE="JavaScript">
<!--
today(new Date(),0)
//-->
</SCRIPT></td>
<TD ALIGN="right" VALIGN="bottom"><IMG SRC="/icons/navbar/authority.gif" HEIGHT=11 WIDTH=435></TD></TR>
<TR>
<TD COLSPAN=2 VALIGN="top"><img src="/icons/line.gif" HEIGHT=1 WIDTH=610></TD></TR>
</TABLE>
<TABLE WIDTH=610 BORDER=0 CELLSPACING=0 CELLPADDING=0>
<TR>
<TD><A HREF="/economy/economy.stm" TARGET="_top" onMouseOver="imgOn('economy')" onMouseOut="imgOff('economy','thoughts')"><IMG NAME='economy' SRC="/icons/navbar/economy.gif" HEIGHT=30 WIDTH=105 BORDER=0 ALT="Economy: Indicators"></A></TD>
<TD><A HREF="/regions/regions.stm" TARGET="_top" onMouseOver="imgOn('regions')" onMouseOut="imgOff('regions','thoughts')"><IMG NAME='regions' SRC="/icons/navbar/regions.gif" HEIGHT=30 WIDTH=90 BORDER=0 ALT="Regions: Data on Metros and States"></A></TD>
<TD><A HREF="/thoughts/archive/archives.asp" TARGET="_top" onMouseOver="imgOn('thoughts')" onMouseOut="imgOff('thoughts','thoughts')"><IMG NAME='thoughts' SRC="/icons/navbar/thoughts_on.gif" HEIGHT=30 WIDTH=100 BORDER=0 ALT="Thoughts: Articles on Topical Economic Issues"></A></TD>
<TD><A HREF="/forecasts/forecasts.stm" TARGET="_top" onMouseOver="imgOn('forecasts')" onMouseOut="imgOff('forecasts','thoughts')"><IMG NAME='forecasts' IMG SRC="/icons/navbar/forecasts.gif" HEIGHT=30 WIDTH=100 BORDER=0 ALT="Forecasts: Prognostications from our Economists"></A></TD>
<TD><A HREF="/industries/industries.stm" TARGET="_top" onMouseOver="imgOn('industries')" onMouseOut="imgOff('industries','thoughts')"><IMG NAME='industries' SRC="/icons/navbar/industries.gif" HEIGHT=30 WIDTH=105 BORDER=0 ALT="Industries: Industry Prices"></A></TD>
<TD><A HREF="/toolbox/toolbox.stm" TARGET="_top" onMouseOver="imgOn('toolbox')" onMouseOut="imgOff('toolbox','thoughts')"><IMG NAME='toolbox' SRC="/icons/navbar/toolbox.gif" HEIGHT=30 WIDTH=110 BORDER=0 ALT="Toolbox: Calendars, Calculators, etc."></A></TD></TR>
</TABLE>
<TABLE WIDTH=600 BORDER=0 CELLPADDING=0 CELLSPACING=0> <BR>
<TR><TD ALIGN=right COLSPAN=2><FONT FACE="Arial, Helvetica" SIZE=4><B><A HREF="archive/archives.asp">Archives</A></B></FONT></TD></TR>
<TR><TD COLSPAN=2><H2><TABLE WIDTH=100%><TR BGCOLOR="#000060"><TD><FONT FACE="Verdana,Arial" COLOR="#FFFFFF" SIZE=2><B>Productivity Puzzle</B></TD></TR></TABLE></H2>
<I>Analysis by <A HREF="/thoughts/archive/bios.asp?Author=Mark+Zandi">Mark Zandi</A><BR>
Written August 3, 1998</I>
<P>
<A HREF="/economy/releases/productivity.asp">Productivity growth</A> has accelerated during the past three years. Accounting for the recent revisions to the <A HREF="/economy/releases/gdp.asp">GDP accounts</A>, nonfarm labor productivity has risen by an estimated near 2% annual rate during the past three years. This is a substantial improvement over the close to 1.5% per annum growth experienced so far during the current expansion and the only 1% per annum growth experienced during the past quarter century.
<P>
A key question for the economy's long-run performance is whether the recent acceleration in labor productivity growth is here to stay or whether it will be fleeting. New economy theorists argue that the recent acceleration in productivity is indeed indicative of a number of structural changes to the economy-accelerated technological change, ongoing globalization, deregulation-that have raised underlying productivity gains.
<P>
The new economy theorists are half-right. Underlying productivity growth has likely accelerated in recent years, but not to the degree suggested by the recent strong gains. Productivity growth is now closer to 1.5% per annum, which is up from the 1% growth experienced since the early 1970's, but well below the 2% currently being experienced.
<P>
Much of the recent measured acceleration in productivity is due to the way the Bureau of Labor Statistics measures productivity. The productivity estimates are for what the BLS calls all persons, which includes wage and salary workers and the self-employed. In calculating the total number of hours worked by labor, the BLS adds the hours worked by wage and salary employees to the hours worked by the self-employed. Since the BLS does not have good estimates of output by the self-employed, however, it assumes that the output of the self-employed is the same as the output of wage and salary workers.
<P>
This is not a problem when the number of self-employed is changing similarly to the number of wage and salary workers. That has not been the case recently. The number of self-employed is plunging. On a year-over-year basis through June, the self-employed have fallen by over 1%. The share of total employment that is self-employed has dropped to near a twenty-year low of less than 7%. Behind the falling number of self-employed is the tight labor market and accelerating labor compensation growth. Many who lost their jobs earlier in the decade to corporate restructuring joined the ranks of the self-employed. Now with many positions going begging, these workers are returning to the corporate ranks.
<P>
The BLS assumption that the output of the self employed is the same as wage and salary workers even though hours worked by the self employed are falling has resulted in a significant overstatement of the recent gains in productivity. The bias in the productivity estimates due to the use of this methodology is longstanding. Productivity growth has historically been very strongly inversely related to the self-employed share of total employment.
<P>
The recent gains in productivity are also in part simply the result of a rebound in the economy's growth. After struggling throughout 1995 in the wake of the Mexican economic crisis and the Fed tightening of the year before, the economy accelerated in 1996. Real GDP growth went from 2% in the year ending in the fourth quarter of 1995 to 4% by the start of this year.
<P>
Near-term changes in productivity growth are closely related to changes in the economy's overall rate of growth. When GDP growth is accelerating, such as early on in economic expansions, productivity growth increases. Conversely, when GDP growth is decelerating or the economy is in recession, productivity wanes. This occurs because businesses are reluctant to hire or fire workers quickly in response to a change in the demand for their goods and services. Initially when demand increases, for example, businesses are generally able to raise output by asking their existing workforce to work more hours and to work more productively. Only when businesses are sure that demand will remain strong do they begin to hire more aggressively. Newer and more efficient capital that was idle when the economy was weaker is also oftentimes available to support productivity gains.
<P>
After accounting for the self-employed bias and the impact of an accelerating economy, the productivity gains of the past two years are not significantly greater than the growth experienced throughout this expansion.
<P>
The near-term prospects for productivity growth are also relatively dim. As the accelerating economy of the past two years has lifted productivity gains, the anticipated weakening in economic growth will soon constrain productivity. The very tight labor market will also exacerbate the weakening in productivity. Labor starved businesses are increasingly hiring those with poorer skills and education levels. The cost and time necessary to raise the productivity of these workers is substantially higher than for those workers hired earlier in the expansion. This is not unusual, since it is primarily for this reason that productivity gains traditionally weaken as expansions mature regardless of the underlying trend growth in productivity.
<P>
Another serious potential impediment to future productivity gains is increasing business expenditures on <A HREF="y2k.stm">Year 2000 problems</A>. The cost of adapting computer systems to handle moving into the new millennium are difficult to measure, but judging by cost estimates of major businesses and government they are significant. Some estimates run into the hundreds of billions of dollars for U.S. businesses alone. While businesses may increase their total investment spending to solve their Y2K problems and continue to make investments in new technologies this is not likely to occur on a large scale. Businesses are more likely to shift some of their investment dollars away from more productive technologies to solve their Y2K problems. Business spending on high-tech equipment has remained remarkably constant at close to 2.5% of GDP since the early 1980's. Moreover, while some Y2K solutions may enhance productivity through upgraded computer code and hardware, most solutions will simply make current systems work in the new century.
</TD></TR>
<TR><TD ALIGN=left> <BR><FONT FACE="Arial, Helvetica" SIZE=4><B><A HREF="archive/archives.asp">Archives</A></B></FONT></TD>
<TD ALIGN=right><A HREF="#top"><IMG SRC="/icons/arrowtop.gif" HEIGHT=47 WIDTH=70 BORDER=0></A></TD></TR>
<TR><TD COLSPAN=2 BGCOLOR="#000000"><IMG SRC="/icons/line.gif" HEIGHT=1 WIDTH=1></TD></TR>
</TABLE>
<TABLE WIDTH=610 CELLPADDING=0 CELLSPACING=0 BORDER=0>
<TR>
<TD VALIGN=bottom ><IMG SRC="/icons/footer_logo.gif" HEIGHT=45 WIDTH=50 ALIGN=left></TD>
<TD WIDTH=550><FONT FACE="Arial, Verdana" SIZE=2><B>Questions, Comments, or Inquiries to: <A HREF="mailto:webmaster@dismal.com"><I>webmaster@dismal.com</I></A></B></FONT>
<BR>
<FONT FACE="Arial, Verdana" SIZE=2><SMALL>©1998, Dismal Sciences®
<BR>The Dismal Scientist and Dismal Sciences are registered servicemarks of Dismal Sciences.</SMALL></TD></TR>
</TABLE>
</TR></TD>
</TABLE>
</CENTER>
</BODY>
</HTML>