Mitchell's Musings

  • 12 Oct 2015 8:32 AM | Daniel J.B. Mitchell (Administrator)

    No, this musing is not about global warming but rather about “climate” as it has been used in the context of universities to characterize the atmosphere for women and minorities. And, yes, I am well aware of the abuses that have occurred in the name of climate used that way in the form of excesses regarding “microaggressions” and “trigger warnings” on syllabi.


    Two items appeared recently in the Los Angeles Times regarding climate in the sense above at UCLA. One involved a fraternity and sorority party in which party goers allegedly appeared in blackface.[1] (Exactly what happened is not completely clear at this writing.) The second involved a survey undertaken by the consulting firm Korn Ferry regarding problems in the climate for women faculty at the Anderson Graduate School of Management (at which I had a full-time, and then a half-time, appointment before I retired in 2008).[2] I want to focus on the latter case but both stories connect climate to students, a fact which led to this musing.


    In the case of the Anderson School, the Korn Ferry report has circulated to faculty members, but is not supposed to be a public document. However, it is evident that the LA Times has seen the report. It has been distributed to faculty by email and thus is an easy “leak.” I will nonetheless not reproduce it here since it is not supposed to circulate. But in general terms the Korn Ferry report largely discusses intra-faculty relations at the School. In addition, as noted above, there is material concerning the atmosphere in MBA classes emanating from students


    In essence, the report describes a macho atmosphere in classes which can make life difficult, especially for junior female instructors. Management schools tend to take student ratings of instructors seriously. So classroom problems can pose an obstacle for advancement of junior faculty. Time spent by them dealing with such problems is time that is not available for research and journal publication, career elements which are critical for academic advancement.


    It is interesting to note that back in 2013, the New York Times featured a report on the Harvard Business School that had similar revelations – and which largely focused on (male) student behavior.[3] One suspects that you could investigate prestigious graduate business schools around the country and come up with comparable findings (just as there have been undergraduate fraternity-related incidents around the country comparable to the recent UCLA event). It is worth noting, however, that when the Harvard B-School report came out, Harvard in response raised the proportion of women in its MBA program to over 40%.[4]


    Which brings me to what might be done at UCLA. Faculty inherently turn over slowly; students turn over fast. A new crop of students is admitted annually. Even if there are major issues in intra-faculty relations (peer evaluations, promotions, etc.), it is hard for a dean or other school administrator to make rapid changes such matters. Hiring and advancement of faculty are handled through a complicated process involving committees, departmental meetings, and external reviews. In contrast, student admissions policy can change quickly. Faculty have little direct involvement in MBA admissions; the process is handled by internal School staff. Admission policy is a lever a dean can pull.


    One suspects that a jump in the proportion women – currently the percent female at UCLA Anderson is 30%[5] - would limit the kind of frat house behavior in core MBA courses that apparently went on at Harvard. One suspects that an increase in the proportion of non-traditional students (for example, those who want to work for nonprofits or the public sector or who have done so in the past) would also improve the climate for junior female faculty in particular.[6]


    My personal recollection – which may be faulty – is that the proportion of women in the 1970s at the UCLA management school was higher, perhaps 40% or so – and then fell back. When I arrived in the late 1960s, the proportion of women was very low so the change in the 1970s was dramatic. But the 1970s and 1980s also saw an effort to change the curriculum to be more like Harvard. In the mid-1970s and before, the UCLA program was much more individualized than it became.[7] The Harvard model – a fixed MBA curriculum with an emphasis on case studies – was implemented. The School imitated Harvard and thus became more like Harvard including, so it seems, the deficiencies found there and identified by the New York Times.[8]


    In an effort to promote the Harvard case method at Anderson – part of the general effort to be more like Harvard - there were classes for instructors on how to do it. Not surprisingly, the classes involved cases. One of these cases told the story of an instructor who walks into his class (the instructors in the stories were male) and finds his MBA students engaged in a food fight, literally tossing their lunches around the room at each other and making a mess. It then describes how the instructor took charge and began the day’s lesson. 


    What was remarkable about the story is that it treated such junior high behavior as if it were something that might be expected from a group of graduate students in their late 20s. When I suggested that the remedy for such behavior was not taking charge but rather discharge (dismissal), this idea was quickly put down. The theme of the case was in essence that boys will be boys and that instructors should expect to deal with that fact.


    In short, the Harvard B-school revelation in the New York Times in 2013 should not have been a surprise. It was the accrual of a long history. Imitating that program at other institutions (such as UCLA) could be expected to produce similar results, even if the dark side of those results was unintended. But if imitation is to be the policy, then why not start by imitating what Harvard did after the story came out in the New York Times? Raise the female/male ratio in MBA admissions substantially. And since it’s likely that it’s not just Anderson that has a problem, all other major B-schools should take notice. Admissions policy is not the only fix that’s needed. But it is a quick one.


    [1] http://www.latimes.com/local/lanow/la-me-ln-ucla-blackface-kanye-party-20151008-story.html. See also http://www.kcrw.com/news-culture/shows/which-way-la/is-the-lapd-over-reacting-to-a-hostile-atmosphere#seg-ucla-students-protest-blackface-frat-party.
    [2] http://www.latimes.com/local/education/la-me-ln-ucla-women-20151008-story.html 

    [3] http://www.nytimes.com/2013/09/08/education/harvard-case-study-gender-equity.html and http://www.nytimes.com/2013/09/10/education/harvard-business-students-see-class-as-divisive-an-issue-as-gender.html. (The latter reference referred to income-related snobbery.) 
    [4] http://poetsandquants.com/2014/01/28/hbs-dean-makes-an-unusual-public-apology/  
    [5] http://www.anderson.ucla.edu/degrees/mba-program/admissions/class-profile
    [6] California voters enacted Proposition 209 in 1996 which bans affirmative action
     in public universities – including affirmative action related to sex. [http://ballotpedia.org/California_Affirmative_Action,_Proposition_209_%281996%29] However, MBA admissions involve a significant degree of subjective judgment; it is not just a matter of test scores.

    [7] Up to that point, the School – it was not yet called “Anderson” – had a variety of separate programs for different fields – finance, accounting, marketing, etc. – as well as a general MBA program. The separate programs were similar to “majors” in an undergraduate college. All of the separate fields shared a limited common core of management classes. 
    [8] One element introduced – I think in the 1980s – was the organization of MBA classes into sections. The students were divided into section groups. Each group stayed together through the core courses. This arrangement may simply have been an administrative convenience that avoided the complications of individual student enrollment in each course. But it meant that a section in which a subgroup of students exhibited bad behavior was preserved in class after class to replicate the problem. 

  • 05 Oct 2015 12:41 PM | Daniel J.B. Mitchell (Administrator)

    Universities have been trying to reconcile notions of academic freedom with various versions of speech codes, meant essentially to protect students from statements varying from deliberately aggressive and intolerant to so-called “microaggressions,” statements not meant to be intolerant that might nonetheless be offensive to somebody.  Often these attempts at non-offensiveness degenerate into ludicrous examples that are widely mocked, particularly in conservative media, but also in mainstream outlets.[1] Closely related are efforts to place “trigger warnings” into course syllabi, warning students that some reading material might be offensive.


    Grafted on to this ongoing brouhaha in recent months was an effort by the Regents of the University of California (UC) to deal with complaints of anti-Semitic behavior related to criticisms of Israel, particularly as it is linked to the “BDS” (Boycott, Divest, Sanction) movement. Various student groups at UC campuses and a union representing UC teaching assistants adopted BDS-related resolutions.[2] Jewish student groups complained of incidents of intimidation ranging from anti-Semitic graffiti to asking a Jewish student who was up for a student government office whether she could be unbiased due to her religion. There was a counter-push to have the Regents adopt the “State Department” definition of anti-Semitism which connects the concept to an excessive concentration on Israel to the exclusion of other nations.[3] 


    At one point on an NPR broadcast, UC president Janet Napolitano indicated she favored the definition personally.[4] The UC Regents – who meet every two months – originally put the issue on their July 2015 agenda but then deferred the matter to their September meeting. Somehow, in the interval between these two meetings, university administrators took up the issue – anti-Semitism – and produced a general resolution against intolerance which did not deal specifically with anti-Semitism and was essentially feel-good Pablum.[5] Meanwhile, the California state legislature came out with a resolution condemning anti-Semitism (but not using the State Department definition).[6]


    At the September 2015 Regents meeting, various Regents condemned the Pablum resolution, essentially for saying nothing and omitting anti-Semitism as a specific topic. The Regents sent the university administration back to the drawing board to come up with something more specific – although it appeared from the discussion that the State Department definition was not what was wanted because of the potential conflict with academic freedom. Regent John Pérez, a former legislative leader, noted in particular that when he asked to speak to Jewish groups on one unnamed UC campus, he was told that he would have to “balance” such a discussion with a discussion with other groups. But he encountered no such balancing requests when he asked to speak with other students.[7] In any event, no time table was specified for a new resolution to be drawn up. What the Regents will eventually do is uncertain.


    Although Jewish groups took the Regents’ action as a victory – since it rejected the Pablum approach and pushed for something more specific to their complaints – the State Department wording was off the table. Even if the Regents had adopted the State Department definition, it was not clear what such an adoption would have accomplished. Any group that made statements that fit the definition could have disputed allegations that it was anti-Semitic. It could simply have argued that the definition was incorrect, whatever the Regents might have thought or said. And no particular penalty was entailed. So – is there any solution to reconciling the conflict between free speech/academic freedom with voicing a view that might be viewed (or even defined) as anti-Semitic? The problem lies in trying to define a motivation or thought process for a specific behavior or statement. Clearly, physical intimidation, property damage, and the like are illegal and violate various university standards of behavior. But when it comes to pure speech and advocacy, the motivational approach runs into a dead end. It’s hard to prove what someone was really thinking or what really motivated a statement when the individual could deny it.


    There is, however, an alternative that avoids assessing motivation and thought. The alternative is to focus on behaviors, not thoughts or motivations. Such an alternative focus won’t solve all problems in this area of controversy, but it will help. The B in BDS stands for a boycott based on national origin – Israeli in this case. And the simple fact is that when it comes to anything to do with employment (hiring, evaluation, selection for a promotion or a particular role, etc.) or admissions to university programs, such a boycott would be illegal. It is banned by federal laws – such as the Civil Rights Act of 1964 – related state-level laws, and university policy. Of course, anyone is free to advocate repealing or changing such laws and policy. But absent such repeal or change, the laws and policy are what they are.


    It really doesn’t matter what nationality is the target, so trying to prove an anti-Semitic motivation is not involved in the behavioral approach. A boycott related to employment or admissions would be illegal whether it involved the Israeli-Palestinian conflict, or Japanese whaling practices, or China’s policy toward Tibet, or Russia’s policy toward Ukraine, or Turkish denial of the Armenian genocide, or women’s rights in Saudi Arabia. If UC or any other university were to engage in a boycott aimed at Israelis, Japanese, Chinese, Turks, or Saudis, it could be sued. If agents of UC or any other university who have major roles in employment or admissions decisions were to advocate such boycotts (say, deans, department chairs, heads of research units, etc.), they could expose their university employer to damages.  


    While faculty have a right as individuals to espouse whatever causes they like, there is no right to be selected for key managerial and administrative roles. Anyone appointed to such roles is expected to carry out university policy and legal duties appropriately. Universities as employers can pick managers and administrators who are most likely to carry out university policy appropriately (and to avoid actions and statements that have the potential to expose their employer to legal liability). Indeed, in a more general context, employers typically screen potential candidates for managerial and administrative roles through interviews, examination of credentials and past employment records, etc., in the hopes of finding individuals who will likely best do the expected job. Someone who publicly opposes university policy and federal and state legal requirements would not seem to be a suitable candidate.


    What about student groups that favor boycotts based on national origin? The problem here is that such groups are seen externally as having an official status within the university. It might be best to make it clear that such groups are independent entities and that they speak and act for themselves only. As such, the Regents might want to reconsider policies that mandate student fees to support such student groups or provide some kind of readily available opt-out arrangements for those students who don’t support the political positions of these groups. Absent such distancing, the public perceptions that there is an official status and that student groups do speak for the university have some grounding in fact.


    And what about labor unions such as the TA union mentioned above? Under California labor law as it applies to the public sector, and to higher education in particular, UC must recognize and bargain in good faith with unions that attain a majority vote in a bargaining unit.[8] Typically, however, union contracts include language more or less taken from federal and state laws that forbid discrimination including national origin discrimination. Such clauses can be made more specific to representation of members of the bargaining unit who fall into the national-origin category that the union wants to boycott. Unions advocating national origin boycotts should at a minimum have such non-discrimination clauses in their contracts and employers can insist on insertion of such provisions in the contract. There are also various opt-out arrangements in place under federal and state law for those in the bargaining unit who do not support the political positions of the union. 


    In short, a refocus of the debate away from thought and motivation (anti-Semitism) and toward behavior (advocacy of boycotts and the legal problems that such advocacy entails) would be a better outcome than continued debate along the lines of what has so far occurred at the UC Regents. Clearly, however, all controversy will not vanish if the behavioral approach is adopted. Nonetheless, UC is the largest public university system in the U.S. If it moves toward a focus on boycott behavior, other universities – public and private - are likely to follow.

    --------------------------------------------------------------------


    [1] http://www.scpr.org/programs/airtalk/2015/06/24/43412/microaggressions
    -should-they-be-censored-on-colleg/

    and http://www.latimes.com/opinion/editorials/la-ed-microaggression-what-not-to-say-at-uc-20150624-story.html.  
    [2] In this musing, we don’t take up the issue of financial divestment from the UC portfolio (including the pension plan). However, the university issued a statement in 2010 setting extremely stringent conditions for divestment from a particular country: http://issuu.com/danieljbmitchell/docs/israel_divestment_2010. It might be noted that calls for BDS as it relates to UC often point to UC’s actions regarding South Africa during the era of apartheid. There was financial divestment in that case, in part owing to a deal between then-Governor George Deukmejian and state assembly leader Willie Brown. (Both were ex officio UC regents. In addition, the governor – who initially opposed divestment - needed the cooperation of Brown on state budget matters.)  But there was never an academic boycott of South Africa in that period. There was no regental ban on South African faculty, etc. I can recall visitors from South Africa coming to UCLA; the library continued to subscribe to South African academic journals, etc. Indeed, it would be hard to imagine American academic heart surgeons, for example, ignoring the early transplant work that took place in South Africa starting in the 1960s: http://www.nytimes.com/2001/09/03/world/christiaan-barnard-78-surgeon-for-first-heart-transplant-dies.html
    [3] http://www.state.gov/j/drl/rls/fs/2010/122352.htm.  
    [4] http://hereandnow.wbur.org/2015/05/21/janet-napolitano-anti-semitism-definition.  
    [5] http://regents.universityofcalifornia.edu/regmeet/sept15/e4.pdf.  
    [6] https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=201120120HR35.

    [7] You can hear excerpts from the Regents’ session at https://www.youtube.com/watch?v=WUC07I-CikM. It might be noted that the Regents themselves have engaged in such “balancing.” A student regent is selected each year for a one-year term. Usually, recommendations as to who would be the student regent come from student government. In one year, the name chosen was a Muslim woman who favored BDS: https://www.youtube.com/watch?v=3LxXT8ocax4. The next year, the regents somehow selected – among all possible candidates – a pro-Israel Jewish student: https://www.youtube.com/watch?v=5l-e3rRDZxE. The balancing approach seems to exist at the campus level, too. For example, UCLA has a Center for Near East Studies which has elicited complaints of being anti-Israel: http://www.international.ucla.edu/cnes/. But UCLA also has a Nazarian Center for Israel Studies which goes in the opposite direction: http://www.international.ucla.edu/israel/home

    [8] The relevant statute is similar to federal labor law. See http://www.perb.ca.gov/laws/heera.aspx.

  • 28 Sep 2015 3:28 PM | Daniel J.B. Mitchell (Administrator)

    When I taught labor markets and got into incentives, there was always the issue of incentives which led to perverse behaviors that eventually harm the organization.  A simple example is a piece rate system that emphasizes quantity over quality and leads to defective, hasty production.  Usually, there is some awareness of the potential pitfalls of incentive programs and additional arrangements may be put in place to prevent perverseness. In the piece rate example, you can hire inspectors who watch for defects. But such supplementary efforts cost money and attaining 100% perfection may not be the optimal decision. Some level of defects may be acceptable. And no incentive system perfectly aligns worker interests with employer interests, despite claims and hypes.


    The idea of incentives should not be limited to explicit formulas that link pay to output (such as piece rates, sales commissions, etc.).  In organizations, there are inevitably ways employees discover to get ahead. Certain behaviors are rewarded, perhaps by promotions, plum assignments, and the like. The notion that organizations have “cultures” – for better or for worse - is closely related to the idea of built-in incentives. Employees figure out what gets them rewards. The actual culture may have little resemblance to grand platitudes in organization mission statements. It’s the substance of numerous “Dilbert” cartoon strips.


    What I used to say in class is that if you observe a perverse outcome, it could just be a one-time fluke or a mistake. But if the perversity repeats, there is probably something in the organization that incents such behavior.


    The recent Volkswagen scandal suggests a corollary to me. In the Volkswagen case, certain diesel engine cars had computer programming built in to allow detection of government emission tests. The engines operated differently during such testing than they did when run in actual driving conditions. In some cases, car buyers may have received government rebates based on the supposedly low emissions. As revelations of the fraud became public, the CEO of Volkswagen resigned hastily saying he was “not aware” of having done anything wrong. He said he was “stunned that misconduct on such a scale was possible in the Volkswagen Group."[1]


    If you think about this episode, you can see that the risks of the fraud becoming known over time had to be large. The programming was in the cars waiting to be detected by authorities that became suspicious.


    There is always some disgruntled employee or some righteous whistleblower somewhere in the system to leak out the information. And the gain to Volkswagen of having somewhat lower emissions really wasn’t all that great, especially when compared with the cost of the eventual revelation.


    So, is it possible that the emissions fraud was a one-off event, due to the action of some “bad apple” in the corporate barrel? Should the CEO really have been “stunned”? Maybe. But it seems very unlikely that – even if this incident was a one-off event – that only one bad apple could have pulled it off. There had to be a lot of folks at some level in the firm aware of what was being done, even if the CEO did not know of - or did not officially authorize - the plan.


    My corollary suggests that a really big fraud of the VW type is unlikely to be a one-off event. The corollary is instead that if there is a really big perverse event, look for earlier – perhaps less risky – frauds that have not yet been uncovered. If there were such frauds, and if those employees who perpetrated them were met with rewards, you can see how a perverse-incentive culture could form that eventually led to the emissions scandal. My suggestion to the authorities who will be probing the VW emissions fraud is to find out what preceded it.


                                                              
    [1] http://www.latimes.com/business/la-fi-vw-ceo-resign-20150923-story.html  


  • 21 Sep 2015 12:08 PM | Daniel J.B. Mitchell (Administrator)
    Last week’s musing involved the then-upcoming decision by the Federal Reserve on whether to raise interest rates. The musing noted that there was pressure on the Fed to raise short-term rates from near zero, not because there was some evident change in economic circumstances that would warrant a hike, but because near-zero rates were abnormal. I argued that the factors that should be the determinants – the general state of the economy in real terms (including the labor market) and inflation – were not indicating a need for a shift in policy. If anything, there were some weaknesses still present on the real side and inflation (whether we are talking wages or prices) was low and not accelerating. 


    Ultimately, the Fed’s decision – taken with only one dissent - was not to raise interest rates.[1] And the factors cited were pretty much along the lines above: the condition of the real economy and the lack of inflation:


    “On balance, labor market indicators show that underutilization of labor resources has diminished since early this year. Inflation has continued to run below the Committee's longer-run objective, partly reflecting declines in energy prices and in prices of non-energy imports. Market-based measures of inflation compensation moved lower; survey-based measures of longer-term inflation expectations have remained stable…”

    In the run-up to the decision, there were editorials and op eds and general articles. The upcoming decision was hyped, in some cases well beyond what was warranted. In view of some observers, there were suggestions that the current low rates were an aid to Wall Street.


    One article suggested that financial markets were addicted – as in morphine - to low rates.[2] But another op ed after the decision complained that it was Wall Street, or at least the holders of capital, that wanted a rate hike.[3] In fact, immediately after the decision the S&P 500 index rose and then fell. For the week ending September 18 as a whole, the index ended up almost exactly where it started.




    Basically, monetary policy is a blunt instrument when it comes to such concerns as income distribution (Main Street versus Wall Street; the 1% versus the 99%). Issues of income distribution are best addressed through fiscal policy. Unfortunately, at present fiscal policy is as gridlocked as ever. Indeed, at this writing, there is looming yet another threat of a federal government shutdown. So, as has been the case for much of recent history, monetary policy is the only macro instrument we have going. We should be grateful it’s in sensible hands.


    Luckily for the U.S., the Fed has been chaired in the Great Recession and its aftermath by two non-ideological pragmatists who let the data talk and avoid basing decisions on outmoded theories: Ben Bernanke and Janet Yellen. Here is Yellen’s policy statement after the recent interest rate decision:


    …Our actual policy actions over time will depend on how economic conditions evolve, which is quite uncertain. If the expansion proves to be more vigorous than currently anticipated and inflation moves higher than expected, then the appropriate path would likely follow a steeper and higher trajectory; conversely, if conditions were to prove weaker, then the appropriate trajectory would be lower and less steep.[4]


    What more do you want?


                                                              
    [1] http://www.federalreserve.gov/newsevents/press/monetary/20150917a.htm
    [2] http://www.latimes.com/business/la-fi-fed-market-influence-20150912-story.html (See the quote at the tail end of the article.) 
    [3] http://www.latimes.com/opinion/op-ed/la-oe-super-fed-rate-min-wage-20150918-story.html   

    [4] http://www.federalreserve.gov/mediacenter/files/FOMCpresconf20150917.pdf

  • 14 Sep 2015 12:40 PM | Daniel J.B. Mitchell (Administrator)
    Imagine you are driving a car. If you depress the accelerator, the car tends to go faster as more gasoline flows into the engine. If you ease up on the accelerator, the reverse occurs. Over time, if anyone cared to do so, you could measure the average angle of the accelerator pedal relative to the floorboard for all the years you had been driving. I suppose you could say that this average angle was “normal” in some sense. But would anyone argue that at any point in time it was wrong to deviate from the “normal” angle? If you were driving up a hill, you would push the accelerator down to maintain speed so the angle would be below normal (average). If you were driving down a hill, you might take your foot off the pedal entirely. In neither of these situations would you say you were doing the wrong thing. You were simply adapting to driving conditions.


    When it comes to Federal Reserve policy on interest rates, however, the notion that deviations from the norm are wrong and should be corrected ASAP, regardless of economic conditions, seem to be in the air.  Interest rates are being kept low presently, which is seen as “abnormal” because on average they have been higher. Because the current level is abnormal, the Fed – it is argued - should raise rates. If the Fed doesn’t raise rates, Bad Things will happen because, well, it’s not normal for rates to be so low. The economy is sure to overheat and break out in inflation.


    When you dig into it, there are folks who believe terrible things will happen if we don’t revert to “normal” because there was much previous monetary creation by the Fed to combat the Great Recession and then to stimulate the recovery, as the chart below shows.
     


    The problem is that there still is no sign of this inflation. Moreover, there is no sign that financial markets are currently expecting inflation.  The 10-year “breakeven” inflation rate – the difference between 10-year inflation adjusted Treasury security yields and conventional 10-year Treasury yields shown below - has bounced around since the Great Recession (when something like the Great Depression was feared and zero inflation or less was projected). But the breakeven rate is currently in the 1.5-2.0%/annum range. So in what sense is the economy overheating or are financial markets expecting such overheating? The problem seems to be that for some folks, facts don’t get in the way of theory.  
     



    There is another strand of the argument that the Fed should be raising interest rates now, even though inflation isn’t high. In this view, there is a “natural” rate of unemployment. Below that level, labor shortages will arise, employers will start bidding up wages (faster and faster the theory goes).  The rising labor costs will show up in prices through markups and therefore will pass into price inflation.  It is true that as the unemployment rate has declined in recent years, the U.S Bureau of Labor Statistics’ (BLS) job openings (vacancy) rate has increased. In fact, the job openings rate (shown below) has exceeded its pre-Great Recession peak. But before panicking, there is a question as to whether we have the right story. Where is the ever-accelerating wage inflation? As measured by the latest release of the BLS Private-Sector Employment Cost Index, for example, accelerating wage inflation has yet to emerge, suggesting that past relationships are not holding.[1]
     
    Maybe the (new) natural rate of unemployment has yet to be reached. Maybe it has changed from whatever level it stood at prior to the Great Recession. Maybe the past is not prologue. There seem to be a lot of maybes here – too many – to make a major monetary policy decision based on old assumptions.




     We know there have been other changes in the labor market apart from the old wage inflation/vacancy/unemployment link. For example, the employment-to-population ratio is well below its pre-Great Recession peak. So there may be more slack present in the labor market than is suggested by measured unemployment or vacancies. No one knows for sure.



    Apart from whether the wage/vacancy/unemployment rate relation is the same as it was, say, ten years ago before the Great Recession, there is still another question to be posed.  The theory or story of the natural rate of unemployment is basically a labor market tale. It involves demand for labor pulling up wages and the resulting wage inflation being passed along into price inflation. The problem is that broad macro measures such as the unemployment rate are correlated with other broad measures such as estimates of the gap between actual GDP and “capacity.” It’s hard to say empirically which index we should be looking at or what the true story is.


    At one time when unions were strong, stories of worker bargaining and labor costs passed into prices may have made sense. But today, the determinants of price inflation (which in the end is what the Fed cares about) may be largely a product market story, not a labor market story. We may be looking in the wrong place when it comes to predicting the point when price inflation will become a problem.  All of which brings us back to our car analogy. The current position of the accelerator seems consistent with our current driving conditions. Current interest rate policy is not causing inflation. Why make a change?


                                                              
    [1] Readers of these musings will know that in a prior post, I expressed the view that it looked like there was some acceleration in wage inflation. See http://employmentpolicy.org/page-1775968/3328512#sthash.1ymQLhfw.dpbs.
    But we have to let the latest data talk and right now that is not the tale they are telling.

     
  • 07 Sep 2015 9:23 AM | Daniel J.B. Mitchell (Administrator)

    The standard Labor Day article either talks about whether organized labor will make a comeback after a long period of decline, or it picks up on some other aspect of labor market trends and problems such as stagnant wages, pay inequality, job insecurity, etc. This musing is being written shortly before the Labor
    Day articles for 2015 actually appear. So what the actual balance will be among these two types is unknown. 


    My own guess is that because of the decades-long trend in falling unionization rates, there will be more of the latter (labor market issues) – probably many more – than of the former (union comeback). You have only to ask what “CIO” stood for in 1955 (when the CIO disappeared into the AFL-CIO and unionization was at its peak) and what happens nowadays if you Google “CIO.” You are more likely
    nowadays to run into “Chief Information Officer” as the meaning of CIO or – even more tellingly – “Chief Investment Officer,” than you are to encounter the 1955 meaning. (If you don’t know the 1955 meaning, you’re just making my point.)


    So assuming articles on problems of the contemporary labor force are mainly what you will encounter, my further guess is that what you will also find is the idea that the jobs of the future will require college degrees. Higher ed, in other words, is the solution to today’s labor market problems, at least in that telling. Let’s put aside the inconvenient fact that according to the U.S. Bureau of Labor Statistics, the top projected job openings are in retail, food service, and other low-education and low-paid occupations.[1]


    What we are talking about here is public perceptions, not necessarily reality. Universities and colleges have long been referred to as “ivory towers.” Presumably what is meant by that phrase is insulation from the “real world.” Given that longstanding view, combined with the more recent perception that the solution for labor market problems is getting a college degree, and you have a circumstance that did not exist in the past. If, in the past, universities and colleges were insulated ivory towers, but you didn’t need to go there, their ivory tower aspects were a mere curiosity. If, on the other hand, you (or your kids) do have to go there, what might have been a curiosity back in the day becomes a potential conflict if you see future barriers to entrance.  


    The problem becomes especially acute in public higher education. Public institutions – because they are supposed to offer lower-cost attendance options than private - thanks to government subsidy – become viewed as the utilitarian route to labor-market advancement.  And if the folks in charge of those institutions seem engaged in odd activities unrelated to efficient and inexpensive student processing, public concerns are raised. What are those folks doing with taxpayer money? Why should I as a taxpayer be subsidizing such activities at a time of rising tuition?


    The most obvious elements of friction relate to admissions (access) and, as noted, rising tuition. During the Great Recession, state governments tended to reduce appropriations for public higher ed as tax revenue declined. As a result, tuitions rose and, in some cases, enrollments were cut. With a piece of their budgets cut away, such adjustments by public higher ed institutions were inevitable. In some cases the response of public higher ed institutions was also semi-privatization, usually admission of out-of-state and foreign applicants at higher-than-local sticker prices for tuition. Typically, however, the actual decisions to raise tuition and/or cut enrollments (or to semi-privatize) were made – not by the legislators and governors who cut the budget – but by the immediate authorities who run public higher ed institutions. So, conveniently for legislators and governors, blame was deflected to those authorities. 


    They made the choice.  While the Great Recession is over, its after-effects linger. Public higher ed authorities – having been cast as the villains in the tuition/enrollment/semi-privatization episode – must now appeal to already offended voters for funding restoration and support. Higher ed authorities may feel that it is unfair to have to shoulder the blame, but that is the reality. They can only go so far in trying to point fingers at legislators and governors since neither are anxious to assume blame, even retroactively.  And both are needed, along with voters, for support. 


    I am most familiar with the case of California, which has the image of a diverse “blue” state that takes generally liberal positions. So let’s look at voters there. You might expect a greater degree of public sympathy for higher ed in California than elsewhere because of its blue reputation. However, it ain’t necessarily so. Like a lot of things, it depends on perceptions.  


    The last gubernatorial election in California was held in November 2014, but the outcome was known well in advance. Incumbent Jerry Brown was expected to win reelection by a large margin. Under those circumstances, with no real contest at the top of the ticket, voter turnout was expected to be low (and it was). So voters who did turn out were presumably biased toward those in the electorate most
    interested in public affairs. 


    In order to predict the results of elections, the California Field Poll attempts to focus on those in the public who actually will vote. A few weeks before the November 2014 election, it polled what it considered to be a sample of “likely voters.”[2] What was the demographic and political makeup of that sample? The table below provides a summary that may surprise. 


    Democrats                43%   
    Republicans              34%
    No party/other          23%
    ===
     
    Strongly conservative      20%
    Moderately conservative  11% 
    Middle-of-the-road          41%
    Moderately liberal           11%      
    Strongly liberal               17%
    === 


    Age 18-29                 11% 
    Age 30-39                 14%
    Age 40-49                 16%
    Age 50-64                 32%  
    Age 65+                     27%
    ===


    White, non-Hispanic      70%
    Latino                          16%
    African-American            6%
    Asian/other                    8%
    ===
     
    Male                       50%
    Female                   50%
    ===
     
    Union household            18% 
    Nonunion household       82%
    ===


    If you want to characterize the median California voter – whose support presumably public higher ed institutions want - that voter is white, nonunion, age 50+, middle-of-the-road politically, independent, and equally likely to be male or female.[3] So the key to political success in California is definitely not denigrating or offending older white males. Other poll data suggest that the median likely voter has just a bachelor’s degree, i.e., 50% of likely voters have educations below that level, 50% have educations at that level or above.[4] 
     
    The notion that California is inherently “progressive” on social issues isn’t suggested or supported by the history of state ballot propositions over the past quarter century despite its blue state reputation. Consider the following election results:
     
    Prop 187 – Ban on public services for undocumented immigrants (passed 1994)
    Prop 209 – Ban on affirmative action in public higher ed admissions and state contracting (passed 1996)
    Prop 227 – Sharp limits on bilingual education (passed 1998)
    Prop 22 – Ban on gay marriage (passed 2000)
    Prop 8 – Ban on gay marriage (passed 2008)
     
    Clearly, if some of these propositions were on the California ballot today, they would not pass. Attitudes do change over time.[5] But to the extent that California – despite its blue state image – is on the leading edge of emerging causes, that leadership is more likely to be true in the environmental area rather than when it comes to social attitudes.
     
    Much of the latest social agitation in higher ed, including in the public institutions of California, has involved such matters as microaggressions, statements – perhaps inadvertent – that might offend. A good deal of this agitation has developed within universities.  It isn’t coming from median voters who  aren’t preoccupied with microaggression, but who do have concerns about tuition and access – based on all those labor market predictions that you must have a college degree in the future. 
     
    Those voters are not committed to public higher ed institutions as centers for promoting social change as California’s ballot history suggests. Particularly given the coarse discourse readily found in everyday political debate, the internet, popular entertainment, etc., what is characterized as a microaggression in university circles seems mild to anyone with a TV or laptop. Ten or twenty years from now, perhaps voters of that future period may have changed their views. But for now, issues such as tuition, access, and student debt are the big issues for higher ed. In contrast, a focus on other matters by those folks running public higher ed institutions is likely at best to appear off-topic and unresponsive to the concerns of the median voter. 
     
    Put another way, being off-topic and unresponsive may be viewed by median voters as a microaggression against them, what they think, and how they talk. And there are consequences if that is what voters come to perceive about public higher ed and believe is going on there. Within academia, there seems to be a body of psychological research on microaggression in the context of interpersonal interactions. It goes along with longstanding research on framing and hidden prejudices. Continued research of that type should be encouraged. But the research so far seems to lack an outward component when it comes to application to higher education.
     
    More precisely, what is odd is not the research in the abstract, but its policy consequences within higher ed institutions. There is much effort at documenting the impact of the microaggression controversy on everyone except those median voters on whose goodwill the fate of public higher ed institutions depends. Put another way, there seems to be great concern about the impact of what might be said within the institution. But there is no concern – or even perception – as to what the impact might be when bureaucratic university policies on microaggression leak outside the institution. 
     
    Thus, when a University of California guide that indicated that asking people where they are from is equivalent to telling them they aren’t “true American(s)” is discovered, and is (predictably) circulated on the internet, the guide – and the official “seminars” at which it was used - become a target of ridicule and offense.6 Did the University really believe that someone saying America was the “land of opportunity” was a micro-aggressor? And, no, it’s not just right-wing news media that pick up such stories. That episode found itself quickly aired in the mainstream.[7] It creates the image of academic administrators gone amuck with political correctness at a time when they should be focused on access and affordability.
     
    Ultimately, the idea that through official policy speech should be constrained so it never offends anyone within the institution, while at the same time its impact on outside political constituents should not only be ignored, but not even recognized, seems bizarre. It is even more bizarre in a world in which those median voters on the outside are being told that the key to labor market success is a college degree and  that their political support is thus needed to fund public higher ed. In an era of economic insecurity, where is the research on the impact of university-generated microaggression against the median voter? Where are the seminars on the external impact?


    [1] http://www.bls.gov/news.release/ecopro.t08.htm

    [2] http://www.field.com/fieldpollonline/subscribers/Rls2489.pdf

    [3] Note that there is a big difference between the general population and the likely voter population. Children don’t vote. Non-citizens don’t vote. Those eligible to vote have to register and then turn out. 
    [4] http://www.field.com/fieldpollonline/subscribers/Rls2490.pdf
    [5] Not all the propositions would necessarily be reversed today. A move in the state legislature to put a proposition on the ballot repealing Prop 209 (affirmative action) not long ago was quickly killed when the Asian community – that felt its kids would be disadvantaged by repeal – vocally objected.  

    [6] http://www.ucop.edu/academic-personnel-programs/_files/seminars/Tool_Recognizing_Microaggressions.pdf  . The official university position seems to be that speech was not being forbidden but that attendees at the seminar
    were being sensitized. 
    [7] http://www.scpr.org/programs/airtalk/2015/06/24/43412/microaggressions-should-they-be-censored-on-colleg/  

  • 31 Aug 2015 5:34 PM | Daniel J.B. Mitchell (Administrator)

    Last week’s “volatility” in the stock market produced predictable results. There was an attempt in the news media to find the “cause.” The cause de jour was said to be investors’ concern over the economic slowdown in China and the devaluation of the Chinese currency. But wait! As we have noted in prior musings, the devaluation was intended to stimulate Chinese net exports and thus stimulate the Chinese economy. So shouldn’t investors have taken the devaluation to be a Good Thing, if they were worried about the growth rate of the Chinese economy? More importantly, if the stock market had gone up rather than down, wouldn’t the news reports have attributed the rise in the market to the devaluation’s effect on improving the Chinese economy? We could go on with this theme. To this author, these stories are what might be called elevator tales.

    What’s an elevator tale? In the building on the UCLA campus in which I have my office, the elevators have never worked well. One of the four elevators in the building is often out of service, although which one it is varies. And when the elevators are working, they do mysterious things such as reverse direction before getting to your desired floor or arriving at a floor without opening the door. Faced with the inexplicable over the years, I have heard fellow passengers express theories about elevator behavior. Examples: You have to press twice or more to ensure getting to your floor. Jumping up and down in the car will ensure that the door will open. (I always offer the explanation, if asked, that the underlying problem is that the donkey in the basement who pulls the rope didn’t get enough carrots.) In short, whether it’s elevators or financial markets, having stories to explain the inexplicable gives folks a sense of security and control.

    There is a chronic description problem when the stock market’s gyrations are discussed. The market went down because everyone was selling. If that were the case, one might ask who they were selling to. Isn’t every stock sold also a stock bought? So couldn’t we just as well say that everyone was buying? In the end, this type of analysis is saying nothing more than that the market went down today because the net consensus among traders was that the price should be lower today than it was yesterday. Some traders thought the new price was a good deal and bought at the new, lower level. Some thought it was a bad deal and sold. Everyone didn’t do anything. Some folks bought; others sold. (And some of these “folks” were programmed computers.)
     
    Despite the fact that stock markets are an amalgam of many traders (and trading computers), the ups and downs are routinely interpreted as if the market was a single struggling individual with bouts of optimism and pessimism. After stumbling, the market was trying to recover. The market was taking a corrective action. Such modern anthropomorphism is no more helpful to understanding gyrations in the market than were old explanations of natural phenomena as the whims of human-like rain gods and the like.[1]

    Finally, there was advice in the news media from “experts” to actual ordinary folks (not computers) with their life savings in 401k and similar plans and in defined-contribution pensions. Typical advice was not to sell because the market goes down since a saver’s horizon – retirement – is off in the future. But wait! Telling folks not to sell in a market downturn because the downs will be followed by ups is inconsistent advice. If you as an expert financial advisor know for sure that the down will be followed by an up, shouldn’t you be advising folks to buy, not just hold? If there is certain to be an up, you will surely gain by buying, no? And, by the way, if the down is sure to be followed by an up, why was there a down in the first place? Was the market god angry?
     
    When there is “volatility” in the stock market, the issue of pensions and pension finance inevitably comes up. Even apart from last week’s turmoil, over the past year, broad stock market indexes such as the S&P 500 have essentially gone nowhere. So defined-benefit pension funds – which generally are built upon assumptions of long-run annual earnings of ±7.5% - have been reporting returns over their past fiscal year of much less than that. In some cases, such plans have cut their assumed future long-term rates of return to 7.25% or less.

    Which brings me to what should have been learned about pensions from last week’s stock market events, but wasn’t. Yes, defined-benefit pensions have to deal with volatility, changes in the long-term outlook for likely returns, uncertain projections about life expectancy and inflation, etc. They can  become underfunded if errors in such forecasts are made. But they have one attribute that defined-contribution plans, 401ks, etc., don’t have; they are collective. They pool risks because many people are covered. The many people are not only those enrolled at a point in time but also over time. Old folks retire; new folks come into the plan.


    The fact is that all of the factors that affect defined-benefit plan funding are also present in defined-contribution plans. If your individual rate of return turns out to be less than you assumed, you won’t have enough money for retirement. If you live longer than you expected, you may run out of funds. The difference is that without risk pooling, the problems caused by incorrect assumptions are more severe to the individual than when there is risk pooling. If you run out of funds, you can’t go back in time and redo your behavior. In contrast, a large plan with many individuals can make adjustments and corrections iteratively over a long period.


    There is much research indicating that people are not very good at long-term planning and investment strategies. There are ways to mitigate some of these issues such as default opting into job-based saving plans and the offering of “life-cycle” investment options in such plans. And there are problems with having defined-benefit pension funding depend on the long-term good economic health of a particular employer. But the loss of risk pooling as the private sector has moved away from defined benefit and towards defined contribution pensions is a problem. The push to have the same thing happen in the public sector is intensifying the problem. And moves to privatize Social Security – individual accounts that move away from risk pooling – would be a disaster. In the end, Social Security is the ultimate collective, risk-pooling pension plan.


    Too bad that when “everyone” was selling last week, they didn’t have time to consider the virtues of risk-pooling. Too bad that when the market was “trying” to recover, it didn’t think about the lessons for collective vs. individual pension systems. Maybe next week, the market god will be in a more contemplative mood.


    ---------------------------------             
    [1] Los Angeles Times  columnist Michael Hiltzik put it nicely: "Stocks staging a stunning
    comeback," declared anchor (CNBC) Amanda Drury around 1:45 p.m. Eastern. A few hours later, one of her colleagues, sounding like a play-by-play announcer at the World Cup, announced that the Dow Jones industrial average were "trying to recover from an early 1,000-point plunge." The truth, obviously, is that as the reflection of millions of individual investment decisions along with algorithm-based trading, the markets don't "stage" anything.

    See http://www.latimes.com/business/hiltzik/la-fi-mh-market-turmoil-and-the-problem-of-cnbc-20150824-column.html

  • 24 Aug 2015 2:48 PM | Daniel J.B. Mitchell (Administrator)

    Of late, political pundits are trying to understand the appeal of Donald Trump on the Republican side of the presidential race and, to a lesser degree, Bernie Sanders’ appeal on the Democratic side. Both are typically viewed as “populists” who have a limited appeal to a base audience in their respective parties. One approach is to depict the two candidates as mirror images, one on the left and the other on the right, who are somehow similar.


    For those pundits who focus on Trump, there is a notion that his appeal is to older white males who became economically outmoded due to some inevitable force – Technology? Inexorable trade competition in a global economy? – and therefore have fading and anachronistic concerns about disappearing jobs. It’s harder to put Sanders in that category, however, because he hasn’t made anti-immigrant comments whereas Trump has.

    I am neither a pollster nor a political scientist so I won’t pursue the question of whether the two candidates are appealing to similar groups or not. What I will say is that if your image of a disenchanted voter is an aging white male manufacturing worker, or perhaps a white male displaced from that sector, you might want to take a look at manufacturing demographics.


    Let’s start with the male/female breakdown. For the entire employed workforce in 2014, almost 47% were women. In the manufacturing sector, a little over 29% were women. So, if you had the impression that there were more men than women in manufacturing relative to the overall workforce, you were correct. But note that almost 3 out of 10 workers in manufacturing were women. Would you really want to take the position that those female workers aren’t concerned about their jobs and/or potential displacement from those jobs?



    What about age? Again, if your preconception is that a typical manufacturing worker is older than the typical worker in the overall workforce, you are correct. However, as the chart above shows, the difference is not striking. The median age of a worker in manufacturing is a couple of years older than that found in the overall workforce, but the share of older workers in manufacturing really isn’t all that different. Presumably, the younger workers in manufacturing care about job loss or the threat there from.




    As for the race/ethnic cut, yes, there is a slightly lower percentage of blacks and Latinos in manufacturing compared to the overall workforce, but – again – the difference, shown on the chart above, isn’t sharp.[1] There are actually somewhat more Asians proportionately in manufacturing than in the workforce as a whole. Of course, manufacturing is a broad term and at a detailed level, the various minority percentages vary. For example, blacks are more heavily represented in motor vehicles and Latinos are more heavily represented in furniture manufacturing, Women form a slim majority of textile and apparel manufacturing workers.


    What we can say about manufacturing workers is not that they are totally different in their response to political appeals that connect to their jobs because of their demographics, but rather that they are responsive because neither political party has been sensitive to their plight. Would a young Latina worker in manufacturing not be concerned about possible layoffs or displacement simply because she isn’t an older white-Anglo male? As long as there is an impression among the political elite that there is nothing to be done about jobs in manufacturing because blind forces of technology and trade are at work, there will be political space for “populist” appeals to such workers, regardless of age, sex, race, or ethnicity. As long as there is an elite stereotype that manufacturing is composed of just aging white males who are a declining share of the electorate, and that therefore concerns of the workforce in that sector don’t matter, there will be such political space. And, no, just telling everyone to go to college, and even offering plans to subsidize going to college, isn’t a political solution – or even an economic solution – to the problem.


    ------------------------------

    [1] The bars on the chart for white-Anglo workers are estimated as residuals after subtracting blacks, Asians, and Latinos. That approach slightly underestimates the white-Anglo group because a relatively small percentage of Latinos also identify as black. But the chart is reasonably accurate for purposes of this musing. All charts and data are from the U.S. Bureau of Labor Statistics.

  • 17 Aug 2015 2:55 PM | Daniel J.B. Mitchell (Administrator)

    From time to time, these musings have talked money, specifically currency exchange rates. There has been much news media discussion in the past week about the recent China devaluation of the Yuan. So let’s start with terminology. A freely floating currency – one that is not “manipulated” – sometimes rises in value relative to other currencies and sometimes falls. It sometimes appreciates and sometimes depreciates. That’s the terminology we use for those movements when they occur due to market forces. No government or central bank or any other official agency is responsible. Appreciation or depreciation just happens due to market conditions.


    In contrast, the word “devaluation” implies that a policy decision was made to change (lower) a currency’s value relative to another. The word devaluation, put another way, implies “manipulation.” (The opposite is “revaluation.”) You can’t talk about a “devaluation” and then ruminate over whether manipulation has occurred. It has occurred by definition.


    Now there is a longstanding history in international monetary affairs about the pluses and minuses about systems of fixed exchange rates, freely floating exchange rates, and arrangements that fall somewhere in between. The old gold standard was a system of fixed exchange rates. The Bretton Woods system set up towards the end of World War II was a fixed exchange rate system. When Bretton Woods fell apart in the early 1970s, a mix of arrangements developed. The U.S., however, largely left its dollar to float. Some countries attempted to maintain fixed arrangements relative to one another but float against others. Some tried to keep their currencies within a band of some other currency. The Eurozone eventually formed and some countries abandoned their internal currencies for a common, international currency. There were experiments with “currency board” systems in which a country pegged its currency to another through a kind of central bank operating by formula.


    To the extent that a country chooses to have some say in its currency’s value, and not leave the exchange rate entirely to market forces, there has to be some regime of regulation and/or intervention in currency markets, i.e., “manipulation.” You can debate whether the result of such manipulation is a Good Thing or a Bad Thing, but (again) that there is manipulation taking place is not a matter for debate.




    When you look at China’s trade with the U.S. as shown on the chart, there is an anomaly. You have a rich country – the U.S. – borrowing from a developing country, essentially to finance current consumption. As the chart above indicates, that odd situation has persisted for a long time. The U.S. trade deficit with China now is close to 2% of U.S. GDP. That figure may not sound like much. But, to put it in perspective, the peak-to-trough drop in U.S. real GDP during the Great Recession was around 4%. So the impact of 2% is hardly negligible.


    China, from time to time, says that it wants to move (gradually) to a floating exchange rate. But the chart surely suggests that what it has been doing is maintaining an undervalued currency. The recent devaluation was said to be part of a move to a market exchange rate. Numerous journalists repeated that interpretation. But any such move would have to be a revaluation (increase in the Yuan’s value), not a devaluation.


    Moreover, as noted earlier, the very word “devaluation” implies an official policy, not some blind market force. So it’s hard to get away from the fact that China views the exchange rate as a macroeconomic tool, not something to be left to the forces of currency markets. It “manipulates” its currency value. Actions speak louder than words, although apparently not to those journalists who repeated the self-contradictory move-towards-the-market story. The Chinese economy was slowing down and to stimulate demand, the Yuan was devalued by the powers-that-be in China. It isn’t complicated. No elaborate interpretation is needed. And it’s a move away from the market, not towards it.


    What is the impact on the U.S.? Again, the story is not complicated. If boosting Chinese exports to the U.S. and discouraging Chinese imports from the U.S. is a stimulus to China’s economy, it has to be a negative, other things equal, for the U.S. Commentators quickly chimed in to say that, yes, there is a negative effect, but it will be small. After all, the trade deficit with China is only 2% of the U.S. GDP and the devaluation will only have an incremental effect on that pre-existing deficit.
    The problem, however, is that for a country such as the U.S., marginal shifts in trade patterns are always in some sense small in their overall impact. But they can be large in individual industries or sectors. As we have pointed out in past musings, manufacturing – which is now only about an eighth of U.S. GDP - is especially affected by trade shifts since much of trade involves manufactured goods. So what happens in trade and exchange rates matters to manufacturing and to jobs in manufacturing.


    Rather than discuss what the Chinese devaluation means – when it’s obvious what it means – isn’t it time to revisit U.S. exchange rate policy, a policy discussion that really hasn’t occurred since the early 1970s when fixed exchange rates were abandoned? At that point, the U.S. essentially said it would not intervene in exchange markets in order to affect the value of the dollar, but that other countries could do what they liked. The problem with that approach is that an exchange rate inherently involves two currencies; it is the price of one currency relative to another. So the decision to float the dollar and let others do what they liked was essentially a de facto decision to let other countries “manipulate” the value of the dollar, if they so wanted. It might have been the best policy choice back then. But over four decades later, it’s time for a review.

  • 10 Aug 2015 2:17 PM | Daniel J.B. Mitchell (Administrator)

    Robert Lawrence of the Peterson Institute posted a video which purports to resolve the issue of why real wages have lagged productivity since the 1970s.[1] He starts with a chart showing a gap opening up between average hourly wages of production and nonsupervisory workers deflated by the Consumer Price Index (CPI) and output per hour (productivity) as measured by the U.S. Bureau of Labor Statistics (BLS). In steps he adjusts the real wage series by adding in employees other than nonsupervisory workers, taking account of benefits received by workers (which are not included in the average hourly earnings series) and noting that the price index used to deflate the output numerator in “output per hour” differs from the CPI and that the former rises slower than the latter starting in the 1970s. So if you use wages and benefits for all workers and if you deflate those wages by the deflator for output rather than by the CPI, the puzzle disappears except for the period after the Great Recession.


    It’s worth noting that there is no law of the universe that says that real wages (however measured) must rise with productivity (however measured). The idea that the two series should be linked derives from the observation that they appeared to be moving together after World War II as an empirical matter. Furthermore, there seemed to be two notions that there should be a linkage beyond the mere empirical observation. To explore the proposition, let’s represent the idea in the abstract:


    Let W = a measure of nominal wages, P = a general price index, Y = a broad measure of national output in nominal terms, and L = labor hours. Saying real wages rise with productivity is equivalent to saying:


    W/P = s(Y/P)/L, i.e., the real (deflated) wage (W/P) is proportionate to real (deflated) output (Y/P) divided by labor hours input L and where s is a coefficient of proportionality.


    Note that you can rearrange these terms to become s = WL/Y, i.e., s turns out to be labor’s relative share of national output. So the assumption that real wages rise with productivity is another way of saying that labor’s relative share of national output is constant. Note, for later reference, that this rearrangement is entirely in nominal dollar terms; there is no price index involved.


    Some observers see (or saw) a moral element in having real wages rise with productivity; some see (or saw) a moral element in labor’s relative share being a constant. In the former case, there seems to a Puritan Ethic-type morality behind the idea that the way workers get ahead is through producing more. Work harder and you will advance! In the latter case, perhaps it is just seen as fair that labor and capital each share proportionally as the economy grows. I am not saying that these are good ways to look at the relationship; only that there is a certain appeal to the concept from various moral angles.


    There is also an historic link to the history of wage-price controls and guidelines and the real wage-productivity relationship. We can also rearrange our starting equation as:
    P = (1/s)[WL/(Y/P)], where the term in brackets [ ] is the average wage cost of a unit of real output or what is called “unit labor costs.” If s is a constant, so is 1/s, and the revised equation says that prices are proportionate to unit labor costs. An interpretation is that firms use some kind of markup pricing when aggregated so that if you can set (or limit) the rise of unit labor costs, you can set (or limit) the rate of inflation. Control wages, the nominal element in unit labor costs, and you can control prices. Crudely, your guideline for nominal wage increases should be your target rate of inflation plus the expected long-term rise in productivity. Such rules were used in the Kennedy-Johnson wage-price guideposts program and the Nixon-era wage-price controls.


    Note that there is nothing about wage equality or inequality involved in these notions. And it is more or less assumed that W is an aggregate measure (an average wage of everyone) and that if some element of pay comes in the form of benefits, it is included in W. Similarly, it is more or less assumed that P is a general measure of prices and that it is used both to deflate output and to deflate wages. Since P is an average, nothing precludes some prices from rising faster than others. Since W is an average, nothing precludes particular wages, say for certain occupations or groups, from rising faster than others.


    The idea that real wages either should, or do, rise with productivity in the abstract doesn’t deal with inequality of wage growth within the workforce and certainly includes payment for labor in the form of benefits. So let’s take a look in the chart below at the BLS data set that most closely adheres to the broad concept. Such a data set can be found in the various series connected to output-per-hour (productivity). The price index used to deflate wages (which include benefits and pay for all workers) is the Consumer Price Index. The broadest sector available is the “business” sector which is essentially all private business plus government enterprises that are quasi-commercial such as the Postal Service, transit operators, etc.




    It is well known that productivity, as measured by BLS, has a cyclical component so the chart above uses business cycle peaks (except for the latest available year, 2014) to adjust for such effects.[2] The real wage and the productivity measures do seems to diverge starting in the 1970s, although pinning down the precise year in which the divergence occurred would be difficult since the two data series never moved precisely in lock step.




    It’s also true, as Lawrence noted, that much of the divergence seems to be based on the faster growth of the CPI relative to the deflator used for determining real output, as can be seen on the chart above. However, Lawrence seems to take the two indexes to be “true” for their different purposes. That is, the CPI is supposed to be truly a valid measure of worker consumption over the decades while the deflator used to turn nominal output into real output is truly valid for that purpose.


    But there are problems in assuming, particularly over long periods, that abstract concepts of worker welfare or estimates of aggregations of the diverse outputs of a complicated economy in real terms are somehow uniquely defined. Consider the CPI. It has undergone various methodological changes over the period shown on the charts. Yet, because it is used for indexing in legally-enforceable contracts, BLS never revises it retroactively since that would upset its consistent history. Instead, one methodological version is spliced onto the previous version going forward.[3]


    For example, during the 1970s, the BLS measure of housing costs was determined by a methodology that gave heavy weight to (mortgage) interest rates. Before the 1970s, such rates did not fluctuate much but then, in part because of a pickup in inflation, the rates began to move. Eventually, a different methodology base on rental equivalents of owner-occupied housing was installed, but not retroactively.


    And there have been other such changes, especially with regard to substitution effects and quality adjustments.


    The deflator used for output by BLS is really part of the national income accounts. There, too, methodology has changed over time, but unlike the CPI, such changes are often incorporated retroactively back to arbitrary dates. And the methodological changes introduced, while they are aimed at actual theoretical problems, are typically chosen from a set of reasonable approaches. Put another way, there are alternatives which might have been chosen that would have produced different results.


    In short, it is hard to say when you look at the divergence between the official measure of real wages and the official measure of productivity, what the question(s) should be. Saying the divergence is largely due to workers’ typical consumption baskets somehow systematically differing from the output basket starting in the 1970s assumes that we have the “right” price indexes for both of the baskets. But maybe productivity isn’t growing as fast as the official measure suggests. If the “true” price index is more like the CPI and less like the official deflator, measured productivity would rise more slowly. It all depends on how much faith you have that we have the right price indexes.


    We can, however, take the abstract concept that real wages should, or do, or used to, rise with productivity and get rid of the uncertain price index element entirely. As we noted above, that concept is equivalent to saying that labor’s relative share of output – at least adjusted for the business cycle – is more or less constant. Labor’s share and output can be measured in nominal terms; no price index is required. We don’t have to worry about price index methodology. So let’s look at the share over time.




    As the chart above shows, the share seems to have started slipping in the 1960s. It flattened in the 1980s and staged a partial comeback in the 1990s. (Did high tech and finance sector pay hikes during the dot-com boom cause the partial reversal?) Then labor’s relative share declined, notably starting BEFORE the Great Recession took effect, and continued to decline thereafter.


    In short, if I had to choose a research project based on these observations, I wouldn’t focus on why worker consumption basket prices differed from output basket prices – because there are too many iffy methodological adjustments in our price indexes. I would instead focus on an issue that doesn’t depend on price indexes at all. What explains the movement, adjusted for the business cycle, of labor’s relative share? Why did it start to decline in the 1960s? What gave it a temporary partial boost in the 1990s? And what happened to the share after the end of the dot-com boom?


    [1] http://blogs.piie.com/realtime/?p=5132. Joshua W. Mitchell alerted me to this source.

    [2] We start the chart in 1953 to avoid effects of World War II wage-price controls and Korean War controls. There was a double-dip recession after 1979 so we skip the middle “peak” of that episode on the chart. Otherwise, peaks are based on NBER business-cycle dating.



Archive

12/28/15
12/29/14
12/30/13 12/24/12 12/26/11
12/21/15
12/22/14
12/23/13 12/17/12 12/19/11
12/14/15
12/15/14
12/15/13 12/10/12 12/12/11
12/7/15
12/8/14
12/9/13 12/1/12 12/5/11
11/30/15
12/1/14
12/2/13 11/26/12 11/28/11
11/23/15
11/24/14
11/25/13 11/19/12 11/21/11
11/16/15
11/17/14
11/18/13 11/12/12 11/7/11
11/9/15
11/10/14
11/11/13 11/5/12 10/31/11
11/2/15
11/3/14
11/4/13 10/29/12 10/24/11
10/26/15
10/27/14
10/28/13 10/22/12 10/17/11
10/19/15
10/20/14
10/21/13 10/15/12 10/10/11
10/12/15
10/13/14
10/14/13 10/8/12 10/3/11
10/5/15
10/6/14
10/7/13 10/1/12 9/26/11
9/28/15
9/29/14
9/30/13 9/24/12 9/19/11
9/21/15
9/22/14 9/23/13 9/17/12 9/12/11
 9/14/159/15/14 9/16/13 9/10/12 9/5/11
9/7/15
9/8/14 9/9/13 9/3/12 8/29/11
8/31/15
9/1/14 9/2/13 8/27/12 8/22/11
8/24/15
8/25/14 8/26/13 8/20/12 8/15/11
8/17/15
8/18/14 8/19/13 8/13/12 8/8/11
8/10/15
8/11/14 8/12/13 8/6/12 7/25/11
8/3/15
8/4/2014 8/5/13 7/30/12 7/18/11
7/27/15
7/28/14 7/29/13 7/23/12 7/11/11
7/20/15
7/21/14 7/22/13 7/16/12 7/4/11
7/13/15
7/14/14 7/15/13 7/9/12 6/20/11
7/6/2015
7/7/14 7/8/13 7/2/12 6/13/11
6/29/15
6/30/14 7/1/13 6/25/12 6/6/11
6/22/15
6/23/14 6/24/13 6/18/12 5/30/11
6/15/15
6/16/14 6/17/13 6/11/12 5/23/11
6/8/15
6/9/14 6/10/13 6/4/12 5/16/11
6/1/15
6/2/14 6/3/13 5/28/12 5/9/11
5/25/15

5/27/13 5/21/12 5/2/11
5/18/15

5/20/13 5/14/12 4/25/11
5/11/15

5/13/13 5/7/12 4/18/11
5/4/15

5/6/13 4/30/12 4/11/11
4/27/15

4/29/13 4/23/12 4/4/11
4/20/15

4/22/13 4/16/12 3/28/11
4/13/15

4/15/13 4/9/12 3/21/11
4/6/15

4/8/13 4/2/12 3/14/11
3/30/15

4/1/13 3/26/12 3/7/11
3/23/15

3/25/13 3/19/12 2/28/11
3/16/15

3/18/13 3/12/12 2/21/11
3/9/15

3/11/13 3/5/12 2/14/11
3/2/15

3/4/13 2/27/12 2/7/11
2/23/15

2/25/13 2/20/12 1/31/11
 2/16/152/17/14 2/18/13 2/13/12 1/24/11
2/9/15
2/10/14 2/11/13 2/6/12 12/15/10
2/2/15
2/3/14 2/4/13 1/30/12 12/9/10
1/26/15
1/27/14 1/28/13 1/23/12
1/19/15
1/20/14 1/21/13 1/16/12
1/12/15
1/13/14 1/14/13 1/9/12
1/5/15
1/6/14 1/7/13 1/2/12
Employment Policy Research Network (A member-driven project of the Labor and Employment Relations Association)

121 Labor and Employment Relations Bldg.

 

121 LER Building

504 East Armory Ave.

Champaign, IL 61820

 

The EPRN began with generous grants from the Rockefeller, Russell Sage, and Ewing Marion Kauffman Foundations

 

Powered by Wild Apricot Membership Software