The Jobs Now bill

[jducoeur]

I think most people can agree on one thing: the current political structure has jumped the shark. Not only is it idiotically contentious, it is *completely* devoid of imagination. Both parties are locked into 40-year-old (or older) mindsets about how economics works, and both tend to favor ridiculously oversimplified answers.

The classic example is how you stimulate the economy. The Democrats still favor highly government-centric stimulus. Sure, it's being doled out to private companies, but the government is still making all the decisions about where it should go. There's a place for that, but it should be focused on infrastructure development (the proper place of government), rather than how to create the most jobs -- basically, it's trying to serve two purposes, so doesn't necessarily do either well. It's a recipe for pointless pork.

The Republicans, OTOH, are wedded to the idea that reducing capital-gains taxes is the answer to all economic ills, which looks increasingly ridiculous. It's quite clear that *simple* tax breaks aren't going to help, because business is already plenty profitable this year. It's just that they aren't plowing those profits into domestic jobs, because that isn't obviously the best use for that money. Giving them more simple profits doesn't change that.

Okay: let's accept the political reality. Due to the breakdown in Washington, not much can actually get *done* aside from reducing taxes. The problem at hand is to increase jobs: almost everyone agrees that that is the central sticking-point in the economy. (There are other severe problems, of course, not least the bust housing market, but unemployment seems to be at the heart of the current economic lockup.)

So here's a not-very-modest proposal:

Starting right now, businesses get a 100% tax break on *payroll increases*. That would be measured against one-year-ago numbers, based on payroll reported to the government: essentially, all new hires get their salary compensated by tax breaks.

Employees wouldn't be quite free -- there are still health-insurance and tax expenses -- but it makes hiring very, very cheap. It specifically gives businesses a reason to expand *now*, by reducing the risk of expansion and providing a reason to believe that that expansion will pay off.

This plan should have aggressive sunsetting built into it. Companies would be compensated at 100% for the first year of a payroll expansion, 66% for the second, 33% for the third. (To discourage a quick hire/fire cycle, and give the recovery enough time for a solid foothold.) The plan would be phased out gradually but efficiently: companies would get 100% of the benefit for anyone hired in the first quarter, 90% the second, and so on, dropping to 0 in 2.5 years. (By which point this plan has either succeeded or failed -- there is no point continuing it any longer than that.) Putting those pieces together: the sooner you expand, the more benefit you get, and those benefits pay out over three years if you maintain that new level of payroll, reducing over those three years. The sunsetting is gradual, to avoid the usual problem of a massive and sudden fall once a program ends. (As happened with the new-mortgage support, for example.)

The idea is that this plan is specifically to kick-start the economy, not support it long-term. We currently have an intractable cycle: unemployment is high because businesses aren't hiring; they aren't hiring because they anticipate unemployment staying high so the economy stays in the doldrums. The country needs to break that cycle, by making clear that employment *will* go up, and providing realistic tools to encourage companies to play along. Frankly, it's all about shifting things back towards "greed".

(A basic of market economics is that things are driven by the tension of "greed" and "fear". We've been *way* too far on the "greed" side for a couple of decades; having gotten our comeuppance, we've swung equally too far towards "fear". The trick is pulling things back towards the center, which is where a healthy and stable economy resides.)

What are the problems with this proposal? Obviously, the deficit hawks will scream bloody murder about it, but it's not obvious to me that they have a valid case. The truth is that the deficit doesn't matter, the debt does, and in the medium term this plan looks likely to be good for the debt. It costs a bunch upfront, but a fair chunk of that comes right back to the government in the form of taxes and reduced unemployment payments. In the longer term, the worst *possible* thing for the debt is an extended period of doldrums: government spending entirely aside, the government's income is shattered by the tax losses. So if this succeeds in its goal of improving medium-term employment, it's probably at least a wash in terms of the debt: basically, it's an investment with a ~3-5 year payoff.

The centralizers will hate it because it provides no government control over *what* those new jobs are. IMO, that's a virtue: it doesn't try to dictate the economy. It just recognizes that the goal is new jobs, and focuses laser-like on that.

It does produce some economic distortions -- growing companies wind up with a definite competitive advantage over established ones, since their costs-per-output are lower. That's probably not too bad a side-effect: that's a normal effect of economic creative destruction. But it would be reasonable for large companies to be just a hair nervous about the effect. (OTOH, if it winds up producing economic growth again, they still probably come out in the plus column.)

And it's sure to produce all kinds of creative fraud. I suspect that even a good bill would wind up more complicated than I'd like, simply to legislate against some of the possible scams. Such is life at this level: it's hard to do *anything* at the federal level that doesn't produce opportunities for fraud. That doesn't worry me too much, although it's important to figure out and rule out ways to game the system legally.

Opinions? This is clearly a scary-big proposal -- but it's frankly less dramatic than many of the proposals that both parties routinely trot out, and seems better-targeted to the problem at hand. It seems to me likely to actually *work* decently well; I'd be interested in arguments about why it wouldn't. If you do think it's a good idea, I encourage you to forward it on...

Comments

[rickthefightguy.livejournal.com]

Actually, large banking firms will love this! They can increase the salary of their ceo by 100% for free!

More seriously, yeah - that sounds like a neat idea. Though I would tend to limit it to new hires, not raise it total payroll. And it could be less radical and still very effective by merely making all new hires no subject to the employers' part of the income tax. That's more like only a 15-20% savings, but that is still a lot, and it wouldn't cost the government anything really, and would still benefit the budget with the employee's part of the tax.

[jducoeur]

Yeah, the "just increase salaries" thing is one possible hack. Ideally, it would be good to exclude that; in practice, I'm not sure how to do so without making monitoring significantly more complex. It isn't clear to me whether this is a major problem to be dealt with, or simply an irritating edge case to ignore.

I'm deliberately aiming for simplicity here, rather than the usual overcomplicated Washington mess that turns out to be worse than the problem it's trying to solve. The cost of that is that there are certainly rules-hacks possible. The question is, which ones are bad enough that they *must* be addressed? Some, certainly, but I'm not sure offhand which. (This would, no doubt, be much of the legislative argument, as it is for any new law.)

Similarly, the reason for suggesting total payroll as the metric is that it is very objective and easy to measure. Limiting to "new hires" introduces possible risks of companies churning employees, which is a real danger: illusory progress that doesn't really help. It might work, but I'm not sure I understand the process as well.

And yes, it's possible that a less-radical solution might suffice. I'm deliberately putting a dramatic stake in the ground, which I'm pretty certain would work. I'm open to arguments for something milder, but the key is to avoid half-measures that are too wimpy to make a real difference. If it doesn't really move the unemployment needle significantly (1% at a minimum), then it doesn't produce the psychological feedback effects that are at the heart of the solution. (Also, it's easier to argue for a less-dramatic solution to become permanent, which probably isn't a great idea in the long term. This is all about turning around the current downturn, and then returning to normalcy.)

[kiya]

I have strong deficit-hawk tendencies, though they should probably be called debt-hawk tendencies, and I think it's a clever notion that nobody will ever be able to get the votes for because it doesn't fit the gods-be-feathered template.

(Mind, I'm the sort of wacko who thinks that genuine universal single-payer health care would be a godsend to small businesses, which would thereby no longer be restricted in hiring by being required to offer competetive health care plans and hobbled by the adminstrative and other costs of health care provision .... I actually have anecdotal evidence that this is a real serious problem, and I'm pretty sure real numbers back it up at least at some level, but haven't done the research.)

[jducoeur]

Oh, sure. I'm firmly of the opinion that our current health-care system is nearly the worst possible setup, and one of the main reasons is that it greatly harms labor flexibility. Tying healthcare to jobs provides a powerful disincentive for people to change jobs, and that's idiotic: job flexibility is one of the US' greatest economic advantages, and it's just plain dumb to impede it.

Single-payer has its own problems, of course. But overall, I don't think I've yet heard a suggestion that I didn't think would be *better* than the currently entrenched model.

(All of which is largely unrelated to the point at hand. But it's a Big Red Button for me.)

[mindways]

Payroll expansion seems like an overly simple metric. There's the "CEO raise" problem mentioned above. Plus, paying existing employees to work more hours is already cheaper than making new hires, in many fields, due to fixed per-employee costs and the expense of the hiring process. With this credit, I think that'd become true even for jobs paying time-and-a-half for overtime or such, meaning you'd encourage wringing more out of existing workers more than hiring new ones.

It would also create an immense incentive for companies to shunt as much of their balance sheets onto payroll as humanly possible - to the extent that this spurs them to hire people for permanent positions this is good, but it also seems like it would result in many accounting shenanigans, as well as hiring people for ludicrously inefficient jobs which would go away in 1, 2, or 3 years. (Which would admittedly serve some purpose towards the declared end.)

[It deliberately ignores the "what about companies that were already planning on hiring?" question, though, which I regard as a plus - that metric is only relevant for talking about "new" jobs created, which seems more important as a political talking-point than an actual metric.]

---

Notional modification: Give a tax credit equal to N% of a new hire's salary, amortized over 10 years - but ceasing if/when the new hire stops working there. This requires more bureaucracy, but doesn't subsidize raises / the forcing of existing workers to take more hours, nor does it provide incentive to artificially shove numbers into the "payroll" category". Define "salary" as W-2 income and it's information already going to the IRS. You can alter the amortization curve from a flat line to provide more up-front benefit if desired, for a sharper "kick" and a gentler letdown (though the "ending when an employee leaves" clause will provide some measure of natural letdown), though this makes the paperwork more complex.

This still leaves the question of what to do about hourly hires without a constant salary, though. (Plus a number of edge cases for salaried employees, but I suspect the hourly workers are a thornier - and larger-scale - problem.)

---

I suspect that any program of this scale would run into some resistance because of a general societal perception that any employment it generates isn't quite "real". Which is true, on one level - it's artificially incentivized - but that can still serve the purpose of pump-priming.

(Not that I claim it necessarily would. I'm becoming more and more a hardline economic agnosticist: the more certain someone is that they've got all the answers, the more likely they're full of it. Unlike some, I do not take this to mean we should throw our hands up and go 100% laissez-faire because that's the best we can hope for - simply that we should acknowledge that our understanding of economics is about on par with various ancient alchemists' grasp of chemistry as we go about our attempts at policymaking. Hopefully it will improve with time.)

[jducoeur]

Yaas. I certainly recognize that I'm proposing a blunt instrument here -- but that's mostly because it's simple enough for the feedback loops to be *decently* predictable.

Your revised version is certainly a plausible tweak; my only concern, as mentioned upthread, is whether there are any job-thrashing possibilities. (Might be, might not; this is more a back-of-the-mind tickle of a concern rather than a serious argument.) I do think the amortization scheme would need to be carefully balanced, to provide strong incentives to hire *now* (which argues for front-loading) but keep people on payroll (which argues for back-loading). But it might well be made to work...

[be-well-lowell.livejournal.com]

I'm confused. I thought that as business expenses, salaries were already exempt from corporate taxes.

[jducoeur]

I'm not talking about exemption: I'm talking about an active kickback -- actually deducting those salaries from the *rest* of your corporate taxes. Essentially, this means that the government pays the salary of the new hire for the first year. That's pricey, but economically healthier than the government hiring people directly (since it allows the market to dictate where those jobs go), or paying unemployment...

[be-well-lowell.livejournal.com]

Okay, that makes sense. Thanks.

[metageek.livejournal.com]

I think the term for that is a tax credit, not a tax deduction. A deduction is subtracted from your gross income; a credit is subtracted from your tax.

[jducoeur]

Okay, duly noted for the second draft that I may well put together...

[cellio]

An interesting idea requiring more thought. One modification I would make is to apply it only to hires (or raises) for US citizens and maybe permanent residents. My reasoning is twofold. First, this is in part a benefit for the employee (who gets the job) as well as to the employer, and tax-funded benefits should IMO be limited to citizens (and those on the citizenship track, probably). Second, it ensures that the taxes the employee will pay on that income will be paid to the US, which recovers a portion of the cost (or maybe all of it depending on duration of employment and benefit phase-out periods). If you increase your payroll by setting up operations in Somewherestan, no tax credit for you.

[jducoeur]

I actually care somewhat less about this (IMO, the illegal-immigration argument tends to be overstated). But it likely makes the bill more politically palatable (indeed, it probably brings some of the current nativist mob on-side), and probably doesn't hurt the idea much economically, so I'm not terribly opposed to this amendment...

[cellio]

I wasn't thinking of illegal immigration. I was thinking of outsourcing new jobs overseas for free, with no payback to the taxpayers who funded it and no reduction in unemployment costs.

Besides, as a practical matter, do employers who hire illegals report that? :-)

[jducoeur]

Fair point -- it's true that this wants to be limited to domestic payroll. I assume that that's already tracked for tax purposes...

[metageek.livejournal.com]

A more limited version which would resist some of the obvious hacks: for the next N years, the government pays the employer's half of FICA.

It wouldn't drive churn, because it applies to old employees, too.

It wouldn't drive pay raises, because the employer doesn't get any more money that way. Plus, the benefit is the same whether someone is right at the FICA cap (currently $106,800) or making $10M.

And it wouldn't require a lot of expensive paperwork, because it would just mean eliminating a line item on the paperwork that already gets filed.

It would have to be fully funded (or gut Social Security), and it would have to be sunsetted.

[jducoeur]

Problem is, this doesn't do much about the problem I'm trying to address, which is very specifically *hiring*.

Keep in mind, companies have *plenty* of money right now; there is no need to improve their bottom line. The problem is, they're nervous about the state of the economy, so instead of adding employees, they are simply making the existing ones do more, and hoarding the profits. And since everyone is hoarding instead of improving employment, the economy stays bad.

So my proposal is precisely targeted as a bribe to hire *more* employees, right now. That produces a virtuous economic cycle, breaking the back of the nervousness by convincing *everybody*, all at once, that a corner is about to be turned, and that if you don't jump on board now, you're going to lose a march on your competitors. (And that the risk of jumping on board is so low that you might as well do so.)

It's all about feedback loops. (Economics is all about system dynamics.) Targeting *all* employees doesn't change the status quo; providing an extremely strong bribe solely and specifically for new ones creates an extremely strong loop, which should be enough to become self-sustaining pretty fast...

[metageek.livejournal.com]

OK, I can see that.

Really, though, I'd be nervous taking a job where the company wouldn't have hired me if not for the tax credit—will my job last past the sunset?

[jducoeur]

Seriously, I think that falls into the category of problems that most people would love to have. At this point, the majority of the folks who are unemployed are having trouble finding *anything*, so a job that might only last a year or two is still a big improvement.

And the whole point of the exercise, like I said, is feedback loops. If this drives unemployment down to a more-reasonable 7%, that makes business a lot less nervous, which means that a lot of them are going to keep more employees on. And the gradual sunsetting is designed to encourage that: there isn't a *sudden* increase in the cost of those employees, so there isn't a clear "okay, time to lay you all off". Boiling the frog can be used for virtuous purposes, if done right...

[metageek.livejournal.com]

Point taken.

[metageek.livejournal.com]

Employees wouldn't be quite free -- there are still health-insurance and tax expenses -- but it makes hiring very, very cheap.

Maybe not that cheap. I remember being told that non-salary expenses added about 60% to the cost of an employee. So a tax credit for the salary would cut the cost of hiring by 62.5%. You can hire almost three times as many people for the same price.

And that 60% may have gone up; that was over 10 years ago, and health insurance costs have gone up a lot since then. (On the other hand, I think that was in Silicon Valley; office space costs were insane.)

[jducoeur]

Oh, sure. It's not free -- indeed, it's quite important that it *not* be free. (If the government *totally* subsidized the job, a host of nasty abuses become too tempting to avoid.)

But it's still a pretty dramatic subsidy, enough that a lot of businesses will take it up. And it doesn't take a huge amount to succeed: my suspicion is that shifting the unemployment rate down by just 1% of total employment would be enough to kick the feedback loop into gear, and 3% certainly would.

So I'm pretty comfortable that, even if this just reduces the cost of hiring by 50%, that is still "very cheap" enough to work...