Saturday, June 1, 2019

Basic Income Pitch Botched by Bad Research

If you want to make the case for a big, new entitlement and the enormous taxes that would be needed to pay for it, then at least make sure you do your homework. It should not be this easy to pick apart the "analysis" produced by a Syracuse professor in defense of universal basic income.

Here we go again... Another push for universal basic income. Hasn't this idea suffered enough? Must its proponents continue to kick this dead horse around? 

The latest contribution comes from the Tax Policy Center. A Syracuse University professor named Leonard E Burman wants to expand the Earned Income Tax Credit until it becomes a de facto universal basic income:
[A] straightforward mechanism to mitigate wage stagnation: a wage tax credit of 100 percent of earnings up to a maximum credit of $10,000 - in other words, a universal erned income tax credit (UEITC). A dedicated, broad-based value-added tax (VAT) of 11 percent would finance the new credit. The maximum credit would be indexed to economic growth. For the first time in decades, low- and middle-income workers would share in economic gains even if the factors suppressing market wages do not reverse. 
The principled idea of a universal basic income has been pitched many times before, even by Nobel Laureates. However, no matter the formal reputability of the messenger, a bad idea is still a bad idea. 

Perhaps the most frustrating part of the pitch for new welfare-state entitlements, this one included, is that the proponents never listen to facts or empirical research. It is well known by now that basic-income handout programs tend to fail when tested; the Finnish experiment, e.g., was proven not to work, and this despite a dedicated two-year effort

No - just like all bad ideas, the basic-income proponents keep coming back, suggesting that the last round just wasn't grand-scale enough to work. This is particularly true with the Tax Policy Center's new proposal. The author, Leonard E Burman, makes exactly this argument; his premise appears to be that we need to make the basic-income program so big that Congress cannot allow it to fail.

There are so many problems with his proposal that I cannot fit them all into one article, but we can certainly start with his ideological inroad to the idea. This is the usual pitch: "income inequality". By making the standard case that middle-class earnings have stagnated, Burman paints the image of a country where the rich get richer impervious of the factors keeping lower-income families in a state of stagnation. In reality, income stagnation cuts across all earnings groups, hitting higher incomes as well as lower.

Yes, this is actually true. In an article on April 19 I explained that the usual superficial look at personal income gives exactly the impression that Burman and others propose, namely that high-earning households have seen their income pull away from the 60 percent that earn the least. But, I noted,
a closer look reveals that things are not nearly as dramatic as they look. In fact, those who want more income redistribution as a desperate solution to raise the standard of living for lower-income Americans are - to brutally fuse two metaphors - clamoring for the straw that will break the camel's back. ... Taxpayers making more than $1 million get less than a third of their income from work; those who fall into the $500,000-$1,000,000 bracket earn just over half their income that way
In other words, those with higher incomes have seen substantial gains not because their work-based incomes have risen, but because they earn substantially from equity-based income. The stock market growth we have seen in the past quarter century is far more responsible for the differences in income that Burman and others observe, than some kind of deliberate stratification keeping lower work-based earnings down. In other words, I noted,
it is reasonable to assume that the performance of the large amounts of liquidity that have been pumped into the U.S. economy over the past 20 years - especially since 9/11 - have fueled the stock mareket in particular, to such an extent that it has allowed wealthier taxpayers to pull ahead.
If you really want to raise low-income families, give them easier access to the stock market. Why not let them save their social-security taxes in a personal account, and invest that money on the stock market?

Anyway. Back to Burman and his dream of a grand-scale basic income entitlements. After having misrepresented income differences in America, he makes a veritable mockery of quantitative reasoning. In Figure 2, page 6 in his pamphlet he tries to suggest that American wages are growing so poorly that we simply cannot say no to his gigantic entitlement. To make this point he delves deep into statistical mythology, comparing - are you ready for this? - the real wage growth in the United States from 1974 to 2015 to the real wage growth in Britain in 1770 to 1869.

Yes, you heard me right. 

I could spend two or three articles explaining how embarrassingly bad this comparison is, but let me just settle for four quick points, and then we will move on to the next flaw:

1. To compare two countries you need to compare them during the exact same period of time. If you don't do that, you allow your data to be contaminated by something that is called implicit axioms, or "different time, different circumstances". Why doesn't he compare America in 1770 to 1869 to Britain 1974-2015? Because he is out to show that the American economy is bad and needs more government. 
2. One time period is 99 years long, the other is 41 years long. Since the time series do not overlap, but together span across four centuries, his selection of years is so unscientific and unscholarly that you would need a particle accelerator to find the difference between his pamphlet and propaganda. 
3. His chart then extends the British real-wage growth line into the future (from 1869) but there is no extension of the American line. Since the American line does not have a documented future, this makes it outright disingenuous - propagandistic - to even put them in the same chart.
4. He compares the mean British wage from the 18th and 19th centuries with the median American wage from the 20th and 21st centuries. Can someone please teach Burman the difference between these two concepts, because apparently he did not learn that in grad school? 

Also, just to make sure his quantitative reasoning falls below every conceivable standard known to man, he uses a British sample for all wages while the American sample is for "full-time males". 

If he were in the car business he'd be comparing a Greyhound bus to a Roman chariot and complain that the sun sets in the west. 

Not even when Burman tries to motivate the choice of time periods does he get anything right. He suggests that "robotics" should have made for a better wage trajectory for American full-time male workers, suggesting that there was no technological evolution whatsoever in Britain from the late 1700s to the late 1800s. In reality, the industrial revolution with mass manufacturing set in precisely during that period of time, with a host of inventions that elevated productivity. Revolutions in infrastructure and transportation (can you say "railroad" and "steam engine"?) allowed mass-market distribution of consumer goods. 

But let's not linger on this sore spot in Burman's brave attempt to look scholarly. Let's move on to the next one. On pace eight he suggests, boldly:
The UEITC is a wage subsidy designed to encourage work. 
He does not explain how the UEITC does this. What he does, though, is reference research showing that a) it is good to work because it makes you live a better life (really?), and b) the Earned Income Tax Credit makes children perform better in school. 

There is just one problem: the two research papers that are supposed to prove that kids get better test scores when we increase the EITC suffer from a curious flaw - the same in both papers. Both of them claim that kids in homes that got more EITC money did better on tests because the EITC was increased, but neither paper explains how that increase in test scores compares to test results for kids in homes who did not get a comparable boost in family earnings. 

Formally, the two papers only look at what factor can most likely explain the increase in test scores in EITC-eligible households. They do not look at how kids from EITC-eligible households performed relative other kids.

If Burman was just out there surveying research, he could be forgiven. The problem is that he is advocating for a major policy reform - and a new tax that would be as big as the personal, federal income tax.  When you are in the business of proposing such big changes to the U.S. economy, you better do your homework.

There is more. Not only is this new UEITC supposed to increase test scores in our schools, but it is also supposed to increase workforce participation. The only problem is that when Burman gets his statistical references right he actually - and unintentionally - makes the case that the EITC discourages workforce participation. In Figure 3, page 9, he reports that workforce participation among men has declined since 1948, but he fails to mention that the participation rate remained fairly stable before Lyndon Johnson's War on Poverty began in 1964. 

It was only after this new welfare-state behemoth went into effect that workforce participation began declining - and then it kept declining. In fact, the decline has continued since the EITC was introduced in 1975, a fact that Burman conveniently fails to mention.

Yet somehow we are supposed to believe that more of the same will have a different result. 

There are several other problems with the pamphlet, but let us skip ahead to the most serious one: the VAT idea. Again, to pay for his new cash handout, Burman proposes an 11-percent federal value-added tax, the revenue from which are supposed to "approximately" pay for the new entitlement. 

As mentioned, this tax would rival the personal federal income tax in magnitude. Before we get there, though, let us note that VATs come with a host of problems. Europe has had a long experience with VATs, and they know that this tax invites fraud. The VAT is also administratively costly, putting small businesses at a disadvantage compared to larger corporations that can afford an in-house tax bureaucracy. Just as an example, when I wrote a study of the Swedish tax system for the Center for Freedom and Prosperity several years ago, I noted that the Swedish national tax agency's VAT manual was 1,000 pages long. 

Imagine what it would look like when magnified to American proportions.

Then there is the cash-flow problem facing small businesses in some countries that already have a VAT. Suppose Joe's Gadgets builds a part that goes into every vehicle that Big Car produces. Joe's Gadgets delivers 100 units every month at a price of $1 per unit. He adds the 11-percent VAT, invoicing Big Car for $111 every month. Now: Big Car wants to make sure that their large cash flows generate some interest as the money flows through their business. When they sell their cars they get paid, say, within 30 days. If they can wait another ten days with paying some of their bills, they can park the sales revenue in the bank and earn interest. 

In other words, Big Car tells Joe's Gadgets to allow them, say, 40 days to pay their bill to him. Joe's Gadgets, facing stiff competition for the contract, has to accept the take-it-or-leave-it wait time. The only problem is that the IRS wants the VAT payment in their bank within 30 days, or Joe's Gadgets faces a penalty. Of course, he gets a refund from the IRS upon filing for taxes, but that money is far and away out. In the meantime, he has to cover the $11 he has to pay in VAT by digging into his bank overdraft. The bank, in turn, charges an overdraft fee that adds to Joe's business operating costs. 

But let us now get to the sheer size of this tax. Burman makes no effort to explain what the VAT tax base would be, other than that it is supposed to be "broad" and pull in $1.3 trillion in net revenue each year. The paper he refers to for his numbers, also a Tax Policy Center study, does not offer any tax base explanation either, except for a hint that it would mimic existing state sales-tax bases. 

I seriously doubt it would be that limited. In 2018, states and local governments took in $610.6 billion in sales and excise tax revenue. If we include the excise taxes, Burman and his buddies at the Tax Policy Center who did the VAT study he refers to, still need to triple the tax burden on the sales-tax base. And they need to be able to do this without significantly adverse effects on consumer spending. 

In reality, Burman wants a draconian tax increase on the U.S. economy. His tax would have to apply to 87 percent of all consumer spending. If we backward-engineer his $1.3 trillion revenue number find that:

a) it is the revenue he expects in 2024, when his VAT is fully implemented;
b) in 2024 private consumption can be expected to amount to $17.8 trillion in current prices (with 2.5 percent annual inflation and 2.5 percent real growth per year);
c) the actual revenue required to generate $1.3 trillion in VAT revenue is $1.7 trillion, because of administrative adjustments of $400 billion;
d) if we divide $1.7 trillion by eleven percent - the tax rate - we find that the tax base has to be $15.45 trillion, or 87 percent of your regular consumer expenditures.

This does not mean that 87 percent of your monthly expenditures (which range from apartment leases and internet subscriptions to gasoline, food, haircuts and car repairs) would be hit with this tax. What it means, though, is that if the tax base is smaller, it will have to be hit harder in order to generate the revenue Burman wants.

And this, we are led to believe, will not have any adverse effects on economic growth. If that were the case, the redistribution systems we already have would not have depressed growth at all. Yet there is abundant evidence that the bigger the welfare state gets, the more lasting is its depressive effect on economic growth

Which brings us to my final point. As growth slows down, more people will be net takers of Burman's universal cash handout program. As more people become net takers, Congress will need to borrow even more money to fill the gap. More deficits, more national debt, more debt costs, more tax increases to pay those costs. More taxes means slower growth - and the spiral continues downward.

There is a better way to grow the economy. Shrink government, cut taxes, reduce regulations and let the private sector thrive. If government were the path to economic success, Cuba and Venezuela would be the wealthiest countries in the Western Hemisphere. 

No comments:

Post a Comment