Income of Gig Work vs. Previous Job in Pakistan

Richard Heeks, Iftikhar Ahmad, Shanza Sohail, Sidra Nizamuddin, Athar Jameel, Seemab Haider Aziz, Zoya Waheed, Sehrish Irfan, Ayesha Kiran & Shabana Malik

Does the transition to gig work improve incomes in Pakistan?

Many workers join gig work platforms in the belief that their incomes will improve, but is this borne out in practice?  To investigate, the Centre for Labour Research interviewed 94 workers based on six platforms across three sectors: ride-hailing, food delivery, and personal care.

Of these, 51 were able to tell us what their previous monthly income had been in their most-recent employment prior to joining the platform[1].  Stated income varied from the equivalent of US$60 per month up to U$1,200 per month, and averaged US$220 per month[2].

After moving into gig work, average gross income was slightly higher, at US$240 per month but, as the graph below shows, there was a much more differentiated picture behind the average, with around 40% of respondents earning less gross income (red-bordered blue columns) than they had done previously.

However, as the graph also shows, things looked worse when comparing net income (orange columns).  For the great majority of prior jobs, work-related costs were small (only work-to-home transport, which we calculated based on typical commuting journeys in Pakistan to be just under US$18 per month; i.e. less than 10% of average gross income).  But for gig work – much of which relies on journeys by vehicle and continuous internet connectivity – the costs of petrol, maintenance and data eat heavily into gross income.  In addition, for some (only a few in our Pakistan sample) there are costs of renting their vehicle.[3]

These costs represented, on average, 65% of gross income and knocked average net income for gig workers down to just US$85 per month.  When we compare before-and-after for net income, then, we found more than 70% of our sample were earning less than in their previous job, and 45% earned over US$100 per month less.

This was especially an issue for ride-hailing drivers and it does reflect the particular circumstances during our interview period of late 2021 to early 2022: a drop-off in demand for travel due to Covid, and a steep rise in petrol prices.  Indeed, so bad was the problem that just over a fifth – 21 of the 94 – were reporting negative income.  That is, they were effectively paying to go to work as their costs exceeded their gross income; something to which the platforms responded in May 2022 by dropping the commission taken from drivers to 0%.

While recognising the challenging period for gig workers covered by our fieldwork, nonetheless, this does suggest that – by and large – gig work is not delivering the income boost that workers often hope for.  They may, for example, be lured by gross income figures, not realising how much lower net income will be.  Gig work does provide a livelihood – 40% of our sample were unemployed in the immediate period prior to joining – but it is not really fulfilling its promise.  It also falls far from decent work standards: five-sixths of those we interviewed took home less than a living wage.

If you’d like to know more, please refer to the 2022 Fairwork Report on Pakistan’s gig economy.


[1] Those who stated what their prior employment had been gave the following job descriptions: BPO operator, Teacher (2), Housekeeper, Shopkeeper, Gas company worker (2), Safety officer, Business person, Tanker driver, Ride-hailing driver with another platform (3), Traditional taxi driver (3), Farmer, Builder, Computer operator, Cook, Technician, Shop assistant, Domestic worker, Government worker

[2] This average is some way above the overall average earnings of US$140 per month but well below formal sector average monthly salary of US$480.

[3] For further detail, see this discussion of the breakdown of ride-hailing passenger payments.

3 thoughts on “Income of Gig Work vs. Previous Job in Pakistan

  1. Dear Richard and team,
    This is a great topic area and seems to be a very interesting data set you have collected. Its great that you are trying to document people’s dynamic journey through gig work and how their income evolves. I am working on a very similar study in Indonesia with Jonathan Donner and Chernay Johnson.
    A few things I wanted to ask:
    1 – You seem to interpret the fact that some gig workers make less than their previous job as a mistake or miscalculation on the part of the worker. But what about a well informed choice born of necessity? At one point after the 2002 recession I lost my job and took another at 1/3rd the pay. It wasn’t something I regretted or felt was a miscalculation, I was glad to have it!
    2 – Related to this is a potential sample bias inherent in only about half the sample being able to recall their past income. Those who can recall would seem more likely to me to have come from formal employment (thus higher wage) whereas others might have been in more unstable or temporary employment (thus potentially lower). I wonder if you have any insights on the differences in the sample between responders and non?
    3 – finally, given the sample period and the level of macro upheaval I think its pretty hard to draw larger generalizations from your data. Lots of things where going on at that time that might make it a really unique moment in history where results wont generalize at all. I know you mention it but it would deserve more discussion in my view.

    Thanks again for sharing this data and all your other great work!

    1. Hi Jake – many thanks for your insightful comments. Let me respond:

      1. For sure, there’s both push and pull in the move to gig work. On push, we note the 40% who had no job immediately prior to moving into gig work, but it’s also reasonable to investigate the pull basis on which many workers were attracted into gig work, with a promise – or perception – of better earnings.

      2. Good point. Those for whom we had no prior income either stated they were unemployed or gave no answer about their prior work, so you could be correct but we don’t have the data to say one way or the other.

      3. True. I plan to take a look at our data from other locations (South Africa and, by coincidence, also Indonesia) including earlier years to see what that says. I did a rough calculation a while back for our data from South Africa in 2019 and concluded, “We did not systematically gather “before-and-after” data on pay but, for those that brought it up, it appeared that delivery riders were earning more than previously, even taking costs into account. For domestic workers and taxi drivers this was less clear: we only had six responses but four of these were earning less than they had previously done.” I have a hunch that, over time, the ‘earning less’ group will have grown.

      Just one other point, which is that it would be very interesting to hear from those who have made the transition OUT of gig work into something else. But I’ve not seen research on that group.

      1. Thanks Richard, all good points. On your last point, we are collecting data on “leavers” from different kinds on online livelihoods work including riders so perhaps we should discuss at some point. Feel free to reach out via email if you want. jake at dfslab dot net.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.