Every $1 spent on food stamps resulted in $1.73 in GDP growth
economy.com
economy.com
The theoretical explanation for this is very simple: in the current economic climate, tax reduction tend to be used simply to pay down debts - what Richard Koo calls a balance sheet recession.
Still, it's nice to have that so clear from a source that is quite unsuspicious of overly left/liberal leanings.
Disregarding that particular issue: Sure, banks are always looking for people to loan money to, because that's how they make a profit.
But what if they simply don't find many people they can give loans to? That is still the problem, and that is an important reason why direct spending has a higher multiplier than tax cuts. People are still repairing their balance sheets, which means there are simply fewer people who are creditworthy and willing to borrow money.
For one example among many, see this link: http://www.ritholtz.com/blog/2010/09/zandi/
More seriously, the link is just a summary of what Zandi's model predicted the recent stimulus would do. It's worth noting that:
1) Pretty much every other economist on the left and right disagree with him. On the right they think all his multipliers are too high, and even on the left they think the way he ranked them is bizarre. He reckons one of the best tax cuts is a lump sum refund - despite copious theoretical and empirical backing for the proposition that it's actually one of the worst (because it's most likely to trigger Ricardian Equivalency).
2) The actual results of the stimulus are starting to become clear, and they disagree with him too, both in terms of overall magnitude and, in particular, his completely batshit rankings. He thinks aid to state governments is one of the worst ways of spending cash, but infrastructure spending is quite good. In reality the aid to state governments was one of the most effective line items, while the infrastructure spending was a black hole in terms of job creation.
My advice is to go hunt up an actual respected economist (Romer or Barro are both great on this area) and read what they have to say. Zandi is neither respected nor, at the risk of being cruel, an economist.
Also, the mere fact that food stamp recipients are prevalent and increasing signals a failure in our system.
Not saying we should get rid of food stamps, they're a necessary safety net, but that figure is misleading. How much should the government spend to increase GDP?
Is it valuable to have a high GDP and a high federal deficit? Or will that just end up in necessarily higher taxes down the road (so we don't end up like Greece) that will wipe out that GDP growth?
It's a sticky issue that can't be boiled down to a simple statement.
$1.73 in GDP growth != $1.73 in revenue to the government.
Why? Money is created by federal government and is therefore worthless to the federal government. It can just create more of it. It just makes no sense to look at the federal government (or any monetarily sovereign government for that matter) in terms of how much money they make.
On the other hand, if government revenues are increased by less than 1$, this means that someone, somewhere in the economy now has more net financial assets because of that.
Is that a bad thing? In some cases perhaps, but certainly not in general.
Is it valuable to have a high GDP and a high federal deficit? Or will that just end up in necessarily higher taxes down the road (so we don't end up like Greece) that will wipe out that GDP growth?
History tells us that, for monetarily sovereign governments, their debt is not paid back. Therefore, the answer to your second question seems to be a very clear "No, it will not lead to higher taxes, but it will lead to better economic development right now, which likely also improves the growth path in the long run."
That is not what the research is claiming.
However, you can't discount that an increase in overall GDP leads to increased government revenues.
Even though the US federal deficit is high, it certainly is not a problem like now like it is for Greece. US treasury bond rates are a pretty good indicator of risk, and right now they're very low.
But how can we know that the net effect is positive? How can we know that, for instance, giving people money doesn't prevent them from working, thereby decreasing GDP?
After all, there is no control group in economics, which is why (frustratingly) there are still totally opposite camps who keep claiming that current events prove them right.
I'm not familiar with the methods used, but the article claims exactly that, doesn't it? By spending money on food stamps, GDP is increased.
Mind you, there might be types of spending with an even higher multiplier. And there might be types of spending that would be seen to be more beneficial for society even if their impact on GDP were smaller; for example: Why not just spend money to create jobs? Then spending on food stamps reduces automatically, as the number of eligible recipients decreases.
After all, there is no control group in economics, which is why (frustratingly) there are still totally opposite camps who keep claiming that current events prove them right.
This is indeed very unfortunate. Perhaps somebody clever enough with enough stamina could work on increasingly realistic simulations on the micro level and see which macro patterns emerge? Though of course then there would be an endless debate about how realistic the modeling of micro behavior is.
The current story told in mainstream textbooks (utility maximization etc.) certainly isn't very credible, given all the psychological insights from behavioral economics; and just the mere fact that marketing exists and how it operates tells you that the classical micro story is rather fishy.
Every dollar in food stamps comes from a dollar of taxation. Their study shows that the multiplier of the original dollar is -2.5. If we have 1.8 multiplier on the food stamp, then additively we have -0.7.
Alternatively, every dollar which is 'spent' by the government in the form of tax cuts has a multiplier of nearly 3.
For example, http://www.federalreserve.gov/econresdata/scf/scf_2007.htm
I sometimes wonder what policy would look like if the dominant economic priority was the increase total wealth, rather than economic activity.
Or if everyone has 10 billion dollars in gold in their vaults in aggregate, if the price of gold rises, on paper their wealth would increase. But if everyone actually tried to expend their gold then the price of gold would decrease, but much more production would happen in the economy. If the price of tech stocks octuples then paper wealth increases but GDP does not necessarily change.
Or, if everyone just spent their money twice as quickly (e.g. the velocity of money increases). The economy would be able to do twice as much, but total 'wealth' need not necessarily change.
'wealth' is entangled with money and prices. The amount of production is affected by those things probably but not as directly.
I'm trying to figure out what government policy might look like if it was directed toward increasing net household wealth. Just hoarding gold does little for the economy. But this is irrelevant. At issue is what it would look like, economically, to focus on increasing a nation's economic wealth (after inflation). It seems like it would be awfully difficult to increase after-inflation wealth without a fair amount of genuine value being created, no?
Could focusing on combined household wealth cause there to be more attention paid to median household wealth? And how might economic policy behave differently if increasing this was its focus?
Another reason to pay closer attention to wealth is that it doesn't suffer from the "broken window fallacy".
Government spending isn't a drain. It is another piston in the financial engine. It also happens to be one of the most reliable and most pliable pistons in the engine.
You can think of the government as Herouku. Most of the time you're going to want to run as few instances as possible. When you make the front page of hackernews, you either need to jack up the instances or your site crashes.
We're drawing our lines in the wrong places. Some government spending is good. In some cases, massive government spending is good. A lot of the time the private markets do a better job. Sometimes they don't.
Paying food stamps is a great way to reduce crime and bolster the stability of your country. If you ignore the root causes you get long term issues. We need to fix the underlying problem and keep paying for food stamps. You can do both at the same time.
This lack of consensus in the empirical findings indicates the inherent difficulties with measuring such correlations [between government spending and GDP] in a complex economy.
And most fiscal conservatives have better things to conserve than food stamps. Let's talk about where the money really goes. Oh, wait, white elephant in the room.
http://online.wsj.com/article/SB1000142405274870447150457444...
As I've quoted elsewhere, this is from the study by Romer and Romer. http://elsa.berkeley.edu/~cromer/RomerDraft307.pdf http://elsa.berkeley.edu/~cromer/RomerDraft307.pdf
You (mostly) had me up to the first point.
How are food stamps a positive / negative on crime?
I think this hypothesis has sociological and psychological merit, and I'm sure its been tested before, though I don't know of any studies off-hand.
Obviously that's conjecture, and all poor people aren't criminals (and all criminals are not poor), but it's not a leap of logic to say poverty can lead normally straight people to mentally justify committing a crime. Welfare programs can and do help people get back on their feet.
Living in Michigan, almost all of my family became unemployed as a result of the factories closing down. Some took the assistance to get retraining and find another job, and some took maintaining the assistance as their new job or as a retirement package. Some just weren't able to get back into the job market no matter how they tried. I got out of my state's "No Worker Left Behind" retraining assistance just before the new governor shut down the incredibly successful program.
It's a combination of willingness to work and ability to work. Welfare programs can do no more than take basic survival off your "worries" list.
Zandi may not have left/liberal leanings, but he certainly has government leanings.
Magically it would generate a 'profit' of $730 billion per printed trillion per year for the general economy.
Except the stimulus programs, from the first that Bush fired off, to the Fed's QE programs, to the trillion dollar fraud Obama made shovel ready (har har) --- all have one thing in common: they've all failed to improve the US economy in any meaningful way.
Work force participation rate? hyper implosion. Government dependency? skyrocketing. Real wages? declining. Inflation? pain at the pump and grocery store. Trade deficit? Near all time record highs. Savings rate? miniscule. Household wealth? Less than 15 years ago. Standard of living? Hasn't moved in in 50 years. Debt? On the moon and rising faster by the day. Entitlements? $100 trillion unfunded disaster. Stock market? Hasn't moved in 15 years inflation adjusted.
and on and on
So, yes, it is probably the case that $1 trillion in food stamps would not generate $1.73 trillion in economic activity. But that's not really a counterargument about whether the current spending has a 1.73x multiplier.
You also can't print magic dollars to continue to fake economic results without suffering consequences that completely wipe out the supposed gains that are being claimed.
For example, even just 6% real inflation times ten years becomes a disaster for a middle class whose real wages are flat lined. Every dollar the Fed shovels into the system, becomes inflation somewhere.
Food stamps = pure inflation in a government system running a $1.3 trillion deficit. The Fed has to monetize it all. Whatever you're supposedly gaining through one program, is completely wiped out by the inflation.
Also: in terms of pure investment (not considering the moral good of helping the poor nor the moral evil of enabling freeloaders), even if the stated GDP growth is accurate, there is always the opportunity cost of not doing something else with that money. How much would it increase GDP to build more roads or fund more research?
I'm not arguing against food stamps here, I'm just trying to put this in perspective. And perhaps the answer to my questions is "nobody knows."
You want to spend $50 billion a year on food stamps? Ok, slash it out of the military, social security, medicare, the NSA / FBI / CIA / HSA / TSA / whatever. You want universal healthcare, ok, disband the US military, or slash social security in half. Something will have to go, one way or another, sooner rather than later; the piper will get paid.
With a $1.3 trillion deficit, it's irrelevant if you argue that the food stamps are paid for by taxes, or if you argue the military is paid for by taxes, or if you argue social security is paid for with taxes. Because bottom line: $1.3 trillion per year is not paid for with taxes. Something is not getting paid for with taxes, it might as well be food stamps as any other program currently.
As another simple example of the inflation spread - if the Fed holds interest rates down, which is part of what the QE programs do, it generates inflation. When you get a mortgage adjustment at 4% courtesy of that Fed program, it is generating inflation by the difference in money you now have versus what you otherwise would have.
If the Fed props up property values with the QE and trillion dollar mortgage purchase programs, that causes inflation as home owners acquire value they otherwise wouldn't have.
Any artificial value, created by the Fed, and shot into the system, must inherently become inflation. Your dollar loses value accordingly. And since 1967 or so, the dollar has lost about 86% of its value.
In particular, nobody who studies the matter argues that all government spending automatically produces inflation; not even Milton Friedman took that line (and empirical data doesn't support it).
I think that you'll find that this model is simple to understand, has far less complexity or uncertainty than you may be used to and can be a lot of fun, too.
You can't invent facts, merely discover them. But you already knew that. /pedantic
EDIT:
Also, I don't think he asserted government spending causes inflation. I think he asserts that government deficit spending causes inflation. Minor difference.
Which is why America is collapsing into a hovel of debt, extreme spending, deficits, and poverty: modern economic policy is completely wrong. Such has been the case since 'modern' economics began ruling America in the late 1960s. The US standard of living hasn't improved since the early 1960s, when America was the dominant manufacturing power; it has been downhill since then in real terms, with brief respites courtesy of technology. 10 year mortgages became 30; a high savings rate became a negative real savings rate; the middle class became the impoverished class.
When you run inflation far beyond the rate of real wage increases, the working class has no incentive to save junk dollars that lose value 24/7. So instead of saving, they spend, knowing full well their dollar will buy more today than tomorrow.
Finally, U.S. income in real terms has increased substantially since the 1960s; there is no credible measure of inflation, not even some of the alternate ones, by which U.S. real GDP per capita today is lower than in, say, 1965. What has failed to increase, or even declined somewhat, is real median income; real mean income has not similarly stagnated or declined. That's a problem of income/wealth distribution, not inflation.
Calculated using the 1980 CPI, inflation is off the charts at about 9% right now. When it was last that high, Volcker was taking extreme actions to crush inflation with high interest rates.
In 1971 when inflation was as high as the Fed's current bogus CPI, Nixon installed price and wage controls.
The Fed CPI intentionally understates inflation by leaving out food and energy prices, which are prime inflation meters. So they get to hide the radical increase in the cost of a gallon of gasoline, heating oil, and food, over the last decade.
The CPI was adjusted during the Clinton Administration to hide the real rate of destruction.
US income in real terms has not increased at all since about 1965. The US Dollar, according to the Federal Reserve's own numbers, has declined by about 86% since then.
So no, real wages haven't increased at all when you have to calculate off of that decline.
It's trivially easy to prove it: run wages against the cost of a new car, the price of oil, the average price of a home, the price of gold, the price of silver, the price of a gallon of gasoline, the price of a gallon of milk, and so on. You'll see the same blatant trend line. How much did those things cost in 1965, and what was the average wage? For your own benefit, take a few minutes and run those numbers; you'll see that the US standard of living has gone nowhere.
The CPI includes both food and energy prices. There is a separate measure called "core CPI" that excludes those, but it's not the one normally quoted, nor the one usually used to correct for real dollars.
You just explained everything wrong with our economy. Don't forget, we also have two parents at work while standard of living has remained the same. If I could sum up with whats wrong with our country right now, that quote would be it.
On the other hand, a comment on top about the true voodoo economics is up and running well...
I think most people want to keep milking the collapsing system for as long as they can, for they're overly invested into that system's survival. Anything else literally scares them. It's why the politicians spout the same drivel year after year, right or left. They're just feuding over how to divide up the last crumbs.
The crumbs are running out.
mine is username + gmail.
Our government is broke and wasting money left and right. The spending spree is not sustainable.
Take a good look at Greece. That is the future of America if we keep spending money this way.
Care to go a bit further into this conjecture?
How is spending more than your revenue a long term plan? The only legit reason would be in a time of war or major disaster. Our entitlement spending and union retirements are a mirror image of Greece but on a much larger scale. We owe more per capita than Greece (don't forget to include unfunded liabilities).
Sure, we can devalue the dollar but too much inflation will be a very bad thing - just ask Zimbabwe.
I'm all for helping the poor and believe in reasonable safety nets, but the problem is they are being chronically abused in this country. I live in NYC and see the abuse first hand every day, in the checkout line seeing able bodied people using foodstamps (EBT cards). Come on man, really?
My point is food stamps and entitlements are definitely not an economic tool.
Economist Mark Zandi (the author) missed predicting the biggest economic event not just of his career, but of three generations before him. Mostly, that was due to conflict of interest: Moody's was busy scooping up money to rate mortgage securities. What fraction of his oversight is due to just being a poor economist is unknown.
There are widows and orphans living poorer because of this clown. Don't waste your time.
Secondly, why be surprised that giving money to those who badly need it would increase spending?
And thirdly that each spent dollar would ripple through the economy for a total value greater than a dollar? This is one of the earliest lessons in Economics 101.
What this shows is that food stamps are a relatively "frictionless" method of boosting spending. There's less overhead for food stamps than for other programs and, since the spending is unlikely to be misdirected (as in "bridge to nowhere" projects et al), you get lots of bang for your buck.