In the United States, wages in real terms have seen very little change over approximately five decades:

from $19.18 in 1964 to $20.67 in 2014. Measured in 2014 dollars.

(source: http://www.pewresearch.org/fact-tank/2014/10/09/for-most-workers-real-wages-have-barely-budged-for-decades/ )

The reasons for this are many and varied.

Over approximately the same time period, union membership has fallen dramatically. BLS data since 1983 shows the following chart

union membership from 20.1% in 1983 to 10.7% in 2017

(source: data roundup via https://www.bls.gov/news.release/union2.nr0.htm with the same chart in a more accessible article at https://www.bloomberg.com/news/articles/2018-01-19/union-membership-rate-in-u-s-held-at-record-low-of-10-7-in-17 ).

The reasons for this too are many and varied.

I think the causal relationship between the two is weak, but I'm wondering if there has been any research on whether or not there is a relationship, and to what degree.

Repeated, in question form: What is the relationship, if any, between declining union membership and real wage stagnation? Thank you.

Edit: UK data.

The UK has a report on union membership: https://www.gov.uk/government/statistics/trade-union-statistics-2016 . One table lists "Trade union membership, UK, 1892 to 2015-16" in absolute terms; the series counting percent of employees only starts in 1989. But it shows that absolute membership peaked in 1979 and has been declining since.

Trade union membership, UK, 1892 to 2015-16

In contrast with the United States, ever since the (absolute) union membership peak in 1979, real wages have increased in the UK (actually shown is since 1980), although they did take a dip after 2008.

Annual Survey of Hours and Earnings, 1980 to 2016

(source: "Real Wages and Living Standards in the UK" )

  • $\begingroup$ I feel like this is more of a political question, but I'm not sure it's a good fit for politics.SE $\endgroup$
    – BurnsBA
    Jun 3, 2018 at 0:53
  • $\begingroup$ "Approximately the same period" when the union participation data starts 20 years later? $\endgroup$ Jun 3, 2018 at 21:04
  • $\begingroup$ @AlecosPapadopoulos Yes, union membership was declining before 1983, but I didn't see BLS data going back further. But see, e.g. en.wikipedia.org/wiki/Labor_unions_in_the_United_States#/media/… $\endgroup$
    – BurnsBA
    Jun 4, 2018 at 12:15
  • $\begingroup$ Have you looked at whether the hypothesis would hold for other developed countries in the same period? $\endgroup$
    – Dan
    Jun 4, 2018 at 16:43
  • $\begingroup$ Wages and legal negotiation authority of shareholder unions and labour unions can be discussed in completely non-economic terms. For example, they can be discussed as issues of representation rights of one or more groups, in an explicitly political framing. More generally, wages are an economic question, even if political. Relative bargaining authority of shareholder unions and labour unions are fundamentally economic questions, even if political. $\endgroup$
    – nathanwww
    Jun 7, 2018 at 19:37

2 Answers 2


My understanding is that theories and empirical methodologies demonstrating the relationship between declining union membership and wages risk overestimating the causality of the relationship.

Anyone who argues that businesses do NOT, in a relevant share of cases, endeavour to suppress union formation, activities or recruitment, whether individually or collectively, is either ignorant or a political shill. (Things are not nice 100% of the time in the other direction either.) This should be recognized, in order to proceed with rational discussion of the economics of the issue.

However, factors like increased international competition in lower-skilled jobs from a very large international labour pool (which is much more accessible to producers than in 1950 or 1980) can be one reason that the causality of declining union membership can be overestimated.

Also, mechanisms of tacit collusion between businesses can act to hold down wages as well, for example with various agreements which prevent an employee from working for a competitor for some number of years after they leave (e.g., Google and Apple could both force employees to sign such an agreement as a condition of employment, thus holding down wages for both firms, which both account for a substantial share of employment in their specific areas of activity). This could be another explanation for some portion of wage stagnation that would be difficult to frame as resulting from declining union membership per se.

Between 1) international competition and 2) a proliferation of methods to suppress wages unrelated to union formation or membership per se, it should suffice to say that significant effort for robustness must be demonstrated before any empirical work can be taken seriously when discussing causality between declining union membership and 50 years of stagnant average wages.

The risk, then, is that an inadequate methodology which involves a) under-specification of a model, and/or which b) includes inadequate robustness checks, could indicate that declining union membership caused stagnant wages, when theoretically this could be a false positive when other factors explain both the declining union membership and also stagnant average wages.


The direct correlation is tricky.

Reason 1: Employees in the union have significantly higher wages than those not in a union. There is not really any debate about this. But...

Reason 2: Employees without a union in the vicinity of an employer which allows/has a union sees a significant decrease in their overall income.

So, overall it may have no net effect.

I hope this helps!

  • $\begingroup$ I'm not sure your second assertion is true; I thought it was generally accepted that non-unionised workers in a company or industry that was more unionised tended to have higher wages than those in less unionised environments. See Booth's "The Economics of the Trade Union" for example. $\endgroup$
    – Dan
    Jun 4, 2018 at 16:28
  • $\begingroup$ I was taught that since unions raise wages, more people enter the industry. That changes the supply and demand of labour and brings down the equilibrium wages for people not in the union (people in the company which does not have a union). But there may be more to that... $\endgroup$ Jun 5, 2018 at 14:37

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.