Popular in Social Sciences

Political Science

V-Dem’s Conceptions of Democracy and Their Consequences

University of Notre Dame Michael Coppedge

Central Monetary Authorities

Macroeconomic modelling of CBDC: a critical review

European Central Bank Ulrich Bindseil

Social Identities and Divisions

A new explanation of antisemitism: Jew hatred as a civilisational phenomenon

Boston University Liah Greenfeld

Central Monetary Authorities

Financing Private Credit

Federal Reserve Bank of New York Nina Boyarchenko

Economics

Failing Banks

Massachusetts Institute of Technology Emil Verner

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

American Politics

Performance-related pay, mental and physiological health

View more insights in Social Sciences - Economics - Labor Economics

View more figures from University of Aberdeen

View the page of Keith Bender

Read the Study

Next in Economics

Do renewable energy investments create local jobs?

Natalia Fabra | University Carlos III Madrid

An early indicator for anomalous stock market performance

Thomas Gries | Universität Paderborn

Keith Bender discusses what Performance-Related Pay is and the pathways to affecting health.

Image courtesy of interviewee. February 29, 2024

Log-in or Sign-up to Faculti
Currently viewing this subject insight as a guest. You have insight(s) remaining for this month. Login to view 8000+ figures on the platform.

Generating Workspace...

Export as PDF

This video video discusses Performance Related Pay (PRP), a compensation system where employees are paid based on their performance rather than a fixed hourly or salary rate. The advantages of PRP include incentivizing employees to perform better and potentially increasing productivity and profitability for employers. However, there are significant concerns about the negative impacts of PRP on workers' health.

Historical references, including a quote from Adam Smith, indicate that performance-oriented pay can lead to overwork and health issues. Recent studies have shown a correlation between PRP and workplace accidents, particularly among blue-collar workers, as the pressure to perform might lead to risky behavior. For white-collar workers, the stress of PRP can result in unhealthy lifestyles due to the need to constantly produce.

This video video outlines three pathways through which PRP may affect health: an increase in accidents, unhealthy behavior due to work demands, and chronic stress that can lead to various health conditions. Experimental research found that PRP raises cortisol levels, a stress hormone, suggesting an increase in stress among those working under PRP systems.

While recognizing the potential health drawbacks of PRP, such as increased blood pressure and mental health issues, the text argues against eliminating PRP altogether due to its benefits for both employees and employers. Instead, it advocates for strategies to mitigate the adverse effects, such as enhancing workplace safety protocols and promoting wellness initiatives to help reduce stress among employees.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.020 --> 00:01.100
Performance related pay,

00:01.100 --> 00:05.545
or PRP is one of many forms of, of payment contracts that

00:06.145 --> 00:07.505
employers can have with employees.

00:07.695 --> 00:10.705
Most of us are used to perhaps an hourly rate of pay

00:10.925 --> 00:13.985
or a, a payment every month as a, as a salary.

00:14.125 --> 00:15.505
But for many types of jobs,

00:15.845 --> 00:19.065
you're actually paid on the basis of your performance,

00:19.125 --> 00:20.425
how much you actually produce.

00:20.845 --> 00:23.585
Um, these can be very directly tied to,

00:23.845 --> 00:25.065
to your production.

00:25.245 --> 00:27.665
Um, sometimes in economics, we call these piece rates

00:27.875 --> 00:31.465
where you're paid a particular sum for every unit

00:31.485 --> 00:32.545
of output that you have.

00:32.575 --> 00:34.385
Sometimes you're paid a commission on,

00:34.485 --> 00:35.785
say the amount of sales that you have.

00:36.125 --> 00:38.745
Um, sometimes you're paid a bonus when you hit some kind

00:38.745 --> 00:40.305
of performance target.

00:40.885 --> 00:43.985
Um, and, and so we have this sort of mix of different ways

00:44.255 --> 00:47.425
that, that workers are, are, are paid,

00:47.485 --> 00:50.185
and economics has, has long looked at all

00:50.185 --> 00:52.785
of these different types of contracts.

00:52.925 --> 00:54.385
Um, looking at both the pros

00:54.385 --> 00:58.185
and the cons of each one of these kinds of contracts,

00:58.185 --> 00:59.465
but performance related pay

00:59.645 --> 01:03.185
or PRP, is one form that that often,

01:03.485 --> 01:05.465
is talked about in terms of its advantages.

01:06.085 --> 01:09.625
Um, one of the key advantages of performance related pay is

01:09.645 --> 01:12.425
as economics research is, has shown, is that

01:12.695 --> 01:15.785
because it pays for performance, it gives incentives

01:15.845 --> 01:17.345
for workers to perform better.

01:17.645 --> 01:20.425
Um, and because of that direct link between their pay

01:20.485 --> 01:23.385
and their performance, we find that, workers

01:23.455 --> 01:27.305
that are paid under PRP, generate more product, which

01:27.305 --> 01:29.625
of course, the, the employer can turn around and sell.

01:29.885 --> 01:32.585
Um, so, so firms can be somewhat more profitable.

01:33.215 --> 01:35.625
They have this direct tie to performance.

01:35.885 --> 01:40.105
Um, it is attractive to, to workers who,

01:40.205 --> 01:43.625
who are actually really good at, at producing,

01:43.625 --> 01:46.305
because they know that they'll get a return, on this.

01:46.645 --> 01:48.905
So from an employer's point of view, economics tends

01:48.905 --> 01:50.665
to say yes, there are lots of good reasons.

01:51.545 --> 01:54.505
Interestingly, the earlier research on performance related

01:54.565 --> 01:57.065
pay also showed that workers benefited those workers

01:57.065 --> 01:58.825
that were paid under performance related pay.

01:59.055 --> 02:01.705
They got more money, so, so they benefited.

02:02.005 --> 02:03.665
Um, and, and,

02:03.665 --> 02:05.465
and so there was this, you know, sort

02:05.465 --> 02:09.305
of virtuous feedback loop, that that went back,

02:09.305 --> 02:12.185
to thinking about this kind of performance related pay

02:12.805 --> 02:15.185
as opposed to all of that research that showed all

02:15.185 --> 02:16.345
of these positive benefits.

02:16.485 --> 02:20.065
Um, there are these potential negative benefits,

02:20.065 --> 02:22.985
these negative, attributes to performance related pay.

02:23.165 --> 02:24.385
And interestingly,

02:24.385 --> 02:27.345
these were observed 200 more than 200 years ago.

02:27.685 --> 02:29.345
Um, as any good economists should do,

02:29.545 --> 02:31.745
they should look at Adam Smith and the wealth of nations

02:31.925 --> 02:33.785
and in, the wealth of nations.

02:33.805 --> 02:37.425
Uh, Adam Smith has, a quote where he says,

02:37.605 --> 02:39.465
men when liberally paid by the peace,

02:39.465 --> 02:42.025
meaning when they're paid by, by their performance,

02:42.325 --> 02:44.985
are very apt to overwork themselves and to ruin their health

02:45.005 --> 02:47.145
and constitution in a few years.

02:47.645 --> 02:51.865
So even back in 1776, Adam Smith, the father of economics,

02:51.925 --> 02:54.305
was making this linkage potentially between

02:54.815 --> 02:56.185
performance related pay.

02:56.245 --> 02:58.865
In the case, in the case of his example, it's peace

02:58.865 --> 03:01.685
Rates where you're paid directly for the output

03:01.685 --> 03:03.445
that you have and their health.

03:03.945 --> 03:06.365
So, when I read this quote, I thought, well,

03:06.365 --> 03:07.965
there must be lots and lots written about it.

03:08.145 --> 03:10.445
And indeed, there was actually very little written about it.

03:10.545 --> 03:13.085
So, over the last several years, we've been developing

03:13.185 --> 03:16.965
of various ways of thinking about what those pathways may be

03:17.125 --> 03:19.285
between performance related pay and health.

03:19.705 --> 03:21.805
So one of the pathways, is

03:21.805 --> 03:23.805
through an increase in accidents.

03:24.145 --> 03:27.285
Um, because what performance related pay does is it puts a

03:27.285 --> 03:29.045
huge emphasis on the output.

03:29.385 --> 03:32.365
So to sort of flip the, the management maximum,

03:32.425 --> 03:35.885
upside down workers are incentivized to work harder,

03:36.105 --> 03:37.525
but not necessarily smarter.

03:37.875 --> 03:39.085
They may cut corners.

03:39.265 --> 03:40.805
Um, they'll do things that

03:40.805 --> 03:42.885
that may actually injure themselves,

03:42.945 --> 03:44.045
at, at the place of work.

03:44.345 --> 03:47.245
And so, back in, in 2012, I

03:47.245 --> 03:49.885
and two co-authors wrote a paper in the journal population

03:50.125 --> 03:53.805
economics that looked at about 30,000 workers, in,

03:54.105 --> 03:57.325
in the eu, and found there was a very strong correlation

03:57.325 --> 04:00.365
between those who were paid by performance related pay

04:00.785 --> 04:03.805
and the rate of injuries, amongst those,

04:03.805 --> 04:05.045
amongst those types of workers.

04:05.545 --> 04:08.245
So, strong relationship that we found,

04:08.245 --> 04:10.165
through accidents,

04:10.165 --> 04:11.485
and was particularly important

04:11.625 --> 04:14.125
for blue collar workers or manual workers.

04:14.865 --> 04:17.245
Now, for many jobs, you know, rates

04:17.265 --> 04:18.725
of accidents are, are quite low.

04:18.745 --> 04:19.765
So for non-manual

04:19.765 --> 04:20.605
or white collar workers,

04:21.115 --> 04:22.525
it's probably not gonna be an issue.

04:22.625 --> 04:24.605
But there's a different mechanism,

04:24.605 --> 04:26.605
that can affect those kinds of workers

04:26.835 --> 04:27.965
because one of the things

04:27.965 --> 04:30.965
that we know about performance related pay as well is

04:30.965 --> 04:32.645
that it puts,

04:33.445 --> 04:35.965
explicitly a cost on not working.

04:36.505 --> 04:38.765
So in performance related pay, when you are paid

04:38.785 --> 04:41.205
by the basis of your production or your sales

04:41.505 --> 04:43.645
or something like that, when you're not working,

04:43.825 --> 04:45.365
you don't get paid, right?

04:45.385 --> 04:47.165
So there's an incentive for you

04:47.165 --> 04:49.405
to be working perhaps quite a bit so

04:49.405 --> 04:53.045
that you can produce a lot, and that increases your pay.

04:53.945 --> 04:55.605
But if you're working a lot, it means

04:55.605 --> 04:56.885
that you're not doing other things.

04:57.065 --> 04:58.845
So, instead of going

04:59.145 --> 05:03.725
and, going to the grocery store and buying, food

05:03.825 --> 05:05.965
and taking it home and making a nice, healthy meal,

05:06.185 --> 05:07.645
you order take away and,

05:07.785 --> 05:10.445
and you eat a, you know, greasy pizza on, at,

05:10.465 --> 05:12.725
at your desk while you're continuing to do work,

05:13.225 --> 05:16.885
or, instead of going, and taking a walk

05:16.905 --> 05:18.965
and doing some exercise or going to the gym

05:18.965 --> 05:20.965
and working out, you stay at the office

05:21.265 --> 05:22.485
and, and continue to work.

05:22.875 --> 05:24.525
Neither of those things will cause you

05:24.525 --> 05:26.525
to be injured in any particular day,

05:27.025 --> 05:29.765
but over time, those kinds of behaviors,

05:29.765 --> 05:31.965
because you're not taking care of yourself,

05:32.215 --> 05:34.285
could potentially lead to worse health.

05:34.745 --> 05:38.285
So in another paper in 2014 in Oxford, economic papers,

05:38.625 --> 05:39.725
myself and my colleague,

05:39.725 --> 05:42.445
Giannis CIO wrote a paper using British data,

05:43.245 --> 05:46.085
starting looking at, at in individuals who all were healthy,

05:46.665 --> 05:48.085
but then tracked over time

05:48.425 --> 05:50.765
how long they spent in performance related pay.

05:51.105 --> 05:53.965
And the longer people spent in performance related pay,

05:54.185 --> 05:56.725
the more likely they at some point would say that

05:56.725 --> 05:57.845
They have heart problems,

05:58.115 --> 05:59.725
that they would have stomach problems,

05:59.725 --> 06:00.725
and they would have mental health,

06:00.785 --> 06:02.805
or they may have mental health problems.

06:03.105 --> 06:06.245
So any of those things could potentially happen the longer

06:06.275 --> 06:08.605
that you stay in performance related pain.

06:08.745 --> 06:11.805
So that's the second mechanism that it, it generates a, a,

06:11.965 --> 06:13.885
a, a change in the behavior of your time.

06:15.025 --> 06:18.765
The third, potential pathway for performance related pay

06:18.765 --> 06:20.645
and health is through stress.

06:21.185 --> 06:22.805
One of the things that we know about performance,

06:22.905 --> 06:25.245
the many performance related pay jobs, is

06:25.245 --> 06:27.165
that the income stream is uncertain.

06:27.585 --> 06:29.205
If you think about the Uber driver

06:29.465 --> 06:31.685
or the, the, the, the delivery,

06:31.705 --> 06:35.485
the food delivery person, when they show up at work,

06:35.985 --> 06:39.605
in an evening, they don't know at how much they're going

06:39.605 --> 06:41.245
to earn by the end because they're paid

06:41.305 --> 06:43.645
for the trip that that they take.

06:43.705 --> 06:45.685
And so if it's a terrible weather night

06:45.905 --> 06:47.205
and nobody's ordering out,

06:47.205 --> 06:50.325
or nobody's, taking Uber, trips,

06:50.715 --> 06:51.805
then they get paid nothing.

06:51.865 --> 06:53.765
Or if they get paid very little, whereas it's,

06:53.765 --> 06:56.005
if it's a very busy night, they could come home

06:56.075 --> 06:57.085
with a lot of money.

06:57.465 --> 07:01.725
So that uncertainty around payment also generates stress,

07:01.745 --> 07:04.605
and particularly perhaps for those with dependents,

07:04.605 --> 07:06.125
if you have kids or if you had a mortgage

07:06.125 --> 07:07.485
or a car payment, any

07:07.485 --> 07:10.165
of those things could generate stress as well.

07:10.625 --> 07:12.925
And, there's well-known medical literature

07:13.035 --> 07:16.565
that ties stress to poor worse, poor health outcomes.

07:17.385 --> 07:19.925
The problem that we had was that in many of the surveys

07:19.925 --> 07:21.525
that, that we had used already

07:21.545 --> 07:24.285
or that were out there, there was not very good information

07:24.285 --> 07:25.845
about stress itself.

07:26.105 --> 07:28.485
Um, certainly nothing objective in terms of stress.

07:28.485 --> 07:30.885
They may ask, well, how stressed are you or, or not?

07:30.945 --> 07:33.085
But economists tend not to like those kinds

07:33.085 --> 07:34.725
of very subjective, issues.

07:35.115 --> 07:38.165
There's also a problem when you use real world data,

07:38.225 --> 07:39.645
and surveys,

07:39.755 --> 07:42.165
that people can select into performance

07:42.165 --> 07:43.485
related pay jobs or not.

07:43.485 --> 07:44.765
It's not a random occurrence.

07:45.425 --> 07:47.565
As I mentioned earlier, there are incentives for,

07:47.825 --> 07:52.125
for particular kinds of workers, perhaps to join, jobs

07:52.275 --> 07:53.845
that have performance related pay.

07:54.265 --> 07:57.725
Um, and so that adds a problem to, to the estimation.

07:57.745 --> 08:01.205
So what we decided to do was to run a series of experiments

08:01.495 --> 08:04.805
where we could, get past this issue of selection.

08:04.805 --> 08:09.765
Because in the lab, you can, exogen exogenously,

08:10.145 --> 08:13.685
allocate performance related pay or not within the lab.

08:14.145 --> 08:17.125
And then working with health psychologists, we have learned

08:17.125 --> 08:18.525
that, one way

08:18.525 --> 08:21.045
that they measure stress in a laboratory setting is

08:21.045 --> 08:22.565
by measuring cortisol,

08:22.565 --> 08:25.045
which is a hormone sometimes called the stress hormone.

08:25.585 --> 08:28.925
Um, but it's a hormone that the body naturally produces when

08:29.065 --> 08:30.485
it is stressed.

08:30.985 --> 08:33.565
Now, cortisol itself is actually a good thing.

08:33.745 --> 08:35.565
Um, it helps regulate the body.

08:35.785 --> 08:38.285
Um, when the body is stressed, there's a,

08:38.285 --> 08:40.805
there's a shutdown in in organs.

08:40.805 --> 08:42.245
It, it all goes back to the fight

08:42.245 --> 08:46.005
or flight, evolutionary thing that,

08:46.145 --> 08:47.205
that we all have.

08:47.505 --> 08:50.685
Um, and what cortisol does is it tries to relax the body,

08:50.905 --> 08:53.405
takes away the adrenaline and, and that kind of thing.

08:53.745 --> 08:56.245
Um, but the challenge is that when you are,

08:56.275 --> 08:59.525
when you have repeated stress, the cortisol isn't

08:59.525 --> 09:03.045
as effective to bring down your body into a kind of balance,

09:03.225 --> 09:05.245
homeostasis as it's called in the,

09:05.305 --> 09:06.485
in the medical literature.

09:06.865 --> 09:10.245
Um, and so, our challenge in these experiments was

09:10.245 --> 09:13.525
to look at, well, you know, if, if we pay some people

09:13.525 --> 09:16.125
through performance related pay and experiment and not,

09:16.145 --> 09:18.525
and then measure their cortisol, what happens?

09:18.745 --> 09:21.325
Do we see a spike in, in, in cortisol?

09:21.665 --> 09:23.165
So the first of our experiments, and,

09:23.165 --> 09:26.925
and this was a, a grant that was funded through the Economic

09:26.925 --> 09:29.325
and Social Research Council, here in the uk.

09:29.745 --> 09:31.485
Um, we set up an experiment

09:31.485 --> 09:33.645
where we would randomize performance related pay

09:33.645 --> 09:37.485
or not across, a number of participants within our,

09:37.545 --> 09:39.085
experimental laboratory.

09:39.705 --> 09:44.165
And, we found that indeed the cortisol,

09:44.195 --> 09:46.125
went up quite substantially for those

09:46.145 --> 09:48.525
who were paid under performance related pay.

09:48.905 --> 09:53.085
Um, and it went up about seven to 10%, depending on the,

09:53.085 --> 09:54.605
the measure that that we use.

09:54.825 --> 09:58.285
And just to put that into context, smoking,

09:58.385 --> 10:02.205
caused a raise an increase in cortisol by about 15%.

10:02.505 --> 10:04.325
So we were somewhere between half

10:04.325 --> 10:08.565
and three quarters of the effect of smoking on, on cortisol.

10:09.095 --> 10:11.445
Again, you could think about that as well.

10:11.555 --> 10:14.085
Okay, so people are a little bit more stressed,

10:14.505 --> 10:16.245
but it, it will relax you.

10:16.785 --> 10:19.125
But one of the things we know about performance related pay

10:19.125 --> 10:21.605
jobs is it's a repeated stressor, right?

10:21.655 --> 10:24.485
Every single day the Uber drivers shows up to work,

10:24.835 --> 10:26.645
they are under the stress of,

10:27.065 --> 10:28.805
of having performance related pay,

10:28.945 --> 10:32.125
or the, the car sales person every month has to hit

10:32.125 --> 10:34.365
that sales target in order to get their bonus

10:34.425 --> 10:36.885
or to get their commission every single month.

10:36.945 --> 10:39.645
And so the cortisol, as I mentioned earlier,

10:39.915 --> 10:43.045
will not relax you when you get this repeated,

10:43.075 --> 10:44.165
this repeated stress.

10:44.825 --> 10:47.845
One challenge with using experimental methods is of course,

10:48.185 --> 10:50.085
you've got a very selected group,

10:50.355 --> 10:53.285
primarily university students, that are coming in

10:53.285 --> 10:55.045
and playing for, for small stakes.

10:55.265 --> 10:57.765
So we really wanted to put this back out into the field

10:57.865 --> 11:02.325
and try to see if in real life, workers who are,

11:02.325 --> 11:06.365
working for, for a living, have similar responses.

11:06.425 --> 11:08.965
Now, we weren't able to find a survey data that,

11:08.965 --> 11:10.445
that actually had cortisol,

11:10.865 --> 11:14.165
but the, UK health longitudinal,

11:14.165 --> 11:17.765
household longitudinal survey, the UK HLS,

11:18.665 --> 11:22.525
has a, a portion of it where they, ask some

11:22.525 --> 11:25.765
of the respondents, a number of health questions,

11:26.185 --> 11:27.605
done through nurses.

11:27.785 --> 11:30.005
So they are sort of objective health measures.

11:30.115 --> 11:33.925
They also take blood samples of, some of the individuals

11:33.925 --> 11:36.565
and they are able to run tests on various,

11:36.575 --> 11:37.725
kinds of biomarkers.

11:37.825 --> 11:39.165
But importantly for us, we wanted

11:39.165 --> 11:43.125
to look at those biomarkers that were related to stress

11:43.185 --> 11:46.675
and inflammation since we didn't have the cortisol itself.

11:47.015 --> 11:49.395
So what we did was we looked within this dataset

11:49.695 --> 11:52.795
and tried to identify if there was a strong correlation,

11:52.815 --> 11:55.245
Be again between those who were se who,

11:55.305 --> 11:57.645
who said they were paid by performance related pay,

11:58.025 --> 12:01.445
and these measures of either self-reported health,

12:01.465 --> 12:03.965
or more objective measures like blood pressure

12:04.225 --> 12:05.845
or these inflammation markers.

12:06.065 --> 12:09.005
And we published this, this study,

12:09.005 --> 12:13.005
earlier in 2024 in the journal industrial relations.

12:13.185 --> 12:16.205
And we found very strong linkages between,

12:16.205 --> 12:17.485
performance related pay,

12:17.545 --> 12:19.685
and particularly with subjective mental health.

12:19.985 --> 12:22.205
So again, reiterating the stress element,

12:22.385 --> 12:24.485
the subjective stress element,

12:24.505 --> 12:27.205
but also with the systolic, blood pressure.

12:27.225 --> 12:29.565
So that increased quite substantially.

12:29.665 --> 12:32.885
And we know the literature out there on blood pressure and,

12:32.885 --> 12:36.045
and how that influences, health in a negative way.

12:36.465 --> 12:40.005
And then also on inflammation markers, FibroGen

12:40.025 --> 12:43.445
and c-reactive protein, particularly in men, again,

12:43.445 --> 12:47.205
performance related pay, in a sample of a broad sample

12:47.305 --> 12:50.405
of workers within the UK context tended

12:50.405 --> 12:53.085
to have these really strong and high biomarkers.

12:53.225 --> 12:55.085
For, for stress.

12:55.945 --> 12:58.005
One easy thing would be to say, well, listen,

12:58.145 --> 12:59.725
we should just throw out PRP, right?

12:59.825 --> 13:01.605
We should just get rid of performance related pay.

13:01.945 --> 13:04.005
Um, and,

13:04.025 --> 13:05.805
and that would get rid of these problems of,

13:05.805 --> 13:08.045
of poor health and, and stress.

13:08.185 --> 13:10.725
Now, some may advocate for that,

13:10.865 --> 13:12.405
but we know that there are benefits

13:12.465 --> 13:14.645
for performance related pay, as I outlined earlier.

13:15.265 --> 13:16.645
Um, employers like it

13:16.645 --> 13:18.565
because of this linkage, workers like it

13:18.565 --> 13:19.725
because they get higher pay.

13:19.985 --> 13:22.445
So I guess the, the key thing is, the first sort

13:22.445 --> 13:25.325
of learning point is, well, one, just recognizing

13:25.325 --> 13:28.925
that there is this issue, we have framed it in the past

13:29.025 --> 13:31.645
as an unintended consequence of performance related pay,

13:31.825 --> 13:33.885
but it's really an unknown consequence

13:33.885 --> 13:35.165
of performance related pay.

13:35.425 --> 13:37.125
And so the more you know about that

13:37.185 --> 13:39.485
and the potential cost for performance related pay,

13:39.665 --> 13:42.445
the more likely, perhaps you can do things,

13:42.445 --> 13:44.005
to mitigate for, for it.

13:44.265 --> 13:45.925
So you could offset it.

13:46.105 --> 13:48.405
Um, if you're going to offer performance related pay,

13:48.405 --> 13:52.605
perhaps a, an employer could mo, could offset it by,

13:52.625 --> 13:55.365
for example, if you're worried about accidents at work,

13:55.365 --> 13:58.325
making sure there's a strong focus on health

13:58.325 --> 14:00.925
and safety, at, at the workplace, making sure

14:00.925 --> 14:03.925
that people adhere to the safety protocols

14:03.925 --> 14:05.045
that, that are in place.

14:05.585 --> 14:08.925
Um, if you're worried about perhaps more, non-manual

14:08.925 --> 14:13.085
or white collar, workers, make sure that there are ways

14:13.385 --> 14:16.445
of being able to de-stress your, your workforce.

14:16.465 --> 14:19.605
So maybe that's even, subsidizing gym memberships

14:19.745 --> 14:22.245
or making sure that people do go out

14:22.245 --> 14:24.565
and take a walk at, at, at lunch or something like that.

14:24.875 --> 14:28.725
Ways of being able to, to build into the system,

14:28.995 --> 14:31.965
some ways to reduce the stress that seems to come about

14:31.965 --> 14:33.605
because of performance related pay.

Understanding Performance Related Pay Dynamics

00:00:00 - Performance related pay, or PRP is one of many forms of, of payment contracts that employers can have with employees. Most of us are used to perhaps an hourly rate of pay or a, a payment every month as a, as a salary. But for many types of jobs, you're actually paid on the basis of your performance, how much you actually produce. these can be very directly tied to, to your production. sometimes in economics, we call these piece rates where you're paid a particular sum for every unit
00:00:32 - Sometimes you're paid a commission on, say the amount of sales that you have. sometimes you're paid a bonus when you hit some kind of performance target. and, and so we have this sort of mix of different ways that, that workers are, are, are paid, and economics has, has long looked at all of these different types of contracts. looking at both the pros and the cons of each one of these kinds of contracts, but performance related pay or PRP, is one form that that often,
00:01:06 - one of the key advantages of performance related pay is as economics research is, has shown, is that because it pays for performance, it gives incentives for workers to perform better. and because of that direct link between their pay and their performance, we find that, workers that are paid under PRP, generate more product, which of course, the, the employer can turn around and sell. so, so firms can be somewhat more profitable. They have this direct tie to performance.
00:01:40 - who are actually really good at, at producing, because they know that they'll get a return, on this. So from an employer's point of view, economics tends to say yes, there are lots of good reasons. Interestingly, the earlier research on performance related pay also showed that workers benefited those workers that were paid under performance related pay. They got more money, so, so they benefited. and, and, and so there was this, you know, sort of virtuous feedback loop, that that went back,
00:02:12 - as opposed to all of that research that showed all of these positive benefits. there are these potential negative benefits, these negative, attributes to performance related pay. And interestingly, these were observed 200 more than 200 years ago. as any good economists should do, they should look at Adam Smith and the wealth of nations and in, the wealth of nations. Adam Smith has, a quote where he says, men when liberally paid by the peace,
00:02:42 - are very apt to overwork themselves and to ruin their health and constitution in a few years. So even back in 1776, Adam Smith, the father of economics, was making this linkage potentially between performance related pay. In the case, in the case of his example, it's peace Rates where you're paid directly for the output that you have and their health. So, when I read this quote, I thought, well, there must be lots and lots written about it. And indeed, there was actually very little written about it.

Performance Pay: Health Impacts and Risks

00:03:13 - of various ways of thinking about what those pathways may be between performance related pay and health. So one of the pathways, is through an increase in accidents. because what performance related pay does is it puts a huge emphasis on the output. So to sort of flip the, the management maximum, upside down workers are incentivized to work harder, but not necessarily smarter. They may cut corners. they'll do things that
00:03:42 - at, at the place of work. And so, back in, in 2012, I and two co-authors wrote a paper in the journal population economics that looked at about 30,000 workers, in, in the eu, and found there was a very strong correlation between those who were paid by performance related pay and the rate of injuries, amongst those, amongst those types of workers. So, strong relationship that we found, through accidents, and was particularly important
00:04:14 - Now, for many jobs, you know, rates of accidents are, are quite low. So for non-manual or white collar workers, it's probably not gonna be an issue. But there's a different mechanism, that can affect those kinds of workers because one of the things that we know about performance related pay as well is that it puts, explicitly a cost on not working. So in performance related pay, when you are paid by the basis of your production or your sales or something like that, when you're not working,
00:04:45 - So there's an incentive for you to be working perhaps quite a bit so that you can produce a lot, and that increases your pay. But if you're working a lot, it means that you're not doing other things. So, instead of going and, going to the grocery store and buying, food and taking it home and making a nice, healthy meal, you order take away and, and you eat a, you know, greasy pizza on, at, at your desk while you're continuing to do work, or, instead of going, and taking a walk
00:05:18 - and working out, you stay at the office and, and continue to work. Neither of those things will cause you to be injured in any particular day, but over time, those kinds of behaviors, because you're not taking care of yourself, could potentially lead to worse health. So in another paper in 2014 in Oxford, economic papers, myself and my colleague, Giannis CIO wrote a paper using British data, starting looking at, at in individuals who all were healthy, but then tracked over time
00:05:51 - And the longer people spent in performance related pay, the more likely they at some point would say that They have heart problems, that they would have stomach problems, and they would have mental health, or they may have mental health problems. So any of those things could potentially happen the longer that you stay in performance related pain. So that's the second mechanism that it, it generates a, a, a, a change in the behavior of your time. The third, potential pathway for performance related pay

Uncertainty and Stress in Performance Pay Jobs

00:06:21 - One of the things that we know about performance, the many performance related pay jobs, is that the income stream is uncertain. If you think about the Uber driver or the, the, the, the delivery, the food delivery person, when they show up at work, in an evening, they don't know at how much they're going to earn by the end because they're paid for the trip that that they take. And so if it's a terrible weather night and nobody's ordering out, or nobody's, taking Uber, trips,
00:06:51 - Or if they get paid very little, whereas it's, if it's a very busy night, they could come home with a lot of money. So that uncertainty around payment also generates stress, and particularly perhaps for those with dependents, if you have kids or if you had a mortgage or a car payment, any of those things could generate stress as well. And, there's well-known medical literature that ties stress to poor worse, poor health outcomes. The problem that we had was that in many of the surveys that, that we had used already
00:07:24 - about stress itself. certainly nothing objective in terms of stress. They may ask, well, how stressed are you or, or not? But economists tend not to like those kinds of very subjective, issues. There's also a problem when you use real world data, and surveys, that people can select into performance related pay jobs or not. It's not a random occurrence. As I mentioned earlier, there are incentives for, for particular kinds of workers, perhaps to join, jobs
00:07:54 - and so that adds a problem to, to the estimation. So what we decided to do was to run a series of experiments where we could, get past this issue of selection. Because in the lab, you can, exogen exogenously, allocate performance related pay or not within the lab. And then working with health psychologists, we have learned that, one way that they measure stress in a laboratory setting is by measuring cortisol,
00:08:25 - but it's a hormone that the body naturally produces when it is stressed. Now, cortisol itself is actually a good thing. it helps regulate the body. when the body is stressed, there's a, there's a shutdown in in organs. It, it all goes back to the fight or flight, evolutionary thing that, that we all have. and what cortisol does is it tries to relax the body, takes away the adrenaline and, and that kind of thing.
00:08:56 - when you have repeated stress, the cortisol isn't as effective to bring down your body into a kind of balance, homeostasis as it's called in the, in the medical literature. and so, our challenge in these experiments was to look at, well, you know, if, if we pay some people through performance related pay and experiment and not, and then measure their cortisol, what happens? Do we see a spike in, in, in cortisol? So the first of our experiments, and, and this was a, a grant that was funded through the Economic

Effects of Performance Pay on Stress Levels

00:09:29 - we set up an experiment where we would randomize performance related pay or not across, a number of participants within our, experimental laboratory. And, we found that indeed the cortisol, went up quite substantially for those who were paid under performance related pay. and it went up about seven to 10%, depending on the, the measure that that we use. And just to put that into context, smoking,
00:10:02 - So we were somewhere between half and three quarters of the effect of smoking on, on cortisol. Again, you could think about that as well. Okay, so people are a little bit more stressed, but it, it will relax you. But one of the things we know about performance related pay jobs is it's a repeated stressor, right? Every single day the Uber drivers shows up to work, they are under the stress of, of having performance related pay,
00:10:32 - that sales target in order to get their bonus or to get their commission every single month. And so the cortisol, as I mentioned earlier, will not relax you when you get this repeated, this repeated stress. One challenge with using experimental methods is of course, you've got a very selected group, primarily university students, that are coming in and playing for, for small stakes. So we really wanted to put this back out into the field and try to see if in real life, workers who are,
00:11:06 - Now, we weren't able to find a survey data that, that actually had cortisol, but the, UK health longitudinal, household longitudinal survey, the UK HLS, has a, a portion of it where they, ask some of the respondents, a number of health questions, done through nurses. So they are sort of objective health measures. They also take blood samples of, some of the individuals
00:11:36 - kinds of biomarkers. But importantly for us, we wanted to look at those biomarkers that were related to stress and inflammation since we didn't have the cortisol itself. So what we did was we looked within this dataset and tried to identify if there was a strong correlation, Be again between those who were se who, who said they were paid by performance related pay, and these measures of either self-reported health, or more objective measures like blood pressure
00:12:06 - And we published this, this study, earlier in 2024 in the journal industrial relations. And we found very strong linkages between, performance related pay, and particularly with subjective mental health. So again, reiterating the stress element, the subjective stress element, but also with the systolic, blood pressure. So that increased quite substantially. And we know the literature out there on blood pressure and, and how that influences, health in a negative way.

Health Risks and Benefits of Performance Pay

00:12:40 - and c-reactive protein, particularly in men, again, performance related pay, in a sample of a broad sample of workers within the UK context tended to have these really strong and high biomarkers. For, for stress. One easy thing would be to say, well, listen, we should just throw out PRP, right? We should just get rid of performance related pay. and, and that would get rid of these problems of, of poor health and, and stress.
00:13:10 - but we know that there are benefits for performance related pay, as I outlined earlier. employers like it because of this linkage, workers like it because they get higher pay. So I guess the, the key thing is, the first sort of learning point is, well, one, just recognizing that there is this issue, we have framed it in the past as an unintended consequence of performance related pay, but it's really an unknown consequence of performance related pay. And so the more you know about that and the potential cost for performance related pay,
00:13:42 - to mitigate for, for it. So you could offset it. if you're going to offer performance related pay, perhaps a, an employer could mo, could offset it by, for example, if you're worried about accidents at work, making sure there's a strong focus on health and safety, at, at the workplace, making sure that people adhere to the safety protocols that, that are in place. if you're worried about perhaps more, non-manual or white collar, workers, make sure that there are ways
00:14:31 - So maybe that's even, subsidizing gym memberships or making sure that people do go out and take a walk at, at, at lunch or something like that. Ways of being able to, to build into the system, some ways to reduce the stress that seems to come about because of performance related pay.

Faculti AI can make mistakes

Research

Policy

Analysis