The U.S. productivity slowdown: an economy-wide and industry-level analysis | Hacker News Reader

The U.S. productivity slowdown: an economy-wide and industry-level analysis

bls.gov

200 pointsrsj_hn179 comments

> The economic gains brought about by labor productivity growth make it possible for an economy to achieve higher growth in labor income,5 profits and capital gains of businesses, and public sector revenue; these economic gains also hold the potential to lead to improved living standards for those participating in an economy, in the form of higher income, greater leisure time, or a mixture of both.

The use of “possible” and “hold the potential to” are telling.

My belief is that as the profits and shareholder compensation have risen disproportionately to wages, the workers who are the key to unlocking the productivity growth have lost their incentives. Why increase productivity when your pay doesn’t increase? This is obvious with the current anti-work movement and labor “shortage”.

If the wealth and growth was shared more evenly, and greed was kept in check, I believe we could easily be at the 2% rate.

I agree with the moral principle of what you're saying.

But, meh. My belief is that this is just measurement error. The linked article is a bunch of yelling about a perceived drop in what amounts to the second derivative of per capita GDP. It's just not a well-characterized "problem" to be solved.

> The figure—$10.9 trillion—represents the cumulative loss in output in the U.S. nonfarm business sector due to the labor productivity slowdown since 2005, also corresponding to a loss of $95,000 in output per worker.

Exactly that. This is characterized as a loss but productivity has actually grown every single year except 2011 (which had zero growth, so not a "loss" either).

Maybe 1998-2004 were just exceptional. My guess is that's when computers started to take over larger swaths of the most productive industries.

> above-average growth of the late 1990s and early 2000s

On top of that, the article (at least to me) reads like there was about a decade of above-average growth whereas it was really more like 6 years.

Edit: This reads a lot like this: https://xkcd.com/605/

aside from the massive adoption of a global distributed communication network enabling communication between almost everyone on the planet, coupled with a vast repository of saved knowledge, I can't really think of anything that might have improved productivity in that timeframe.

Growth as a norm in industrial sections should be expected even given "fixed" levels of experience from industrial learninh curves as processes grow more optimized and efficient and features move from revolutionary to old hat.

Of course there is a matter of growth in what too - product count and quality (utility value) in some combination or profit. It is a known irony paradox that sectors which grow more efficient become less paid relative to what is produced. Part of the general economic trend of diminishing returns via saturation.

> The linked article is a bunch of yelling about a perceived drop in what amounts to the second derivative of per capita GDP.

The Bureau of Labor Statistics is not in the habit of doing a bunch of yelling. This is a scholarly report, and is commenting on a widely reported phenomena, namely the productivity slowdown. If you don't like this analysis, there are many others that draw similar conclusions. The reason why this analysis is interesting is that it breaks down the productivity decline into component factors and identifies which factors have slowed down (MFP).

The labor shortage is plenty real and is going to be for a long time.

We could massively expand immigration, but we won't.

Which would also massively help with our population growth and funding entitlements problem.

We celebrate quickly giving citizenship to pro bball players (Enes Freedom, who btw I think he's awesome and congrats). But it's inaccessible for so many that would be net contributors to our country!

I firmly believe it's rooted in racism & the electoral benefits of Republicans rather than an actual policy/economic debate.

Doesn't help when one political party and their news media empire catastrophize and fear monger dangerous, gun toting, drug muling, cartel infested, caravans of chain migration invasions every October on the dot. But perhaps now I'm the one politicking ;)

Expanding an underclass of poor migrant workers does not help anyone. How would population growth help us when we also supposedly have a housing shortage in almost every city? Of course we should try to get elite athletes and academics to come to America, but the picture is much less clear for other types of immigration.

The sum is greater than parts

So can we let skilled tradesmen immigrate then?

Your thinking goes against the entire history of the US. The goal of having immigrants is that they will come to work and improve their and everyone else's lives, not that they will come rich to support the country.

we already have a large pool of migrant workers and it seems like there isn't enough.

right now there is a lack of this 'underclass' labor for farms and other intense, low pay industries. they can't function with covid restrictions on that non-U.S. citizen migrant labor pool. So they aren't stealing jobs they doing jobs that otherwise go unfilled [1]

your comment also just writes off an entire class of people and their contributions. America is built on the skill & labor of the 'underclass' and historically there has been movement up the economic ladder.

This article and top comment points out that this economic improvement is currently lopsided against low income workers who are producing the net productivity gain for the owners/corps.

Housing shortage doesn't have anything to do with immigration. But that's a separate and worthy policy change to discuss as well.

There is a lot of research that shows current undocumented immigrants are a net benefit to the US purely economically speaking. Plenty of sources here [2]. I'm agreeing with parent comment in it would also solve our population growth & entitlement problems.

another interesting point in this artificial distinction between skill and unskilled: there is a talk on HN of tech corporations exploiting the high skilled tech migrants which you seem to advocate for.

using the really harsh immigration laws - for instance can't legally be unemployed or between jobs - to get lower than market price labor at the expense of US citizens on this board. Not as bad as holding passports but it is effective

[1] https://www.nytimes.com/2021/10/25/business/economy/foreign-...

[2] https://en.wikipedia.org/wiki/Economic_impact_of_illegal_imm...

> Housing shortage doesn't have anything to do with immigration.

It does if immigrants want to live in housing. (And I say that as someone who is overall quite pro-immigration.)

Housing demand is highly location significant and immigration for the relevant jobs are in areas of the country where there is a demand shortage. So little in this context.

Unless you are retired or a recepient of passive income chances are that if housing in an area is very cheap, you won't be able to afford it. That is the reason prices got so low is a lack of either income generation or demand from those well enough off (vacation homes do not need to be within commuting 'distance-factored-as-time').

Strongly disagree. Especially in tech, many of our success stories are from people who themselves are first generation Americans with parents that arrived here with literally nothing.

American poet Emma Lazarus wrote “The New Colossus,” which now hangs in bronze in the Statue of Liberty’s base. The end of the sonnet goes:

“Give me your tired, your poor, Your huddled masses yearning to breathe free…”

America remains competitive by keeping the door open.

If we slam the door shut, we could be missing out on the next Steve Jobs.

>“Give me your tired, your poor, Your huddled masses yearning to breathe free…”

We already have millions of tired, poor and huddled masses. Drive down the street in San Fran, LA, Philidelphia or virtually any other big city in America and you'll see their tents on the sidewalks. The idea that its somehow a good idea, let alone necessary, to import millions of poor, unskilled foreign laborers while we have millions of poor Americans already here is wrong and offensive. The somewhat widespread idea that opposing immigration is racist rather than a show of support for the millions of poor Americans we are allowing to languish is similarly offensive.

The US is going to get to a population of 333,333,333 people in short order. Is that not enough?

While importing cheap labour and working them hard is a road to prosperity - probably quite a good one for all involved - it is hard to see how there can be a pure labour shortage with that many people, that much wealth and that much history of advanced manufacturing.

There is probably an organisational problem here somewhere.

The absolute number of people tells you nothing about the state of the labor market. Every additional person demands services, which creates jobs to fufill the demand, and then the employees of that job use their income to pay for goods and services they like, and the economy grows... that's the entire basis of capitalism and economics in general. The economy isn't a zero-sum game.

For example, when Orlando had a sudden influx of people move there from PR after Hurricane Maria: "we find that employment in Orlando increased, especially in construction and retail, and find positive aggregate labor market effects for non-Hispanic and less-educated workers. While we find that earnings for these workers decreased slightly in construction, this was balanced by earnings growth in retail and hospitality. These results are consistent with small negative impacts on earnings in sectors exposed to a labor supply shock, offset by positive effects in sectors impacted by an associated positive consumer demand shock."

https://www.nber.org/papers/w27718

Instead what you have is a sudden negative supply shock, due to the pandemic, while simultaneously having a demand shock for goods because of stimulus. Of course prices for the labor demanders will increase.

Much more likely is we increase automation in jobs that no one wants to do like burger flipper or artichoke picker. And this would be a huge productivity win. It also might be rough for society but for productivity it would be huge.

Automating crop picking might be profitable for the growers, but it isn't going to unlock much productivity, it's currently done in a few weeks by a (relatively) small number of people.

Probably the same deal with quick serve. They already automate individual tasks as it makes sense, they aren't trying to automate the job.

Freeing up labor resources can, in theory, increase productivity in other areas.

Say we have all these artichoke picker people, they get replaced with machines. Those people can now do other work while the artichoke picking output remains the same.

Labor is not fungible, because education and training is far more costly than most people think.

Especially when transitioning from a low-skill job to a high-skill one.

If you think there's a big productivity slowdown now, then oh boy are you going to be surprised when you start massive retraining programs.

Manual crop picking is an economic bottleneck - if prices drop to zero or below (including subsidies) leaving it to rot in the field becomes the main viable option, barring another process which could transform the crops into something profitable.

Not sure why you are being downvoted. The immigration policy especially for high-skilled workers is rooted in racism and is ignored because no party benefits from it. The only solution is for workers to form a PAC and raise funds to influence the political agenda.

Labor productivity = output per hour worked. Adding people does not change labor productivity.

Mostly true, to a point... But there will usually be a point of diminishing returns on marginal labor productivity, when we hit the limits of capital utilization.

But that's not really relevant in thr current labor market... it seems clesr enough that we're way over at the other end of the scale.

Ive reached the same conclusion, with some additional ideas to why. In recent decades the availability of capital has increased dramatically. Low interest rates, spray and pray venture capitalists, creation of new dollars at a rapid rate.

The idea behind capitals returns were that it took great risk as capital is rare and hard to get. If capital becomes easier to get, that alters the risk/reward ratio. I think we are just watching the reward part shift to where it belongs in contrast to the lower risk.

> In recent decades the availability of capital has increased dramatically.

Yeah, if the receiver of the capital is already rich and/or well-connected. If you're not in the top 10%, all you get is 10%+ APR credit cards, payday loans and other usurious crap.

You don't need to be in the top 10%, you just need to be likely to be able to repay your loans.

Government subsidizes home loans, student loans, and small business loans to the tune of hundreds of billions of dollars a year through the Fed's asset purchase program. The reason broke people can't get anything better than a 10%+ APR credit card is because they're so unlikely to repay the loan that the issuer has to charge that much interest to break even.

I actually took a look at my banking app and I'm offered "personal loans" up to $20k at under 6%, which is actually incredible considering this is an unsecured loan to a random person.

Do loan providers typically break even?

Even with the federal loan systems you mentioned, they’re making a profit each year.

You make fat profits in normal years and potentially lose out catastrophically during Great Recesssion-like events. Lots of lenders straight up went bankrupt due to defaults.

Of course since the government made it so you can't discharge student loans in bankruptcy that market isn't functioning anymore.

Is there any evidence of this in the data above (or elsewhere)? Labor productivity is a low degree-of-freedom datapoint and your explanation is indirect and has a lot of degrees of freedom.

My opinion was anecdotal. But I think it could be argued the entire field of economics has a lot of degrees of freedom - which is why they do not have firm conclusions as to the causes of productivity loss in the article.

Edit: also see: https://www.epi.org/productivity-pay-gap/

It is possible that the drop in productivity growth, if it goes on long enough, will have the effect of closing the gap.

Why/how are the workers the key to unlocking the productivity growth? I imagined productivity growth came through mechanization/electrification/computerization, not from the workers themselves being more skilled.

IE if warehouse box-stackers productivity has tripled since 1950, it is because companies have invested in forklifts for the workers to use, not because the box-stackers are just way better at stacking boxes than their 1950s counterparts.

Strange that we're talking about a shortage of workers to unlock productivity then, and not a shortage of box-stacking machines or forklifts.

Obviously someone needs to operate the forklift. But the point is the productivity gains aren't coming from the person driving the forklift, they are coming from the capital investment in the forklift. Because if the forklift wasn't there, the person driving it would instead be stacking boxes by hand.

If the driver was going to capture all of the upside of the productivity gains of the forklift, the business would never invest in a forklift in the first place.

Most capital investments do require a more skilled workforce to get the expected return on investment. In your example, a box stacker requires virtually no training, while a forklift operator has to take a training course and be skilled enough to operate the vehicle without costing the employer thousands or millions in liability claims. And no one is arguing that the employee should capture 100%, just more than roughly 0%.

> Most capital investments do require a more skilled workforce to get the expected return on investment.

Does it? I have no idea whether this is true or not, but there have been plenty of counter examples throughout history. The spinning and carding machines during the industrial revolution, automated checkouts, ... Even in your example, the next step in an automated warehouse where the employee is just there to make sure everything is working correctly.

> employee is just there to make sure everything is working correctly

this employee is going to require more skill than the original box stacking employee.

The savings to the company is in the form of a lower number of employees, making the total cost lower. AKA, one maintenance engineer, replacing 10 manual stacking employees.

In this case, productivity per employee grows, and it required capital investment, both in the form of machines from the business, but also from society in the form of education (presumably, the maintenance engineer needs to understand how the machines work to make sure it works correctly). The business doesn't really pay the education cost directly.

They're not coming from the capital investment in the forklift. They're coming from the design of the entire logistics and marketing chain, which - for now - happens to use forklifts in one part of the process.

That part of the process could just as easily use robots. Or - potentially - genetically modified chimpanzees.

Value is contextual. If productivity is dropping instead of increasing, one possible reason is because the friction in value chains is increasing and both intelligence and efficiency are decreasing.

Another is because returns are being measured in the wrong ways, and the contextual element is being disguised or lost by current accounting practices.

These systemic issues are more plausible causes than any suggestion that front line workers are all somehow becoming less good at their jobs.

This is a worthy idea, but not it's not necessarily substantiated.

FYI 'Productivity' is not 'How Hard We Work' - it's a measure of profit/hour.

If profits go down for any reason, including more leverage by consumers due to increased competition, productivity will go down.

'Low Productivity' does not mean the economy is doing poorly.

One other theory, is that consumers have more power then ever, consumer surpluses are greater than ever (through lower prices) and that the money is not going into the companies and therefore not recorded as productivity.

Facebook, Google, Web Browsers, News, email. So much free content.

My favourite Video Game is free to play.

You can buy at Wallmart today what would have cost a lot of money in the 1960s.

Travel is cheaper than ever, more people are travelling, Airlines are always teetering on bankruptcy.

All of that means the productivity numbers are going to be lower than otherwise all things being equal.

> FYI 'Productivity' is not 'How Hard We Work' - it's a measure of profit/hour.

Productivity is revenue/unit of time (generally hour).

Yes, I stand corrected.

Or more like 'How much stuff/value is created with what amount of money per hour'

Productivity is based on labor hours not cost.

In my labor intensive business processes, I’ve used technolgy to control cost. My CFO doesn’t give a rats ass about headcount, just dollars. The people we hire or contract are dumber and cheaper today than they were in 2005.

Whole industries run like this. Hotels, government, manufacturing, etc. You might have a company with 3,000 people where key decisions are made by 5. Smartphones and services replaced supervisors and dispatchers with text messages. Many of those labor hours are inefficient as a result, but very cheap.

>Productivity is based on labor hours not cost.

Exactly, a distribution warehouse is a fixed size, it has a maximum capacity (floor space & racking), so despite being able to bring in more boxes and also hire more (agency) staff Hours to process the Units in and out, at some point that warehouse will grind to a halt and become like a traffic jam caused by too many people and too many Units much like we see with too many vehicles on the road at peak times.

Automated distribution warehouses (computer controlled conveyor belt systems) also have this problem, as do theme park rides, rail networks and airlines as other common examples. In the 90's automated warehouses in one company I was familiar with would also crash at peak times, namely Xmas because it couldnt handle the increased volume. When it crashed work was distributed to other warehouses in the network to balance the workload amongst the remaining operational warehouses. However with automated warehouses, management/board know its maximum handling capacity and just accept it, unlike human manned warehouses where management keep pushing for more until workers revolt!

This is why the Reddit AntiWork subreddit is interesting right now, we are seeing a country wide revolt which could spread to neighbouring Western countries. The spooks will be watching this like hawks. During a UK fuel protest in the UK, a protestor who was also a small business in the run up to his attendance at a second protest had a visit from 4 big burly guys in a new expensive Black Range Rover, one of the guys got out and told this small businessman, if you go to the next fuel protest, you wont have a business to come back to. Thats how the security services and others operate here in the UK! Its the security services job to protect the country which includes the economy and other things, thats also why the illusion of privacy and tech security is just an illusion!

So other factors that can affect productivity in a country is poor transport networks, if people are spending more time travelling to and from work, then that country will be facing problems in years to come unless changes are made.

Look at commute times dropping during covid lockdowns. It was like getting to work 20 or 30 years ago, less traffic on the road yada yada yada.

Now changes in legislation can offset some of this which is in the hands of a Govt, like sunday trading (being open on Sundays) and also 24hr trading (being open 24hrs a day), internet shopping can also offset some or redistribute the demand on infrastructure but even the internet retailers have their problems and inefficiencies but these are now offloaded to the consumer, so legislation has created a cost saving for internet retailers in some countries.

I do agree that management are largely detached from the situation on the ground and you can see this in many ways with the problems that exist, but either they dont care or they dont know, some really are detached from the reality on the ground. Its also cheaper to address these on an ad hoc basis when intelligent consumers complain, management rely on stupidity and laziness to get away with poor goods or services. Remember inflation only measures the cheapest price of a good or service, it does not take into account the decline in quality or life span of a good or service.

> This is why the Reddit AntiWork subreddit is interesting right now, we are seeing a country wide revolt which could spread to neighbouring Western countries.

how big is this movement, actually? personally, i quit my job without knowing about “anti-work”, and without the goal of making any point. then i was very confused when multiple people IRL told me they respected me for it, and learned about anti-work. it happens that there’s some overlap between their complaints and my reasons for quitting… but their complaints cast such a wide net that it’s not surprising. so i’m not yet sure how much the anti-work label or the community actually affects things on the ground v.s. just showcases popular sentiment and actions that people would have taken anyway.

I read the parent as the anti work reddit is a symptom, not a cause. That the malaise you felt is common enough that large amorphous groups are loosely coordinating in a way that is seen clearly on one of the biggest websites, but if that subreddit didn't exist, the malaise still would.

Goldman Sachs gave it some public recognition recently in an advisory to investors or something. News orgs picked up the Giant Squid advisory as you can read here https://www.forbes.com/sites/jackkelly/2021/10/21/unemployme...

https://finance.yahoo.com/news/antiwork-movement-may-be-long... The subreddit is about a tenth the size of the WallStreetBets thread of GameStop mania fame, but Goldman points out that Antiwork now ranks above WallStreetBets in comments per day."

Its English speaking, so it can spread to 5 Eyes country's easily and other country's especially if native news outlets run with it. US social media dominates so it will have an impact in other country's.

Its aimed at the millennials which is why its on Reddit.

Lots of factors or trends could be influencing it even sock puppets, however my belief is it is an attempt to stimulate inflation short term to get interest rates up. BBC mentioned Bank of England could see interest rates going upto 5% by the middle of next year!!! Who can afford debt repayments where the base rates gets to 5% and the lender has added a few % more on top for them??

We havent seen anything like that in at least a couple of decades.

Inflation also forces people back to work after the Covid lockdowns but the effect is, some businesses have realised some staff can work from home and save money on expensive premises like office blocks, such is the technology today namely the internet, voip and remote working in general. Others are realising that some sectors are best avoided for employment because of the way they laid people off during COVID lockdowns. Karma I guess!

However GameStop is also an indication of a rebellion which is why GS have mentioned aNTIwORk, even the regulators know they need to write a more honest fairer rule book and their inaction and coverup is angering some, which is why Gamestop got popular.

You can measure this sentiment by analysing the use of words and phrases used in comments across different social media platforms. People give away so much meta data on social media, its getting easier to predict when someone might sneeze! LOL This is also what Goldman Sachs hinted at when they mentioned how many comments its getting now. GS monitor social media comments for insights into financial trends which would be entirely expected of them in my opinion, just like many other big companies will be monitoring social media for trends, including this site.

>The subreddit is about a tenth the size of the WallStreetBets thread of GameStop mania fame, but Goldman points out that Antiwork now ranks above WallStreetBets in comments per day."

Is that really surprising considering that the gamestop mania has fizzled out?

>Lots of factors or trends could be influencing it even sock puppets, however my belief is it is an attempt to stimulate inflation short term to get interest rates up.

Are you saying the government/central banks are doing this to "attempt to stimulate inflation", or that the readers/commenters on the subreddit are doing it?

Govt's & Central Banks want and need inflation. https://en.wikipedia.org/wiki/Helicopter_money

When some parts of the world have a deflationary economy due to the boomer population hump flattening out as they retire and die out more quickly, and you have been printing money under the label of Quantative Easing since 2008 so need to cancel out this money printed for liquidity reasons otherwise you risk killing your currency, a way to do this is to stoke up inflation and blame the public & businesses as its always been one of the ways to cancel out debts.

When the financial crisis in 2008 kicked off, the velocity of money dropped sharply as you can see here. https://en.wikipedia.org/wiki/Velocity_of_money#/media/File:...

This was called a liquidity trap so QE money was used, like grease on wheel axles, to get transactions increasing which then helped get the economy move again. However all of these different actions, just like different types of taxation has a range of effects, some become ineffective over time, so the default "reset" is generally to stoke inflation as its erodes debts more quickly.

So if I was a young person just leaving education right now, I'd be taking on debt to purchase assets like property, mindful that population numbers in some countrys are shrinking which will create a surplus of property and drive down prices now. Over a decade or two, their wages will be significantly higher in a high inflationary period than their wages over a decade or two in a stagnant or low inflationary period, because they wont have external factors to justify going to their boss asking for a pay increase.

Likewise Reddit AntiWork is another way to give these young people ideas to ask for better pay and conditions when they perhaps wouldnt have got it from the older generation who tend to carp on about their poor pay and conditions whilst ignoring the fact that back then one wage could buy a reasonable sized house and support a family unlike today where two wages are need to achieve the same results.

The existence of this subreddit also makes it easier to quantify negative sentiment towards current pay and conditions when one didnt exist previously so it also becomes an intelligence tool for law enforcement, policy makers, investors, and traders.

TBH I'm surprised its taken this long to appear on the radar, but I guess other subreddits have served their useful life span.

Edit. You also have this https://en.wikipedia.org/wiki/Maslow's_hierarchy_of_needs evolving as people age because we dont stop wanting things but these things we want change as we age unless we stop aging!

https://en.wikipedia.org/wiki/Wasp_(novel)

This has been my impression for the last quarter of a century. Better communications ultimately give lower efficiency. Why plan when you can micromanage?

Definitely. Another factor I think about a lot is what I call "managerial diversion". As a simple example, imagine one of those meetings where a boss calls it to feel important. E.g., a "progress report" meeting where each person tells the boss what is going on while everybody else just tries to look alert.

To me this is a diversion of resources from productive use. Managers get to feel powerful, important, smart. But neither investors nor customers benefit. At the team level, it's often just wasted hours here and there. But as you scale up, it gets worse. I've plenty of expensive corporate projects that were really just resume-builders for some executive.

How much more productive would each of us be if the management chain was reasonably effective and reasonably selfless?

This opinion feels to much like a cliche. Management needs to know what’s going on, even in well run companies. You might not like giving progress reports, because you feel like it takes time away from your job, but it’s possible that management is using the progress reports to better the company as a whole. For instance, if you’re behind schedule they can allocate more resources to your project, etc. I’m sure you have many anecdotes where management wasted peoples time. That happens often too.

i think the complaint is not giving status reports, but being forced to listen to others’ seemingly irrelevant status reports.

Exactly. As well as more efficient ways of keeping people updated. Think of all the people who have to enter a ton of stuff in to Jira and then give the boss a verbal status report.

And let's not forget obviating the need for status reports entirely. I've been part of teams that got solid work done without ever having to give status reports to anybody. We'd ship early and often; our "report" was frequent results.

Big meetings are more time effective and easy for a manager but worse for many workers and productivity.

This has been a huge advantage of the WFH culture lately. I can just keep working while I listen to other people update the manager.

Why do this?

Use that time to read or browse HN or something.

Lets say one hour meetings per day. This means 8-1 = 7 hours of real work. If you work the 8 hours + the meeting, if you ever have to go back to the office, either you will have to work 9 hours to produce the same as WFH (since now you have to sit on the meeting), or drop in productivity and get scolded by a manager (you are producing 12.5% less)

More often than not, I’ve spent about 20% of my time giving status reports or being around while others do. Very little of it has seemed like it was important or valuable for anyone and more of an exercise in most people finding something to say for the day.

Reporting does have value but has become way overdone.

> For instance, if you’re behind schedule they can allocate more resources to your project, etc

I have never seen this happen in my entire life, nor would it be useful

> For instance, if you’re behind schedule they can allocate more resources to your project, etc

See: The Mythical Man-Month

I agree that making sure that your work is observable is important, and usually Management's attention needs to be called to where the observed changes are being made. This is not the same as holding you hostage in a meeting culture where you need to sit through everyone else's reports on their progress. Instead of people taking five minutes to write out their progress report to their manager, who can then skim the progress reports and the work for a full team in fifteen to twenty minutes, instead we hold everybody hostage for thirty to sixty minutes. Typically this is a syndrome of a culture where labor is not expected to document their work, and nothing will scale without pulling senior engineers off their IC work to synchronously bring people up to speed. This necessarily reduces engineering velocity to a crawl.

Nobody can speak as fast as a progress report can be read. Verbal meetings cannot be indexed or searched. All meetings should have a product - a decision, language that is agreed upon, written out, saved to a knowledge base. If you call too many meetings, it's a sign that you're not delegating authority enough.

The original poster said resources, not necessarily human "resources".

The Mythical Man Month was also not saying adding manpower to a project makes it slower, simply that it potentially could and there's limits to parallelism.

Everything else mentioned rings true though.

The mythical man month does not apply as a default. Often in orgs, there are other employees equally versed in all aspects of the work you do, but are just currently working on another aspect of the project, or codebase.

Thus, some work may be immediately taken off your plate, without you having to dedicate your own time to training, etc.

Meetings suck. However they absolutely have a purpose, and really, sadly, you are never "held hostage" in a paid-for scenario.

> Often in orgs, there are other employees equally versed in all aspects of the work you do, but are just currently working on another aspect of the project, or codebase. Thus, some work may be immediately taken off your plate, without you having to dedicate your own time to training, etc.

Not everybody's time is worth the same, and it's a classic organizational fallacy to treat your employees' time as having equal worth. Simply put, five minutes of a senior engineer's time is worth more than five minutes of a junior engineer's time. This is clearly obvious even if you only compare their salaries (and not the value that each worker brings to the company). If their time was worth the same amount, we wouldn't hire juniors in the first place.

Thus, it's inherently unreasonable to use five minutes of a senior engineer's time to save five minutes of a junior's time. I expect juniors to struggle with something for at least a few hours before asking a senior for help.

This is besides the point that, if employees are documenting their work properly, there is less need to synchronously synchronize. "I don't remember, read the documentation, that's why I wrote it."

Yes it does, you are not taking into account communications overhead, which as a project grows larger becomes a significant factor.

More resources doesn't mean more man months. It may mean assigning the most productive and expensive people to the project for example.

This is not the 1950s, we are on a software forum on the internet. Lots of metrics can be extracted directly at source, lots of status can be extracted from auto-distilled tickets.

The management situation you posted is also a bad management decision: https://blog.devgenius.io/why-adding-more-people-to-a-projec...

But i digress, fractally, outwards, not really caring..

Wouldn't the companies that allow this "managerial diversion" quickly fall behind competitors that did not allow it?

Companies also have different ratios of managers to employees. If managers drag down productivity wouldn't you expect to see much flatter orgs in successful companies.

Only to the extent that a) there is direct competition sufficient for this to make a difference, and b) bad management isn't an endemic problem across wide swathes of American business. But there are widespread concerns about declining competition. [1] And some, me included, see American business culture as captured by managerialism. [2]

By your theory, all companies should be pretty efficient and managers pretty effective. But how many people would say that about their companies? If that's how most of your friends talk about their bosses, let me know where they live and I'll move there.

One good example to look at here is the notion of bullshit jobs: https://en.wikipedia.org/wiki/Bullshit_Jobs

[1] E.g.: https://hbr.org/2018/03/is-lack-of-competition-strangling-th...

[2] This is a good book on the topic: https://www.amazon.com/Confronting-Managerialism-Business-Ec...

From bullshit jobs it looks like 37% of some surveyed Britons thought that their jobs did not contribute 'meaningfully' to the world. So that means the vast majority do think they contribute meaningfully.

Most people I know do seem to like their manager, if they don't they switch jobs. But tech jobs are also plentiful here (SV).

About half the places I’ve worked did fail and the other half didn’t have competition adequate enough to cause failure.

Once you achieve a certain amount of success and market share success stops being driven strongly by quality fast execution and you can spend decades in a slow decline unless you’re lucky enough to have a disrupter targeting you which you can usually just buy.

Employ a lot of people to inefficiently maintain what you already have and you’ll be fine for a good long time. A lot of this work doesn’t need to be done but nobody gets a promotion for doing more with a smaller team.

It’s like the problem with the Navy buying ships, big ones are less useful and more vulnerable but captains and admirals feel more important with bigger ships and drive decisions that way.

A lot of what happens is to make middle management more money, look more important, and keep busy.

This assumes that competition exists and self-regulates, it doesn't (always, or much in later states of a system), contrary to a lot of economic theory that's peddled as fact by those who benefit the most from it.

The market only cares about aggregate performance. If every company has its own set of problems that are difficult to fix, the situation is stable. Especially if the market is dominated by a small number of companies, which makes it unlikely that anyone manages to fix their hard problems.

I feel like flat orgs are a resounding success. Valve comes to mind. Netflix to a degree.

I'll sometimes just give a 10-second status update: "Working on feature X, it's coming along nicely", or "Fixed a dozen pesky bugs".

Eventually, others start to do the same. Granted, there will still be a couple of "long talkers".

The MBA hacks need to slither in and “create value” from things that they have no real expertise in. Look at the debris field they created value GM, GE, Intel

You should try working in management before coming to conclusions

Nah, that would require doing work and something that is outside their comfort zone.

I think that having a person on staff who has special training in organizing would be helpful for large teams, but MBAs are absolutely not that.

You should try not making assumptions about what people have and haven't done.

> the workers who are the key to unlocking the productivity growth

What makes you say that?

> Why increase productivity when your pay doesn’t increase?

You're conflating two things - industry-wide pay changes (which any individual worker's productivity-related decisions have ~zero influence on) and individual pay changes, which are in fact determined (partially) by individual productivity.

The title of the article omits the word "labor" but the article is obviously focused on labor productivity so lets starts with a simple definition from Investopedia: Labor productivity, also known as workforce productivity, is defined as real economic output per labor hour. Growth in labor productivity is measured by the change in economic output per labor hour over a defined period.

I believe (through observation/experience) that workers are the key because without their buy-in and support, none of the business strategies or technological innovations management can introduce will matter. Individual workers have a lot of control over the economic output they choose to contribute. This is, IMO, very much about motivation.

RE your second point, I'm not sure if I'm conflating the two vs using a micro example to extrapolate a macro trend.

Do you think that ~any of the productivity growth over the last 200 years has to do with increasing laborer motivation rather than increasing technology? There is scant evidence for the former and plenty for the latter.

Oh I absolutely agree with you here that technology was the driving force for the last 200 years. But it was coupled with high motivation. As motivation decreases (see the current anti work movement and anti capitalist sentiment in this country), all the technology in the world will not make up the difference.

Worker motivation to take advantage of new technology is extremely important. The more you pay for workers to learn a given skill, the more likely they are to learn it. No amount of extremely efficient technology can help if your employees are incompetent at using it.

> Worker motivation to take advantage of new technology is extremely important

They're motivated by not being able to get a job if they don't know how to do the job. There's an equilibrium state where workers and employers are both investing some amount into training.

Wild thought here: how does changing demographics influence this? Retiring people in many cases literally can't be productive anymore, but demand capital returns. But because they, in aggregate, didn't have enough kids, there isn't enough productivity to service that demand.

I’ve wondered this myself. Boomers are 25+% of the population. That large of a percentage dropping out of the workforce is going to leave huge areas for younger people to move up and fill in, leaving tons of the bottommost jobs (most service jobs) unfilled. At some point the Republicans will have to come to grips with their xenophobia, since immigration seems like the only solution.

Japan never came to terms with their xenophobia and opened up to immigration, so I would be careful about assuming it’s a foregone conclusion.

> bottom most jobs (most service jobs) unfilled

which would encourage even more automation - filling those jobs with immigrants aren't the only option. I would garner that using immigrants for cheap labour is a form of exploitation which should be discouraged too (tho i dont think it's likely).

There's definitely some very big problems with people demanding fixed interest returns when demographics aren't good: https://realinvestmentadvice.com/the-unavoidable-pension-cri...

I'm having trouble following your logic, if productivity is defined as revenue per labor hour, and you're saying this is due to the anti-work movement and labor shortage, that's decreasing the denominator.

Or are you saying workers have become less engaged and less productive at work but maintain they're jobs? That might be plausible

I took it to mean that workers did what was asked and no more — so productivity stayed constant.

Which was contrasted with engaged employees, who sharing in more of the profits, have an incentive to increase productivity — eg, by inventing a tooling improvement for the assembly line.

I think the overall point was employers are “penny wise, pound foolish” in that small productivity “gains” caused employees to disengage which is causing long-term slowdowns in improvements — improvements typically driven by employees stepping up and doing more.

Toyota empowers workers to stop production to fix quality issues. But they earn less than unionized workers.

Empowering line workers makes sense when there's growth and healthy margins to pursue. If the industry is mature or even shrinking then there's a heavier focus on cost cutting and metrics/Taylorism to maximize free cash flow and pay it out as dividends and stock buybacks.

Absolutely the latter. A huge part of the movement is sticking it to the man, do the bare minimum, do not go above-and-beyond etc etc.

This isn't economics, but I will say that, in much of the past, Americans had a belief in being productive, contributing members of society, in anyone achieving anything ('you could be President some day!'), and in an obligation to maximize your potential.

As wealth has become concentrated and the wealthy have openly and arrogantly adopted a philosophy of taking as much as they can - that it's what capitalism is about - rather than producing and contributing and thereby earning, it's no surprise that workers might not feel they should do more than the minimum.

> My belief is that as the profits and shareholder compensation have risen disproportionately to wages

It's not that simple: https://www.brookings.edu/wp-content/uploads/2015/03/finalin...

(From https://www.brookings.edu/bpea-articles/deciphering-the-fall...)

The takeaway is that these additional gains in productivity are being captured by land (as expected based on the law of rent). The problem is that this is popular when the majority of voters (or at least, the ones active in politics and making their opinion heard) are homeowners.

I believe it’s a mix of an explosion of recreation and entertainment options that people can’t wait to get back to and it getting increasingly hard to maintain the paradigm of work being a virtue.

The promise was it was going to get much cheaper to live. That didn't happen, and here we are.

Some make the case that cheap goods from China via Walmart was precisely that.

Was not enough.

On family wage income, yes. However, moving that stuff offshore also reduced wages as service jobs became the norm. For many, buying power dropped.

The metric in question might be "labour share of income" and it has indeed been decreasing for a long time.

But it's not obvious that incentives in individual jobs are affecting overall productivity (can one factory worker really make the conveyor belt move faster?). A plausible explanation could be that it's becoming easier (and profitable) to hire more low-wage workers than systematically increasing productivity.

My understanding is that cheap labor (China) is why there was nearly a complete stop of factory automation. Instead of increasing automation, low cost labor was utilized instead. Now that wages are rising around the globe, there should be increased innovation around automation. Explicitly that would be factory robots.

"Why increase productivity when your pay doesn’t increase?"

Because the neoliberal belief is that pay should remain largely fixed and the productivity bonus comes through as reduction in prices or increase in function.

So this season's iPhone does more than last at roughly the same price.

Which then leads to the paradox of productivity - since profits cannot then grow either unless there is ever greater leverage from banks.

The problem with 'productivity' measurement is output is based on 'sale price'

* Sell, 1,000,000 lottery tickets (or CDOs or Carcinogenic products) -- Boom, you are productive.

* Cure Cancer (or Linux or Wikipedia or an ad-free Educational YouTube Channel) and open source it for free -- Nah, not productive.

As we digitize, more things will be free and yet valuable and these things won't show up as productivity.

It's time to come up with better measurements of productivity

If the everyone receives the cure for cancer, or benefits from wikipedia or linux, then you would expect to see this show up in productivity numbers elsewhere in the economy.

If instead we invest in US productivity destroying activities such as dismantling factories, price gouging, speculation, mass diss-information and distraction machines, net loss corporations, and indefinite warfare. Then you would expect productivity to fall eventually.

Not necessarily.

Today we produce 1,000,000x more videos/memes/gifs/content 50 years ago and we also consume more of them.

But, those numbers don't show up at all in our productivity measures.

Why would watching "Gone with the Wind" 30 years ago more productive because we paid $10 than watching the same "Gone with the Wind" on a streaming service today (amortized at probably $0.05)

But memes have zero value. Or perhaps negative value?

Obviously they have value to the people making and viewing them. Entertainment and content do have value.

The ability to make sufficient for mass market consumption quality combinations of visual, audio and textual content has gone from being purview of real professionals who do this as their day jobs to something literally anyone with a smartphone, an option and 5min of free time can do.

That is a change to human communication as revolutionary as the telegraph or the telephone. There is definitely some value there. Lord knows who'll capture that value.

The attention memes draw has value and the productivity should show up in ad sales and increased want creation. Rather than unproductive sleeping, if we all stay up an hour later looking at memes + ads, that should increase total productivity (assuming we buy an extra coffee the next morning to keep us perky at work ).

That would increase GDP but I'm not sure it would increase productivity as a barista can only make so many lattes in an hour. Extra demand would require more staff who would work at the normal industry rate.

> Today we produce 1,000,000x more videos/memes/gifs/conten

Truth is nobody knows how to value those things in terms of money, to be honest I'm not even sure they can be valued the same way as we used to value physical things that got built during a physical production process.

I've seen this problem partially addressed by slightly leftish-leaning economists such as Mariana Mazzucato in "The Value of Everything" [1] and I'm sure there may be others like her, but afaik mainstream economics is still ignoring the issue, I think mainstream economists don't even acknowledge this as being an issue.

[1] https://marianamazzucato.com/books/the-value-of-everything

Thanks for the link. I've been building a hobby looking into this space.

Personally I think we can use Vickrey Auctions for this problem. Vickrey Auctions are able to give not only price discovery, but also value discovery. Everything digital has infinite supplies, that we artificially constrain so as to pump prices. If instead of using price discovery, we use value discovery, the price can stay at 0, so long as a certain (limited) amount of people are faced with a choice that would have them go without.

So for example perhaps we sample 50,000 people and say 75% of people keep access, say when they vote. If you bid in the top 75%, you keep access for the period, and you pay the value that the bottom person that kept access bid. That money is then not used to pay for the IP, some other tax will fund the IP or Open Source Software or Meme, that 25% will then go to all the voters for providing their values. So the average bidder will pay nothing, the below average bidder will lose something they value low, but gain, and the highest bidders will fund that. We then take some other collection and sample another 50,000 people, until we find out the value of all of these digital items to people.

If you do the subsidization process correctly, I think the idea of IP owners holding their supply back would go away, as releasing it to all would always provide more value than holding it back. I think at the very least this would be an obvious answer to Open Source Software, and this would easily fund things like Linux and Firefox. I think this would stretch further to research, and colleges would now be funded through this mechanism, and the professors who currently waste enormous amounts of time doing grant writing would be able to be productive. Thus the whole issue of journals goes away. Even things like ads, people would bid on their service less because of them, so there would be no fundamental reason to include them as it would reduce the subsidy.

There is a reason they are ignoring it. Their notions of value aren't workable or coherent. If I manufacture a gun and it is used by a rancher it reduces predator-on-livestock losses. If I manufacture it and an eventual buyer uses it in a school shooting it causes large losses.

There is this weirdly persistent leftist idea of demanding absolute precognition with values for things which are determined in the future and utterly irrelevant and unknown. If I buy coffee beans from locals of Cape Horn growers for my use at the local prices I am helping by supporting local farmers. If I roast them and sell them overseas for 100x the cost I am somehow harming them now.

The Marxist derived notions of value and exploitation are downright certifiably insane.

You aren't harming them but you should share some of that 100x? Of course the market would take care of that by transmission of information through a horde of you buying for the 100x gain. Eventually they'd share in that. So you see, the concept isn't wrong, just that there is already a mechanism for it.

I read her book, it's quite good and makes good points. I studied economics and her points make sense. There truly is a big value hole, intellectually. Our current default is actually not a theory of value at all, it's a hand waive that allows us to not get stuck in a deep conversation about what is worthwhile.

Of course the problem is that any theory of value like the labor theory has massive holes as well. That's why the current subjective value theory is not really a theory: it's not wrong, it's a placeholder.

Surely videos/memes show up in Google/Youtube's quarterly reports.

30 years ago people valued watching "Gone with the Wind" at $10 and today at $0.05. To serve that, the distributor did a proportional amount of work 30 years ago and today (because it's easier to distribute today). Separately, it would take some effort to produce something equally engaging today with so many entertainment options to choose from.

By what measure are memes and gifs productive?

Ha. Art is one of the last stops in the productivity chain. We are productive so we can have leisure.

Does someone performing for free at your bar produce value or are they simply enjoying their leisure time? If they produce value without currency exchange, do the denizens of the bar trying to impress potential dates produce value?

"Productivity numbers", much like GDP, are mostly worthless because they don't measure what you are producing. If you pay one person to dig a ditch and another person to fill it in, the productivity numbers will be higher than if no ditch was dug, but have you really produced anything? The metrics we use don't just measure what we are calling "productivity" and "economic growth" but also drive and influence our collective behavior. When these metrics are flawed our behavior in pursuit of changing these metrics also becomes flawed.

If you were to compare two individual people using that type of metric, I would agree. But if you are measuring multi-decade trends for the 161 million people in the US workforce, the measure they use seems adequate.

We ostensibly have a "better" measure of productivity, called the GDP. Every time a carton of cigarettes gets sold, GDP goes up. every time a person is diagnosed with cancer, GDP goes up. for a capitalist, the GDP is generally a positive litmus.

the concern I think is most evident is that a nearly theatrical number of short-term and long-term issues are becoming insurmountable obstacles to progress at all. having coasted on Quantitative Easing and bond buys since 2008, the market has cheated recession at all turns and subsequently created a corporate credit bubble that has turned the prime interest rate into a third-rail for anyone seeking to raise it to counter now rampant US inflation. many point to 2020 as a recession period, however bond prices and home prices remained high, and it only lasted a month at most as the fed simply injected more cash into the system to "correct" the uncomfortable decline.

the minimum wage hasnt moved in a decade, and most service economy workers (those which make up the backbone of neoliberal capitalist society) faced with the near Sisyphean task of caring for COVID patients at home, educating their kids remotely, and working multiple jobs that offer no healthcare or medical leave reached its absolute breaking point when the government and corporations deputized most of them as mask police to be spit on and assaulted. paying people more isnt working.

finally the fed and the gov arent helping. the looming threat of regular petulant government shutdowns coupled with states that refuse to in many cases even acknowledge their covid numbers, is butting against Federal reserve dogma that laughably insists somehow this is just "transient" inflation and its just going to go away, despite the first decline ever in cyber monday sales on record.

the silver lining analysts all rally around is a trillion dollar stimulus bill just that wasnt even submitted to the house until nine months into the year that will arrive just in time for a 2022 meldown over what are widely anticipated to be poor christmas sales amid a driver shortage, shipping gridlock, and chip shortage.

to see the BLS flog capital intensity platitudes and labor composition functions is just bad comedy. its the same sort of bureaucratic blinders we had right up to 2008.

They won't show up directly but they should (hopefully) be observable via secondary effects. Linux itself doesn't have price but employees using Linux have salaries, Redhat support has a price. SaaS providers do this for countless open source projects.

Linux and free software are also used for thousands of small commercial websites and projects that show up. If everyone had to use Windows Server and SQL Server for their .NET wordpress-style site there would be fewer sites and fewer businesses.

Educational YouTube channels should lead to more educated and therefore more productive people. It could be argued that for better or worse the higher-ed and post-high school systems (degrees, credentials, etc.) have been very slow or even opposed at allowing this extra learning/knowledge to improve career/productivity outlooks.

e.g. Youtube is free but text books have increased X% in price so the net effect is diminished. Learning is easier and cheaper than ever before but most colleges are not using the more efficient learning systems (online classes, MOACs(?)). The cost of college has increased so much that college itself has decreased in productivity, etc.

Also a lot of the time spent on these free/non-profit projects are not counted as labor hours so if anything they should have zero or positive effect on the measured amount of productivity.

(Personally I think HN is responsible. /jk)

The 'secondary effects' are bigger houses than ever, more toys than ever, more entertainment than ever, safer, more fuel efficient cars, more vacation (it's not much in the US though), more vacation travel etc..

"Educational YouTube channels should lead to more educated and therefore more productive people."

Possibly, but also less demand for a big source of labour productivity, which is education.

And of course there is home life: if a Mother (or Father) stays home to mind the kids, all of the effort they do doesn't count towards labour input, GDP etc. but if they were paying someone for it, it would.

This is economic productivity. Why wouldn’t you measure it in dollars.

How do you account for differences in the purchasing power of those dollars you are making the measurement in?

An example of why this is relevant: https://www.investing.com/indices/zimbabwe-industrial

They do affect productivity and disproportionately one might say. The problem is that these things are very hard to measure. Say you have an article about how to generate Qr-codes and a few hundred people everyday use it to build Qr-code solutions for their businesses. The alternative is that they don't use Qr-codes because they can't afford to buy a solution.

How are you going to measure the productivity gains for that one? Many things were involved along the chain: Internet (and Cable networks), Search Engines, Publishing platform, somebody free time, etc...

Yep, what about child-rearing. Why isn't it incentivized/uplifted in this society? It's weird so many successes credited to self help, therapy, that petereson dude, elon musks inspiration, etc but why aren't we giving it up for moms??

My cynical explanation:

In every workplace, it's usually the few productive people who do all the work and others just dragging their asses around, mostly being a nuisance - they exist only because HR is doing a bad job on one hand, people learned how to pass interviews and then kiss asses to stay on the job very well, on the other hand, and managers want to stroke their egos by having as many people "under them" as possible.

Trick is, in the last 10-15 years those few productive people, thanks to the infinite scalability of success in the digital world, and the good stock market performance, have saved up enough money to kiss the job goodbye forever, and only those losers remain. Almost all good developers i know, quit actual development - they either sit on multimillion savings, or went into well, hardly productive sectors like crypto. No one works for hire as developers or even runs development companies (startups are too hard, many tried but no one pulled it off). And these people are barely 40. 20 years ago people like that would be those primary workhorses, now they are lost for the labor market forever.

On paper things look well, but productivity obviously suffers.

You make retiring in your 30's sound like it's a victory. No one does that unless they feel unwelcome and have enough money to scrape by after saying goodbye rather than being forced to tolerate an unwelcoming environment.

I don't agree, many people just don't like work. Smarter people may be less influenced by social pressures/morals so they can see it in a more unbiased way (society forces on us perception of work as a virtue), so they like it even less.

After all, i for example, have never participated in a single project that would ever go live, and only a few times, built small pieces of code that were for some time used in live products. How can i probably fool myself into believing that work is a virtue?

It depends how you define work. I have built systems that are used in large production services. But even if you're just going to dinners and meetings then that's still work in a way. Possibly a more important kind. We all need to have some reason to exist. Tech is weird historically speaking since once the work is done the code runs itself which leaves builders as the architects of their own demise, since few folks are interested in disrupting cash cows with a better mousetrap.

Why would it not be a victory? Who likes working for someone else when you don’t have to. I can think of an infinite number of things to do with free time if I didn’t have to work for money.

Smart and productive people are happiest when they are free to do whatever they feel is worthy of their time and attention. One of the key to their productivity is that they don’t need to be told what to do every step of the way. Retire at 30 does not mean being unproductive, it means working on what you want, when you want. I have even met execs in large orgs that were essentially doing the job because they liked it. They did not need the salary anymore.

And in 20 more years, they would retire, die or become managers. And new engineers enter the field, without millions in the bank, establish themselves as workhorses and so it continues.

This is a bit oversold. Here's the FRED graph of nonfarm labor productivity over the last 80 years or so: https://fred.stlouisfed.org/series/OPHNFB

In fact productivity has been going up at very roughly constant rate ever since it's been measured. And it's been growing since 2008 too. It's true there's an inflection around then, and that the growth rate has been lower than it was during the peak years of the late 90's. But (1) it's not that much lower and (2) it's not remotely outside historical norms. There were flat spots in the 70's and late 50's too.

Basically: the free lunch from the heyday of VLSI scaling, that enabled the enormous automation economy, is probably over. Now we're waiting for another breakthrough.

Exactly. I'm very surprised by this article and all the comments that apparently didn't bother to go check historical data and immediately started looking for causes (they are lazy! It's management! It's inequality etc). But it seems the article is really about the fear of losing no. 1 place to China as can be seen in the conclusion: "the productivity slowdown of the past decade and a half has left the U.S. economy in a weaker position"

Maybe its just people are working too much, leading to lower output due to not being fresh. I just wish we could try different things, a 4 day workweek, or a day off every other week and measure productivity. Where I have a hard time is we have a problem but we are changing nothing. Microsoft Japan has showed that a 4 day workweek improved productivity by 40%, so maybe less is more - https://www.npr.org/2019/11/04/776163853/microsoft-japan-say...

Also given that tech has almost fastfood levels of turnover cannot be great, people constantly leaving and knowledge being lost and the intense time sink of training new hires and interviewing not to mention recruiting and scouting out new people is not helping any productivity. Pay people decently, don't work them into the ground and watch your productivity jump. Very simple.

Even four days a week is too much to be honest though I don’t know if our current societal paradigm could handle anything less than that. With how productive and involved work is now, working white collar jobs are hard. It isn’t like 30-40 years ago when everyone worked at a relaxed pace at relatively low mentally intensive tasks.

> Pay people decently, don't work them into the ground and watch your productivity jump.

That's a question of accounting. Paying people better so that they don't churn is a long-term obligation (opex), whereas recruiting costs are capex - and especially for "cannon fodder" aka junior developers you don't have much in that area anyway. Constant churn keeps wages as low as the market will bear, prevents organizing of employees and guarantees a constant supply of "fresh blood" aka new ideas.

Personally I believe that paying people better is the way to go, the problem is that stock markets reward low-opex operations more than low-capex/long-term sustainable ones.

Surprise, surprise, increasing concentration of market power in tech and other industries seems to explain much of the productivity slowdown:

Computer and electronic products incurred a massive slowdown, with a contribution to MFP growth of 0.45 ppt. from 1997 to 2005 dwindling to 0.10 ppt. from 2005 to 2018....The MFP slowdown in computer and electronic products represents 66 percent of the slowdown in durable manufacturing and 31 percent of the slowdown in the private nonfarm business sector.

...

Mordecai Kurz [...] finds growing market power in the IT sector, which may be stifling the entry and growth of young firms. Kurz reports that “declining or slow growing firms with broadly distributed ownership have been replaced by IT based firms with highly concentrated ownership,” and that “IT innovations enable and accelerate the erection of barriers to entry and once erected, IT facilitates maintenance of restraints on competition.”

Foster, Grim, Haltiwanger, and Wolf also reference the concentration within high-tech industries, noting that, in contrast to the late 1990s, when “the productivity surge in the high-tech sectors [had] a high contribution of increased within-industry covariance between market share and productivity . . . the productivity slowdown in the post-2000 period in high tech is due to both a decrease in within-firm productivity growth but also a decrease in this covariance.”

Titan Alon, David Berger, Robert Dent, and Benjamin Pugsley offer further evidence to support this finding, noting that “over the last three decades, the U.S. business sector has experienced a collapse in the rate of new startups alongside an enormous reallocation of economic activity from entrants and young firms to older incumbents.” Alon et al. clarify that this finding is not just particular to high-tech industries but is “widespread across industries and geographic markets,” so that while this could be relevant in high-tech industries, it could also help explain the productivity slowdowns in other industries. And, more generally, Grullon et al. observe that “more than 75% of U.S. industries have experienced an increase in concentration levels over the last two decades.”

Since the title is a little misleading, here's how the article describes itself.

> Labor productivity—defined as output per labor hour—has grown at a below-average rate since 2005, representing a dramatic reversal of the above-average growth of the late 1990s and early 2000s. The productivity slowdown during these years has left many economic observers wondering why this situation has occurred and what factors may have contributed. To clarify potential sources of the productivity slowdown, this article presents an analysis of labor productivity and its component series—multifactor productivity, contribution of capital intensity, and contribution of labor composition—at both the economy-wide and industry levels, complemented with a survey of the contemporary productivity literature.

My guess would be internet browsing and social media.

bureaucratic costs?

If a small company can "handle" your medical insurance problems in one hour, and a "too big to care" megamerger company takes ten hours, and the economy encourages mergers to profit off the financial transaction, the resulting megacompanies will have everyone's productivity drop. That times hundreds of other business operations ranging from getting a toner cartridge from the supply closet to departmental meetings.

Odd that it describes a slowdown in growth of labor productivity as a "productivity slowdown." As though we should be able to sustain growth infinitely. Seems like the low-hanging technological fruit has been picked, though.

  > As though we should be able to sustain growth infinitely.

isnt that the basic assumption that underpins all of our economics though?

maybe im missing something, but minus that (overall expansion/growth), things start to look more like zero-sum?

You're not missing anything. Yes, this basic assumption underpins all orthodox liberal economics. No, it can't be true.

The recent trend the capital component was a little light, labor was a bit more than usual. Its the 'mfp' that is really lagging.

My speculation is business investment in r&d has vanished, in part because of stock buybacks and in part executive level compensation. Think about what the r&d spending was like back in the 50s-mid 80s and then the initial flipping of the online switch in the late 90s early 00s. The easy gains seem to have been made?

> My speculation is business investment in r&d has vanished, in part because of stock buybacks and in part executive level compensation.

No, it only has shifted. In ye olde days, companies took the risks of R&D spending upon themselves. These days, R&D spending has been shifted towards the capital market (especially venture capital) and in some areas universities, and companies buy startups that have shown results, and the VC investors/taxpayers/university endowments bear the risk.

Not sure most people would consider venture capital funded startups as even close to the same as the large corporate basic R&D of the past. A better 'name' might be: niche short term applied r&d.

Likewise, university "research" is not the wide open expanse that it once was either.

The starting point of this article is misleading. It would be more accurate to say that most Western nations saw historic and extraordinary productivity gains for much of the early 20th Century, building up to the astonishing economic activity of the post war boom, 1945-1973, what the French call "les Trente Glorieuses." Since that time, every Western nation has suffered the Great Stagnation, but the Great Stagnation has been punctuated with several short revivals. France saw productivity boom in the early 1980s, but that boom died out. The USA saw productivity return to the boom levels from 1995 to 2005, but then die out.

This article starts its analysis with 2005, but lacks the broader historical context that might offer deeper explanations for the malaise that it is about.

> Namely, though low productivity growth may be leading to rising inequality, it may also be that rising inequality is reducing productivity growth, by stifling “the ability to harness the talents of potential innovators across the income spectrum.

"The ability to harness talents" is a weird way of saying people don't want spend their time working to benefit someone else. When the value extraction knob turns up, the incentive to work goes down. When you're getting cucumbers and the other guy keeps getting grapes[1] it makes sense to stop playing the game.

1. https://www.youtube.com/watch?v=n_9RjDqJ7Zs

https://en.m.wikipedia.org/wiki/History_of_Facebook

The timing may be totally coincidental, of course…

At the most basic, production is defined as resources_in x efficiency.

American energy consumption stopped growing so all growth has to come from efficiency increases, which has a physical maximum of 1. Energy efficiency of computation has been increasing exponentially - and orders of magnitude increases are still possible before the Landauer's bound is hit. Cooling and heating is close to physical efficiency limits. Transport in theory could be near lossless, but required infrastructure makes that impractical any time soon, and open-air transport is close to efficiency limits.

Therefore, in the absence of an increase in energy consumption significant productivity gains can only happen in sectors where compute/storage/bandwidth expense is a significant factor. This very lopsided growth profile is likely to lead to an increase in social tensions.

The only other sector I can think of where exponential efficiency increases are also possible is biotechnology, and, by extension, medicine.

https://en.m.wikipedia.org/wiki/Energy_in_the_United_States#...

Like profit (measured only as money left after expenses), productivity is not the only important measure. What is traded for productivity? Who gains from productivity increases?

It would certainly appear that over the last 30 years in the US, most of the productivity gains have gone to a relative few people - certainly not uniformly distributed amongst the people who created most of that productivity.

The average worker is not getting more productive, because they are being homogenized to great corporate benefit.

Workers are homogenized when software and hardware automation replace some of their skills, even though the job still exists.

1. Reduced utilization of worker skills, reduces worker's on the job value.

2. Reduced utilization of worker skills, reduces worker replacement costs.

3. Reduced need for training, also reduces worker replacement costs.

4. Reduced need for training, decreases on-the-job value gains for workers.

5. Homogenized jobs are easier to offshore to areas with lower wage expectations.

6. Homogenized workers allow for greater manager-to-worker ratios, either reducing the number of management jobs or increasing the number of homogenized jobs.

More of these jobs are being created very quickly since:

1. They save companies in wages.

2. They save companies in time.

3. They save companies in management complexity.

4. They save on resources for dignity and comfort.

5. They allow for faster scaling.

6. Entering new markets that require many workers is easier and more profitable where job homogenization is possible.

Think of automation from the employer's view:

1. Automation by software.

2. Automation by hardware.

3. Automation by worker homogenization.

Workers (for many types of low-level high-count jobs) and capital goods are "better" when they are standardized in quantity.

The point isn't to make them more productive, but to save money, which may go to profits, but it might also just go to reducing product and service prices since homogenization is not a difficult differentiator to copy.

So no "net gain" in GDP dollar terms, just reduced prices in line with reduced wages. But with dollars shifting away from companies not doing this yet, to companies that are.

This looks like the falling rate of profit (https://en.wikipedia.org/wiki/Tendency_of_the_rate_of_profit...) by a different name.

Many economic models assume economic profit is zero, meaning the owners could not use their time or money better in any other business.

Huge productivity boom after the 2008 recession when everbody had to do the same thing with a ton fewer people, I would expect to see smaller growth as things normalized again over the following few years. That said, not a data scientist - just an armchair economist.

Can't wait to see the numbers "per hour" when they normalize a full time job into 40 hours a week and try to figure out how remote workers perform and also have childcare duties (or work all the time) for two years from your house. It will be quite the challenge for them to normalize!

These statistics are scary. I believe that instead of worrying that "the robots will take all of our jobs" we should worry that the robots aren't taking our jobs quickly enough.

Should robots appear tomorrow and replace 99.95% of the workforce do you really believe the robot owners will be altruistic enough to provide charity to the displaced workforce?

The attractive people who are out of work will end up in the sex industry. The rest will go to protection zones, ghettos, and will be sterilized to prevent reproduction, and possibly starved depending on how things go.

Society is not shock resistant and needs time to grow into things. Look at oligarchies and tell me socialism, capitalism, and communism are really any different beyond the rhetoric.

How many dev shops said they are agile when they were clearly waterfall?

No matter what you call it the ruling class will always call the tune and the poor and middle class will dance.

And who are the 0.1% going to sell the output of their robot labour to if everyone else is in ghettos?

Anyway, I’d hope in such a scenario some smart underdogs would illegally repurpose some robotics to create their own abundance. Might make a good TV show.

Let's take the states for example 99.95 are nonentities so that leaves 330,150,668 *.0005 = 165,075.

That doesn't count the sex workers, security, or workers programming and running the bots. That is still a lot of people for one country, and shopping will change a bit, and they will all pat themselves on the back for saving the planet now that the population is under control.

You will have resistance but in the end the deathbots will win.

That is frankly a deeply stupid sensationalist assumption of "structural unemployeement? straight to genocide!" The main origins I can find for it are outright projection from The Party's leadership who treated their people as tools and property who existed only for them. It showed in many ways from their military doctrines to lack of investment in even their intellectual workers like doctors. Lumpenproletariant were treated far worse than even homeless vagrants in the most callous of capitalists. Not even late Rome with its later high levels of unemployment and unrest was so cruelly stupid. They knew panem et circenses was the cheaper option in spite of the expense and the whinging of patricians about politicians appealing to the masses with quality of life issues. If the marginal cost of goods goes to nill guess what becomes cheaper than dedicated genocide bots which will lead to retaliation? Building more and using the excess to support them. Let alone what sorts of tasks become possible with a large excess, and the nature of veblem goods.

Romans attempted to feed its slaves to lamprey. They also refused automation because it would displace the slaves. They worked slaves to death on occasion. Rome had crucifixion and boiled people in oil for trivial reasons. Human cruelty knows no bounds and as a species we haven't changed much. If you look at modern examples:

https://www.history.com/news/ukrainian-famine-stalin

https://www.dailymail.co.uk/news/article-2017839/Madman-star...

Communism/socialism are sorta different. Look at cuba look at the values that people died for throughout the last century. Humans are imperfect but ideals are ideals, and show me the ideals that capitalism has gotten us but mass exploitation. We have to be more creative than the past and see better alternatives. Mass movements/protests are the only thing that have ever moved the needle https://en.wikipedia.org/wiki/Neozapatismo

>Should robots appear tomorrow and replace 99.95% of the workforce

In what world is that going to happen?

> No matter what you call it the ruling class will always call the tune and the poor and middle class will dance.

Ever heard about the French revolution? Ruling class had to take dance lessons.

Maybe that's the problem. We stopped giving dance lessons to the ruling class a long time ago, but those lessons have always been the only reason why, as working class, we have the few good things we have.

And just like in the classic Animal Farm, the animals threw off their shackles and became the elites who crushed the downtrodden.

Would robots always show up in productivity numbers or are there certain ways robots could be employed that don’t show up?

I was kind of expecting this to be blamed on three things:

1. Supply chain issues

2. Businesses that are still closed/reduced in size

3. People working from home

But instead it's a long-term analysis from far before covid

Why should we be more productive if all the additional value we create goes up? Why should we do charity for the rich?

I wonder how much of this is due to boomers being suckered by their parents to eradicate their own corporate safety net. These people are in their 70s and can't afford to retire. Many of them enjoy seniority for just existing for so long and they would cost a fortune to lay off under most VSO programs. The term "shoot yourself in the foot" referred to someone with a rifle making an error. That generation decided to use the semiautomatic variety.

Reply on news.ycombinator.com