The More Fungible Worker

http://unpleasantfacts.com/the-more-fungible-worker

by dlss • 12 years ago

92 173 12 years ago

When I discuss 'Policies and Procedures' (including but not limited to tech like the OP's Amazon example) I often refer to "Plug and Play" staff [1]. If, for any reason, an employee leaves a position then the business is able to replace them with minimum fuss. The more the role can depend on technology, the more 'true' this is, but it needn't be seen as tech-specific.

Staff of course have a different view as to their "fungibility". I tend to overcome this in two ways:

First, "If you can't be replaced then you can't be promoted." That resonates a lot.

Second, I replace the 'Bus Factor'[2] context with the 'Lottery Factor'. More directly, "How many of you, if you won $35M in Powerball, would still be working here next year? So let's plan for at least one of you to win the lottery."

(They're not bad at maths, they get that it's the same conversation as being hit by a bus, but the energetic flip leads to faster engagement in the process.)

[1] Maybe I need a new Cloud metaphor?

[2] http://en.m.wikipedia.org/wiki/Bus_factor

chii3• 12 years ago

First, "If you can't be replaced then you can't be promoted." That resonates a lot.

that is a very sneaky way of saying that their wage/salary cannot increase.

JacobAldridge2• 12 years ago

How so? Most of my clients give 5%+ payrises each year.

Animats4• 12 years ago

Yes. I refer to the Amazon warehouse operation as "Machines should think. People should work". All the thinking is done by computers and robots. Watch one of the Kiva robot videos that shows the human picker reaching where the laser pointer tells them to reach, waving the item under the scanner, and putting the object in the bin where the light is on.

Those people will be unemployed as soon as the Amazon Robot Picking Challenge (http://amazonpickingchallenge.org/) succeeds.

This is the future.

AndrewKemendo• 12 years ago

Wow, you weren't joking.

http://youtu.be/Fr6Rco5A9SM?t=2m33s

I knew about Kiva and the widescale automation in warehousing but I had no idea the UI was so...simple. I see no reason the person couldn't be taken out of the loop completely here with a robotic arm.

Animats4• 12 years ago

The Amazon Picking Challenge, to do that job with a robotic arm, is underway. Finals, with live robot demos, are on May 26-28, 2015, in Seattle. That's the next step, and it's not far away. In a year or two, robots will probably be handling everything rectangular, like boxes and books. There may still be a few people around for odd sizes and shapes. Not many.

As for "there will be jobs building and repairing the robots", Amazon has about 30,000 pickers during peak season. Kiva Systems has about 600 employees. Because all those little mobile robots are interchangeable, there's not much need for on-site maintenance. There's someone on site to replace wheels and batteries and do basic diagnostics, but seriously broken ones are just shipped back to the factory.

The Kiva system requires little installation, which is why it's being deployed so fast. Previous systems required conveyors, stacker cranes running on tracks, and lots of specialized, custom machinery. Kiva just needs a flat floor, markers on the floor, and WiFi.

Here's what US employment looks like, by sector: http://www.bls.gov/emp/ep_table_201.htm Manufacturing is only 8% of the labor force, yet the US makes more stuff than ever. US manufacturing is doing great. The drop in output after 2008 has been completely recovered. Manufacturing just doesn't need many people. The jobs lost never came back.

In 2012, retail employed 10% of the US labor force. Remember malls? (See deadmalls.com) No new indoor mall has been built in the US in 10 years. (OK, American Dream Meadowlands in New Jersey is still under construction, after 12 years, three bankruptcies, and a roof collapse, but that's a political boondoggle, part of a stadium complex.) Online ordering doesn't need many people. The Bureau of Labor Statistics estimates show a small decline over the next 10 years, but those numbers predate Amazon's recent growth and are probably optimistic. The BLS shows employment growth in transportation and warehousing. Probably not going to happen.

This is the future, and the future is now.

TeMPOraL7• 12 years ago

Holy fuck.

That's basically Manna, sans headphones.

http://marshallbrain.com/manna1.htm

toomuchtodo7• 12 years ago

How will anyone purchase anything if they don't have money from a job because robotics and software can do the job better? Basic income?

Or are we resigned to a dystopian future where the landscape is covered by Amazon fulfillment centers sitting silent, stocked to the ceiling with product that can't be sold because no one can afford said merchandise?

politician2• 12 years ago

No. That's a silly question - if the `Idiocracy` future you're describing comes about it'll only last for a moment until the distribution centers are pillaged by regional warlords.

However, long before that we will have adopted some form of Basic Income, or we'll have decided to go to space and restart the resource race on hardmode.

tarikjn• 12 years ago

Those who will have income will either design or maintain the robots, or produce value which robots can't produce.

When a job disappear, people either retire or change job, that's pretty much what happened when agriculture became mechanized.

tveita• 12 years ago

"Machines should think. People should work"

This dystopia has been envisioned for more than 80 years: https://www.youtube.com/watch?v=q030WNZvXrA

It's a good reminder that social and technological progress must often be fought for separately.

The market is brutal when applied to people. Worker fungibility is a value add for the company, and yet it drops wages to the minimum, since the worker's individual bargaining power is practically non-existent. The traditional answer is unions, but even that becomes impractical as the cost of replacing a worker approaches zero.

This is especially relevant when the buzzwords of the day are crowdsourcing and independent contracting. Are we really making things more efficient and creating value, or are we just helping drive down the wages of people that already have a hard time scraping by?

TeMPOraL7• 12 years ago

The market serves Moloch[0], not humans. We have bootstrapped a system that is close to become self-sustaining, and humans are increasingly becoming cost rather than value. Unless something is done it will either collapse or become a human-free (or almost human-free) system. Neither outcome is good for us.

[0] http://slatestarcodex.com/2014/07/30/meditations-on-moloch/

noonespecial2• 12 years ago

I'm afraid for a future that ends up like this:

Nobody hires people because robots can do it cheaper, so no one has a job, so no one can buy anything, so no robots do those things that they could do cheaper.

The technology doesn't even have to exist to cause poor-world. Only be possible.

Put me firmly in the basic minimum income camp.

chii3• 12 years ago

there will be people who have money (for example, those who inherited their wealth, or those who own capital/businesses). Their needs may vastly differ from the poor, and because they have money, the market serves them. The jobless poor will just either live on what little the state/charity can provide, or just die on the streets. I don't see basic income as feasible politically nor in practise. Therefore, what you must do now is to ensure that you are one of those who have money.

TeMPOraL7• 12 years ago

I don't think this will be likely outcome - no change happens instantaneously worldwide, so the very period of (possibility of) deployment will - I believe - make this system turn into robots working for rich while most of the humanity starves.

So, while I fear a different future, I'm also firmly in the basic income camp. Right to live should not in principle be dependent on slaving away your life, so let's deal with that when the technology allows us to.

a3_nm• 12 years ago

You still need humans to explain to the machines how to think. The problem is the training that it requires.

jeffreyrogers• 12 years ago

This kind of thing makes me sad. We know that one of the things people need to feel satisfied is a degree of autonomy. And by taking that away and treating them as a commoditized component in some machine you're taking away a lot more than you realize from them.

walterbell3• 12 years ago

Spreadsheets made it easier/possible for "users" (accountants) to write financial "programs", reducing the need for custom development of financial software. If there are more innovations like spreadsheets, there will be less or different needs for programmers as the labor category exists today. Fungibility cuts both ways.

bobbles• 12 years ago

The circle of life:

Finance task :

finance system to address task
excel spreadsheet to address finance system shortcomings
finance system to 'formalise' the spreadsheet
excel spreadsheet to address finance system shortcomings
finance system to 'formalise' the spreadsheet

etc...

walterbell3• 12 years ago

This circle could also apply to enterprises and startups.

VLM4• 12 years ago

My experience is the need for programmers rapidly increases under those conditions because

1) New capabilities and demonstrations of capabilities mean its easier to make the new thing a business requirement, now implemented and supported by IT.

2) Labor required to do something vs skill level in programming scales way beyond linear, maybe exponential. So pay a genius 10 hours of labor at 3x pay, your average grunt "real programmer" 50 hours at 2x pay, or front line sorcerers apprentice doesn't even have programmer in his job title 250 hours at 1x pay to do it in a spreadsheet.

3) Sorcerers apprentices have no idea how to use tools making them incredibly inefficient both in calculating results and the immense labor they expend for their result. The most expensive way to get a result is to pay people who have no idea what they're doing.

walterbell3• 12 years ago

How do you see this dynamic playing out with SaaS services vs. IT? In theory, an external SaaS provider could create economies of scale in usable software for apprentices. Would custom enteprise apps reincorporate those learnings? Or would the enterprise pay a cost to integrate the output of SaaS services (possibly with apprentice errors) into corporate systems?

crdoconnor25• 12 years ago

The Luddites tried futilely to stop progress that benefited society

Luddites were not anti-technology nor were they anti progress. They destroyed technology as a tactic to raise the price of the more automated competition. Their competition made cheaper, crappier equivalents of their artisanal work made with de-skilled McJob equivalent workers whose wages were pressured to abhorrent levels (via other means like the enclosure movement/Napoleonic wars).

Similarly, people who strike are not "futilely trying to stop the existence of work". It's another tactic.

In a way, the Luddites are pretty similar to the anti-H1B crowd of hacker news.

Before these technological advancements, workers enjoyed mini-monopolies.

This is a hacky way of looking at it but the conclusion sort of works. It makes much more sense to look at relative market power of employees vs. employers rather than use the all-or-nothing terminology of monopoly vs. perfect competition.

If Amazon workers attempt to sabotage Amazon's stock picking machinery, I expect it will result in an increase in wages, much like it did for the Luddites. They will be called Luddites, but similar to the original Luddites, they kind of only really have a problem with technology that is used to control them.

dominotw• 12 years ago

In a way, the Luddites are pretty similar to the anti-H1B crowd of hacker news.

H1B crowd == de-skilled McJob equivalent workers ?

crdoconnor25• 12 years ago

de-skilled McJob equivalent workers = Outsourced (often Indian) workers that are paid peanuts and who consequently don't give much of a shit about the quality of their work.

Elance with its screen recording feature = the equivalent of stocking frames

mc32• 12 years ago

Then that would make the people who refuse to train their replacement a 'Luddite' of sorts? Afteral that would entail the replacement worker ramping up on their own.

cbd19842• 12 years ago

Their competition made cheaper, crappier equivalents of their artisanal work made with de-skilled McJob equivalent workers whose wages were pressured to abhorrent levels

... which enabled working-class people to own fairly large amounts of clothing for the first time, and even have fashionable clothing, which had once been the exclusive preserve of the rich.

Don't try to paint this as all rich-vs-poor. The poor got a lot out of mechanization, even at the time, and reforming that system worked out a lot better than trying to work with people who were so threatened by change they would destroy other peoples' method of earning a living.

IndianAstronaut2• 12 years ago

Very true. Consumers are often ignored in all of this but it is large scale automation and industrialization that is allowing for the mass consumer culture. A visit to other places which aren't as industrialized or have a lot of protectionism and you fill find very few consumer goods available and often for very high prices.

crdoconnor25• 12 years ago

which enabled working-class people to own fairly large amounts of clothing for the first time

Actually...

"English peasants didn’t want to give up their rural communal lifestyle, leave their land and go work for below-subsistence wages in shitty, dangerous factories being set up by a new, rich class of landowning capitalists. And for good reason, too. Using Adam Smith’s own estimates of factory wages being paid at the time in Scotland, a factory-peasant would have to toil for more than three days to buy a pair of commercially produced shoes. Or they could make their own traditional brogues using their own leather in a matter of hours, and spend the rest of the time getting wasted on ale. It’s really not much of a choice, is it?

Faced with a peasantry that didn’t feel like playing the role of slave, philosophers, economists, politicians, moralists and leading business figures began advocating for government action. Over time, they enacted a series of laws and measures designed to push peasants out of the old and into the new by destroying their traditional means of self-support."

http://www.nakedcapitalism.com/2013/08/the-rise-of-bullshit-...

In today's world, austerity serves the same function as the enclosure laws. Reduce public sector demand for labor so as to push down wages across the board.

Perdition• 12 years ago

which enabled working-class people to own fairly large amounts of clothing for the first time, and even have fashionable clothing, which had once been the exclusive preserve of the rich.

No, that was due to social reform as a result of the threat of anarchist/socialist/communist uprisings against the established order.

Average living standards went down during most of the industrial revolution.

reforming that system worked out a lot better than trying to work with people who were so threatened by change they would destroy other peoples' method of earning a living.

Except the capitalists attempted (mostly successfully) to destroy alternate ways of earning a living other than industrial work.

America was flooded with immigrants because it was a chance for peasants to have land of their own.

x0x0• 12 years ago

Anyone modestly familiar with labor and economic history would know that the actual poor / working class got screwed by the industrial revolution. It took a very long time (on the order of perhaps a century) for them to see improvements in standard of living; instead, they lived in squalor and suffered a significant increase in disease. Which doesn't even get into the use of force by government to drive them into cities. In sum, the claim that the "poor got a lot out of mechanization, even at the time" couldn't be more wrong.

yummyfajitas16• 12 years ago

In a way, the Luddites are pretty similar to the anti-H1B crowd of hacker news.

I'd describe it a bit differently. The anti-H1B crowd here is typically (not exclusively [1]) like the old Jim Crow/New Deal/Chinese Exclusion crowd - wanting to pass laws to reduce competition by people who have had a different accident of birth.

I'm curious to see if anyone can cook up an argument why protectionism against orange, white and green is morally justified, but protectionism against black is not.

[1] Some people, (including myself), have non-protectionist opposition (or at least reservations about) to immigration. I'm very unsettled that my reservations could also justify anti-black protectionism if the same empirical criteria were satisfied.

dropit_sphere5• 12 years ago

You're conflating race with citizenship.

One of the ostensible goals of a state is to protect the interests of its citizens. Less politely: a state is an in-group that knows it's an in-group, and excludes the out-group, in accordance with all in-groups ever, everywhere.

I feel perfectly justified valuing the interests of, say, my family, over someone else's family. I extend the same to my countrymen, over some other country's citizens.

crdoconnor25• 12 years ago

I'd describe it a bit differently.

Of course you would.

I'm curious to see if anyone can cook up an argument why protectionism against orange, white and green is morally justified, but protectionism against black is not.

Four aspects of the debate that the, uh... 'curious' tend to misrepresent or outright ignore:

* Race-baiting doesn't change the fact that at its heart it is about wages vs. profits.

* H1Bs are not just about bringing in workers. They are also about giving employers the power to fuck those they bring over by linking their immigration status to their job. This isn't just bad for them.

* The effects of brain drain on 3rd world countries where the brightest people flee to richer countries.

* Brain drain is not just about losing the brightest people but also about losing the (often large) investment in their education paid for by the 3rd world, that the 1st world gets the whole benefit of.

venomsnake2• 12 years ago

Interesting how much of the anti H1B will support green card for competent ITs. I suppose - quite a lot.

stretchwithme3• 12 years ago

People who kept doing things the old inefficient way will see their wages go down.

If a worker can make ten times the widgets with newer technology, the amount that can be charged per widget, and thus the amount a worker can be paid to produce each one, has to drop eventually.

stretchwithme3• 12 years ago

Of course, the upside is that the worker thatchers up with technology eventually gets their widgets for 1/10 of the cost.

And this is literally true. Cars, for example, used to cost ten times as much. You can buy a car as reliable as any 1914 model for a few hundred bucks that is more efficient and cheaper to maintain. You can also start it without getting out of the car.

TeMPOraL7• 12 years ago

And as the inefficient workers get outcompeted and more skilled competition shows up, the end result is always the same: new widget costs 1/10 of the original costs, those who make them earn the same amount of money as they did before, but now 9/10 of workers are without a job. It has always been like that and it's responsible for the age of plenty we live in - but the problem of today is that those 9/10 of workers without a job are increasingly unable to find another one, as everything gets slowly automated away.

stretchwithme3• 12 years ago

9/10 of workers are not out of a job. Unemployment is not 90%.

politician2• 12 years ago

A timely article considering the deformation of traditional IT departments threatened by Docker.

k__• 12 years ago

I remember talking to a few people in high school about me studying CS because I wanted to become a developer.

They all my IT-friends were like, "Don't do it man. No one needs new software! Better be an administrator, in the future they will be needed to manage all the IT stuff that already exists, but there will hardly be anything new on the marked!"

That was 2003, right before the social and mobile boom.

toomuchtodo7• 12 years ago

Docker isn't anything new (perhaps nicer toolset on an old idea), and admins aren't getting replaced by devs any faster than devs getting replaced by admins. If anything, the amount of work continues to get abstracted away for both camps.

Maybe you won't need sysadmins in ~10 years. Maybe. Those of us still doing IT will just pivot into another niche.

eropple2• 12 years ago

IT departments are threatened, but it's really not Docker doing it. Docker is somewhere down the list behind AWS/GCE/Rackspace, Chef/Puppet, Mesos/Deis/Flynn, and really good hosted Active Directory services.

toomuchtodo7• 12 years ago

AWS/GCE/Rackspace, Chef/Puppet, Mesos/Deis/Flynn, and really good hosted Active Directory services.

Which all still needs ops. IT departments aren't threatened; their capital expenditure budget is. Someone still needs to make sure the gears at the end of the day.

hacknat• 12 years ago

I don't understand the argument that just because technological progress in the past has led to higher wages and employment (indeed, it actually hasn't always) that it's a given that this will always happen.

It seems to me that it is self-evident that most low-skill labor, at the very least, is in the process of being automated and is therefore going to be unnecessary. There has been no other time in human history when ALL low-skill labor was in the process of being automated in such a generalized way.

While it is certainly true that there is and always has been huge economic pressure on low-skill labor to be destroyed, it is being destroyed at an unprecedented rate and the economic forces that are destroying it don't seem to be creating any residual low-skill labor (like the assembly line, etc did). Making sure that the labor force is educated enough to not need the low-skill labor is good, and it is undeniable that everybody is getting more educated. Still, I worry a great many people are going to get left out in the dark.

SomeCallMeTim2• 12 years ago

This is why we should support basic-income or minimum-income.

Then NO ONE is left out to starve, ever. People would still get jobs to get MORE money, because people always want more, but you could eliminate minimum/living wage laws, so the marginal cost of production at jobs that people enjoyed could actually go down, while at the same time more people would have money to spend.

The transition is the hard part.

_yosefk• 12 years ago

People always want more money, but they won't necessarily want to work for it. Is there any evidence for a sufficient willingness to work under a basic income arrangement to keep the economy going?

ahupp• 12 years ago

"It seems to me that it is self-evident that most low-skill labor, at the very least, is in the process of being automated and is therefore going to be unnecessary. "

I'm not sure this is restricted to low-wage labor. There are plenty of low-wage jobs that are just complex enough to make automation more expensive than wages for the near future. Driving is a good example here (for now).

And there's plenty of high-wage jobs which are quite automatable - accounting (TurboTax), financial management (wealthfront/betterment), lots of software operations work (AWS tools), etc.

copsarebastards2• 12 years ago

Making sure that the labor force is educated enough to not need the low-skill labor is good, and it is undeniable that everybody is getting more educated.

That's very deniable in the US. The rising cost of education and lack of government funding for education means that many, many people are not getting more educated.

hsod• 12 years ago

This graph:

http://www.census.gov/hhes/socdemo/education/data/cps/histor...

and others on this page:

http://www.census.gov/hhes/socdemo/education/data/cps/histor...

seem to disagree with your intuition

Futurebot2• 12 years ago

To be more precise, it's not actually 'low-skill' labor that is vulnerable - it's 'routine' labor that is. Autor and Acemoglu talk about this in their paper 'Skills, Tasks and Technologies':

'Following, ALM, we refer to these procedural, rule-based activities to which computers are currently well-suited as "routine" tasks. By routine, we do not mean mundane (e.g., washing dishes) but rather sufficiently well understood that the task can be fully specified as a series of instructions to be executed by a machine (e.g., adding a column of numbers). Routine tasks are characteristic of many middle-skilled cognitive and manual jobs, such as bookkeeping, clerical work, repetitive production, and monitoring jobs. Because the core job tasks of these occupations follow precise, well-understood procedures, they can be (and increasingly are) codified in computer software and performed by machines.'

JetSpiegel4• 12 years ago

This is a very important distinction. A robot to wash dishes would be ridiculously expensive for such small payoff.

snowwrestler3• 12 years ago

There has been no other time in human history when ALL low-skill labor was in the process of being automated in such a generalized way.

Resist hindsight bias here; the work we consider low-skill today--working in factories--was the high-skilled labor of its day. The low-skill labor it replaced was largely the large-scale manual labors of raising food, producing energy, and making basic everyday items.

The need for more educated, better trained, smarter workers during the Industrial Revolution led directly to the establishment of free, publicly funded schools for all through high school. That was a major cultural shift, a huge public investment, and a huge jump in the mental capabilities of the workforce as a whole.

I think it's arguable that now we're facing another similar shift, and we need to make another jump in how we educate people. The number of people who get left behind depends on how well we recognize and commit to this shift.

I would argue that a starting point is to accept that the economy is changing, and we have to do something to prepare everyone for that. Wishing that textile or car factories would go back to centers of mass employment won't really protect anyone.

SatvikBeri• 12 years ago

Workers becoming more fungible happens even in creative domains. For example, I've heard several companies cite how advances in sales software have dramatically reduced the training time for new sales people as well as reducing the gap between top performers and everyone else.

Futurebot2• 12 years ago

Those that can 'defer to the machine' are going to do better in many jobs (that will remain) in the future. This is a big part of Cowen's "Average is Over." Here's a good review of it which goes into that aspect of it:

http://ieet.org/index.php/IEET/more/searle20150109

Excerpt:

'The moral Cowen draws from freestyle chess is that the winners of these games, and he extrapolates, the economic “games” of the future, are those human beings who are most willing to defer to the decisions of the machine. I find this conclusion more than a little chilling given we’re talk about real people here rather than Knight or Pawns, but Cowen seems to think it’s just common sense.

In its simplest form Cowen’s argument boils down to the prediction that an increasing amount of human work in the future will come in the form of these AI-human teams. Some of this, he admits, will amount to no workers at all with the human part of the “team” reduced to an unpaid customer. I now almost always scan and bag my own goods at the grocery store, just as I can’t remember the last time I actually spoke to a bank teller who wasn’t my mom. Cowen also admits that the rise of AI might mean the world actually gets “dumber” our interactions with our environment simplified to foster smooth integration with machines and compressed to meet their limits.

In his vision intelligent machines will revolutionize everything from medicine to education to business management and negotiation to love. The human beings who will best thrive in this new environment will be those whose work best complements that of intelligent machines, and this will be the case all the way from the factory floor to the classroom. Intelligent machines should improve human judgement in areas such as medical diagnostics and would even replace judges in the courtroom if we are ever willing to take the constitutional plunge. Teachers will go from educators to “coaches” as intelligent machines allow individualized instruction , but education will still require a human touch when it comes to motivating students.'

KedarMhaswade• 12 years ago

The essence of the article is about effectiveness of technology/trade in reducing the 'gaps' between the capabilities of workers/artisans results in lowering their average compensation. But I think there's something missing. Professor Katz famously suggested not to become scientists [1] because getting your research funded becomes top priority than the assurance that 'all you'll do is research'. In a way, since we live in a society, workers need to feel personally responsible for /constantly/ finding the appropriateness (value) of their skills and elevate them as necessary to make sure they have a steady income and their craft is still valued by 'customers'. If you stagnate there, then you will gradually (or rapidly) become replaceable. And as you age, being indispensable becomes harder. So, perhaps, it makes sense to learn things that ensure stability of demand for your skills.

1- http://physics.wustl.edu/katz/scientist.html

The More Fungible Worker

Related Stories