Hacker News Reader

I run payroll, and everytime I do it's a potential mistake waiting to be made. We use Gusto, which is about as good as it gets for online / easy payroll, and there are still occasional issues. Payroll doesn't just pay an employee, it also handles proper withholding for each employee with an ever changing system that is tied to a huge swath of federal/state/city tax rules, which includes how often you have to pay the locality, usually based on how big payroll was in that locality. Basically, doing this everyday, having it trickle in, seems like a nightmare. I think for it to work properly the end to end system needs to be reworked, and I don't think the potential benefits outweigh the risk and work required.

"If it hurts, do it more often"

That works when the pain is something you can fix, and the cost of experimenting isn't too high. I'm not sure government bureaucracy meets that description, and experimenting can become very expensive if you screw up and the Eye of Sauron turns upon you and starts auditing.

I love the spirit of this, but there are so many industries where it would be nearly impossible to pay daily or even weekly. The administrative overhead would be huge and costs to customers would go up to accommodate. I'm thinking about my own past experiences, where trying to automate away (the only solution here to more frequent billing + payroll) would be tremendously challenging and industry / company specific.

In code you write, not in real deployments where it can cost you millions.

The cost/benefit doesn't pan out. We're not talking about exercise here, we're talking about regulatory liabilities for any mistakes being very potentially costly.

Preparing city, county, state and national governments for daily automated payments seems like a good first step. That and creating guidance for companies.

I can see it being a pro-labor perk for those companies or industries that chose to adopt this type of scheme without being forced by legislation. That stated, it overlooks the number of capital-intensive industries and small businesses that simply cannot afford to pay on a more regular basis than every two weeks. In fact, I'd go so far as to say that legislation of this would be well-meaning, but would only serve to harm a lot of small /early-stage businesses.

Payroll isnt quite that simple though. It takes manpower to make sure folks clocked in and out correctly, for instance. There are automatic deductions beyond taxes - everything from health insurance to pensions to work shoes (seriously). Sometimes child support or other legal things. So many complications.

Heck, banks won't even let you have the money immediately when it is on the weekend or a holiday.

There are real costs to processing payroll more often, even though checks are not printed as often.

> However one thing that's been on my mind lately has been the lag between work and getting paid for that work. Even for regular employees.

Regular employees already benefit from legislation on this point (which often covers both minimum frequency and maximum delay from end of pay period or termination to delivering pay.) [0] So, with regular employees, it's just a matter of tweaking regs to better fit current conditions.

With gig economy contractors, the situation is less simple.

[0] A description of the CA state rules: https://www.dir.ca.gov/dlse/FAQ_Paydays.htm

So are you saying you'll be increasing the frequency of ACH transactions by 14x (e.g. daily vs. biweekly) or 30x ?

There is likely a cost with that, even if the bank allows the frequency change for free. Most banks are starting to provide APIs, but many still charge fees for large volumes of ACH formatted files for example. The costs in making sure things are tied out (ie, all amounts match to the penny) seem pretty high.

Every payroll cycle is difficult because you have things like [benefit|FSA|ESPP] deductions, and the occasional bonus earning or 0-dollar mandatory RSU vesting (to account for taxes paid).

The benefit for a daily payment is pretty low compared to the costs.

>The costs in making sure things are tied out (ie, all amounts match to the penny) seem pretty high.

Could you elaborate on why it seems high? What you're describing is computations on distributed systems, and I'd be surprised if it's computation or communication that is the bottleneck for handling ACH transactions.

Manual effort for handling discrepancies for each payroll run. You might be surprised but lots of companies' payroll processes are not automated to avoid manual processing at some point in the process.

> The benefit for a daily payment is pretty low compared to the costs.

Agreed. I can't imagine having to balance my checking account with a daily transaction. It would be much harder to catch an error or a one-off when you didn't get paid too. Is my mortgage going to start deducting daily too? Both those things seem insane.

Also you are limited to 6 ACH transactions a month to a savings or money market account.

> Also you are limited to 6 ACH transactions a month to a savings or money market account.

You're limited to 6 withdrawals, you can make as many deposits as you want.

Until you deal with unions, tax rates across different states and countries, corrections, benefits, inaccurate time recording, etc.... Payroll isn't that easy.

The opposition you would get there is that makes it sound too much like actually getting paid for time worked, i.e., hourly wages. It'd be much harder then to justify the "salaried and exempt from overtime" dance that most companies do for employees.

This is a great comment. Imagine pulling a 15 hour shift and getting 200 bucks, just like your 8 hour days. It would be infuriating and nobody would put up with it. Yet they do right now. This is another great reason to switch to a daily pay system.

You have it exactly backwards unfortunately. There being such a short feedback between working a long shift and getting a relatively bigger chunk of change would incentivize people to do it more frequently if their employer were willing to pay for it.

> if their employer were willing to pay for it.

Behringer is starting from the supposition the employer wouldnt. And thus it would cause unrest.

Yeah, parent was discussing from the point of view of reality, where employees are not willing to pay for it (the opposite, they'd like to milk their employees even more).

On the other hand, if I am sick with the flu, getting 200 bucks for just sitting at home with chicken noodle soup is a good feeling.

In the US corporate world how often are people "sick with the flu" versus forced to work overtime?

In my experience it's fairly common to not have to work much overtime even on a salary. It's just a few certain types of industries and certain types of positions where this is heavily abused.

I worked in the gaming industry in canada for a brief stint and there was a period where we had mandatory team overtime of 4-5 hours M/W/F for a few weeks.

The only saving grace of going through that was that team productivity tanked so badly that the company ended up losing progress toward the milestone and payed out of that contract. After that experience dev estimates carried more weight and the company was more cautious about tight deadlines.

There is an interesting, well-funded startup tackling this problem: https://www.earnin.com/

I am not affiliated with this company, just like their idea a lot for all the reasons mentioned in your post. Obviously they have to make money and presumably take some cut or added fee for providing the service of getting you paid before your employer does it, but at least they shorten the time between work done and getting paid.

I get paid twice a month and I wouldn't care if I got paid only once a month considering all my bills are monthly or longer. Credit cards, rent, and internet are monthly. Utilities are monthly or biannually. I can't think of any other bill than that.

I agree that for some, once one has enough cash sitting around it just becomes a sea of cash that barely changes order of magnitude.

However, for many of America's working class they are < 1 paycheck from bankruptcy. I speculate that for many of them getting the "liquidity" of getting paid daily would stop some of them from missing payments.

The one piece that I had a hard time understanding, is there is a range of income for the group of people affected. Some make $10 an hour, others make $15, yet they are both in the same boat. If the person making $15 an hour lived like they were making $10 then in 2 units of time they have 1 unit of time of expenses saved (i.e, in 2 months, they have an extra 1 month of their expenses saved).

But then I came to the realization that real life doesn't work that way, even looking at my (lack of) savings through the years. That is because life is full of stress, and throwing money at the causes of stress is sometimes the only way to reduce that stress. For example, having a teenager that has a drug issue, if you don't have the extra money then you deal with it, but if you do have extra you can spend it on therapy which may help thereby reducing stress.

And the various stress items really don't go away until you are bringing in about 70K or so a year -- at that point you can finally afford to pay off various debts, build an emergency fund, drive a reliable car, and live in a slightly better neighborhood.

Adding to your comment, also comparing $10 to $15 is really hard because that could be $10 where rent is $500 or $15 where rent is $2000. Because price levels vary so much nationally we need a way to level those out. It makes little sense to compare people who have fundamentally different purchasing power. Also, IMO, this applies to taxes (why should someone "just making it by" in the bay area, but makes $100k be taxed like someone who pays 50% less for everything they buy in another region of the country)

I'm glad that you're covered. But have you ever driven over to the seedier part of town and seen most of the businesses revolve around money lending? "Money Tree", "Payday Loans", etc.

We're talking about a lucrative industry that exists around extracting money from the poorest people. And most of it has to do with them waiting to be paid for work they've already done.

And most of it has to do with them waiting to be paid for work they've already done.

Is it really though?

If someone is really in this poor of a financial situation, I'd bet that it could actually be worse to receive money day by day. This gives them access to use it prior to when larger bills are typically due: the start or end of the month.

That is, if you get paid on the 31st, you'll have that full paycheck available to pay your loan, mortgage/rent, car, phone service, and so on on that day or within the next few days. Then you have the rest leftover to blow.

Day by day payments involve control and making sure you are setting aside enough.

Have you ever noticed how excited people in this situation are for "payday", since they're able to go buy stuff? Now, every day is payday with the associated temptations.

Have you considered that it's not about when they get paid but how much they get paid, and how much money they are spending each month?

If you get $300 on the 1st or you get $10 a day you'll have the same amount on the 30th. If that amount is less than your monthly expenses you're in trouble and payday loans start to look attractive.

A lot of people in the seedier parts of time are working jobs that don't pay a livable wage, and a lot of people are also spending more than what they should be. Sometimes irresponsibly, sometimes out of necessity (see: the boot theory of economics).

Paying irresponsible people more often will exacerbate their problems.

Paying poor people more often won't make them not poor any more.

Considering the time value of your money-

If you get a $1500 paycheck every 15 days, as opposed to a $3000 paycheck every 30 days, then you earn 15 days of interest on $1500 that you would not otherwise earn. At Ally bank, that could be 1.85% in savings accounts. This two-paycheck cycle would repeat 12 times a year, so for half the year you earn interest you would not otherwise earn, totaling to roughly (compounding negligible) 1500*.0185/2= $14.

If you assumed you could get 9% in the stock market, the time value would be $68.

Depends on the company. I'm paid on the 15th of the month for work done that month, so that works out fine.

That $68 earning is either going to the employee or the employer. If law forces it to go to the employees then the employers could try to hire people for $68 less salary.

> = $14

.. per year. Or $1.16/mo or under .04%. Although that's not literally nothing, it's awfully close. Oh, and that's before income tax.

> If you assumed you could get 9% in the stock market, the time value would be $68.

Although that assumption may be a bit of a stretch, something a typical consumer might be far more likely to encounter (and need not apply income tax adjustments to) is consumer loan payments.

A 20% revolving balance paid 15 days early results in $12.50 monthly, which can pay for a streaming subscription.