Saturday, May 28, 2005
Paul Krugman speaks to Princeton alumni
The audience was heavily skewed toward older alumni -- say, 40th reunion on up -- and very young alumni. The former applauded Krugman for his strident opposition to the Bush Administration's proposals to reform Social Security, and the latter asked freighted questions about Bush's "lies" in connection with the Iraq war. No apparent conservative asked a question.
I typed six single-spaced pages of notes -- an abbreviated transcript -- during both his short introductory remarks and through the questions and answers, and if I posted it all it would be too much Krugman by almost any standard. There were, however, some highlights.
Krugman's "reluctance" to write on politics
In his introductory remarks, Krugman characterized himself as an economist with a facility to explain his profession in plain English. When the Times approached him in 1999 to write a regular column, he imagined that he would write about such things as the currency crisis in Argentina and fiscal policy. Instead, he says, he found himself increasingly concerned about the direction of the country during the 2000 election campaign.
It became clear to me that something funny was going on. People were being profoundly dishonest. The issue that radicalized me in 2000 was the issue of Social Security, so in some sense we have come full circle. Then from then on things became increasingly political.
And I hate it, by the way. I wish we didn’t live in that kind of world. I’m a moderate Democrat.
The big thing in The Great Unraveling is that I went out on a limb and said that these are not just bad guys from my point of view, but that we are facing a radical challenge. I have slimmed it down: Basically we have a coalition between a radical religious right and a radical economic right that has come to power in Washington, and does not take for granted things as they were. That we are trying to completely change how the Senate does business as just one small example.
That Paul Krugman thinks we are trying "to completely change how the Senate does business" and that the audience nods approvingly does indeed suggest that somebody is being profoundly dishonest.
Daniel Okrent's "very peculiar blast"
Newshounds know that the outbound public editor of The New York Times, Daniel Okrent, took a shot at Paul Krugman in his farewell article last Sunday. Okrent went out of his way to snark:
Op-Ed columnist Paul Krugman has the disturbing habit of shaping, slicing and selectively citing numbers in a fashion that pleases his acolytes but leaves him open to substantive assaults. Maureen Dowd was still writing that Alberto R. Gonzales "called the Geneva Conventions 'quaint' " nearly two months after a correction in the news pages noted that Gonzales had specifically applied the term to Geneva provisions about commissary privileges, athletic uniforms and scientific instruments. Before his retirement in January, William Safire vexed me with his chronic assertion of clear links between Al Qaeda and Saddam Hussein, based on evidence only he seemed to possess.
No one deserves the personal vituperation that regularly comes Dowd's way, and some of Krugman's enemies are every bit as ideological (and consequently unfair) as he is. But that doesn't mean that their boss, publisher Arthur O. Sulzberger Jr., shouldn't hold his columnists to higher standards.
I didn't give Krugman, Dowd or Safire the chance to respond before writing the last two paragraphs. I decided to impersonate an opinion columnist.
One of Krugman's "acolytes" asked him in sympathetic terms if he had anything to say about Okrent's charge. Many of the people in the audience did not understand the reference to Okrent, so Krugman explained that "the public editor" of the Times -- Krugman never once used Okrent's name himself -- "took a very peculiar blast" at him about the misuse of numbers without supplying any evidence. Krugman said that he had exchanged emails with "the public editor" in the last few days in response to the article, and that Okrent had not supplied any instance of Krugman misrepresenting numbers. He attributed Okrent's criticism to pressure from conservatives, and said that Okrent had questioned him about his columns via email since Okrent had come to the paper a year and a half ago, but that he (Krugman) "always had an answer." Okrent, who was "under constant pressure" from conservatives, finally gave up asking Krugman about the columns and "built up a list of grievances in his mind" which he uncorked in his final column.
Basically, Krugman believed that Okrent had a psychological need borne of pressure from conservatives to find misrepresentations in Krugman's work. According to Krugman, there are no such misrepresentations.
On the distribution of income
One old alumnus fairly bitterly asked whether we were heading toward a world in which a small group of people would make "a minimum of $3 million a year" and most everybody else would earn "a maximum of $3 per hour". Even Krugman had to admit that "it doesn't work that way," but went on to claim that all the improvements in income distribution that had occurred between the Crash of '29 and the late 1970s had been erased, that "we are back to 1920s income distribution -- the Great Gatsby has returned," and that social mobility is declining and "is actually a lot lower in the United States than in Western Europe." Any of the Krugman fans out there want to point us toward the data that prove that last point?
On politics and universities
Another person asked whether universities were "cultivating" this trend toward extreme income distribution. This question had some promise if answered honestly -- one might ask whether the bloated inefficiencies of most universities didn't push tuition beyond the reach of the middle class, for example -- but Krugman dodged it. He emphasized that it was "important to maintain independence of thought," and digressed that David Horowitz was leading an "organized campaign" against that. "Whenever I give a large class, I assume that somebody in there is feeding any example of bias back to him."
A subsequent questioner suggested that Krugman was "censoring" what he said in class, and he denied it. But he did say that he was very careful to label political opinions as such and that when he gave an assignment on a political subject such as fiscal policy, he asked the student to argue both sides on the exam. Krugman quite clearly offered these two precautions into evidence as steps he had taken to innoculate himself against the accusation that he was politically biased in his classes. When the audience tisk-tisked and looked grim, as if these concessions were themselves evidence of rising fascism, it was genuinely hard not to burst out laughing.
The crushing of dissent
One of the younger generation asked why the "Downing Street" memorandum had not received more coverage in the United States, and whether Krugman thought "that the Bush Administration has initimidated the media to the point where they can’t even report on the truth any more?"
"Hell yes," said Krugman. Huh? "Everybody understands that if you are critical of the administration, they will come down on you. They will put pressure, they will scream at the editors, and they will try to destroy your credibility."
Well, Paul, destruction of credibility is the sine qua non of refutation, and any editor who can't deal with somebody from the White House staff screaming at them isn't very committed to free speech.
Shorting the dollar
Krugman was at his most interesting when he talked about economics. He was asked "what are the changes for a big currency crash" and what would its effects be on politics and the economy? Krugman replied that this question is best answered by asking and answering four other questions:
Is the current balance of payments sustainable? If not, will it end with a wimper or a bang? Whether with a wimper or a bang, how bad is the end going to be? What are we going to do? My answers are, in order: "No," "bang," “I don't know, but have a bad feeling about it,” “and run away.” The current balance of payments is not sustainable. Almost always these things come to an abrupt end. We are going to have a “Wile E. Coyote” moment. The case for a big fall in the dollar is overwhelming, with the one caveat that if you think about the alternatives you say, oh, wait a minute, what is so attractive about the Euro or the yen. But we are living way beyond our means, and the Europeans are not. So I think it happens quite suddenly when it happens. How bad? There is one big difference between the United States and Argentina, for example. Our foreign debt is in our own currency. We do not have the Argentina problem, but there may be other stuff. Housing already looks like a bubble. If interests rates go up – which they would if the dollar plunged – housing would be in trouble. Ultimately, I don’t know, for sure, how bad it would be. For what it's worth, Robert Rubin and Paul Volcker, who are the two clamest men I know, are both very concerned about this.
As regular readers know, I have argued repeatedly that the dollar will generally rally against most foreign currencies because the alternatives are so poor. However, I agree with Krugman that there is something going on with American investment flows that is historically anamolous and potentially troubling:
[Our balance of payments] is a very weird situation. The way it is suppose to work is that rich advanced countries lend some of their extra savings to poor countries where the opportunities are. Capital is supposed to flow from the rich countries to the poor countries. What we have instead is that the U.S. is sucking in huge amounts of capital from the rest of the world for the purpose of high consumption, instead of investment, and a lot of that capital is coming from China and oil producing countries. It should be unstable. As a share of GDP, the current account deficit is bigger than most precedents. So far, however, the Chinese are still buying lots and lots of dollars. I believe it is going to blow, but if you had believed me all along and speculated on that you would have lost money.
Krugman also answered several questions about universal health care, said that Brad DeLong was his "favorite econo-blogger" and sympathized with a half dozen "questioners" who stood up and made speeches about the United States becoming an authoritarian country. One woman who looked about 60 claimed she "grew up in Stalin's Russia" and that a lot of the things she saw in the United States today reminded her of Russia under Uncle Joe. Another perfectly respectable-looking member of the audience offered a long rant about the rise of fascism and compared the intimidation of the press in the United States today to the suppression of the press in Germany during the 1930s. Krugman politely deflected these questions with a smile, probably aware that some blogger would be all over him if he agreed. He did not, however, take issue with these people, but only suggested that it would be a poor tactic if he wrote so bluntly in his column:
You have to modulate yourself so you don’t lose people. I cannot write two columns a week saying “gee, this looks like the road to fascism.” It won’t do any good. We all have to pick and choose.
Finally, Krugman confided some optimism.
Weirdly, I’m feeling somewhat encouraged. I was deeply encouraged that Joe Lieberman voted against John Bolton yesterday... If you compare the way things are today to the way you felt after the election, you want to give Harry Reid and Nancy Pelosi a Nobel prize for politicking.
12 Comments:
By 
TigerHawk, at
Sat May 28, 10:52:00 AM:
And how wonderful to see that you're still out there and full of piss and vinager! :)
By Beth, at
Sat May 28, 01:50:00 PM:
What, no pies thrown at him? ;-)
By Edgehopper, at
Sat May 28, 02:51:00 PM:
See, conservative students could have been there to question Krugman...but we have better things to do. It's reunions, and seniors like me get a free (OK, included) wristband that allows us access to all 12 or so reunion tents and unlimited alcohol at all of them. Why waste the time listening to Krugman?
(Yes, I'm writing this during the P-Rade...recovering for a few minutes from back pain after marching through with the band. I'll be back out at the festivities shortly.)
By Gordon Smith, at
Sat May 28, 04:26:00 PM:
Hawk,
Thanks for attending and reporting on one of the leading voices of the left.
You mention that any editor who is committed to free speech ought to be able to handle a little White House intimidation. I couldn't discern from this statement if you support the White House intimidation part or just the committed to free speech part...
By TigerHawk, at
Sat May 28, 05:19:00 PM:
Brian, congratulations! I walked the P-Rade myself ('83), and participated in giving your class a cheer. Perhaps you saw me. In any case, there will be extensive Reunions blogging later.
Screwy, query what the "intimidation" is other than calling up the editors and "screaming" at them. Whether voices are actually raised or not, one wonders what is so intimidating about a politician's butt boy calling up to complain about press coverage. Indeed, any politician's butt boy that doesn't complain about press coverage is probably not doing his job.
There are forms of intimidation that might be worrisome. For example, if Krugman's tax returns had been audited for the last four years, or if a van with big strong guys with ear plugs parked outside his house, then you might say that attempted intimidation was going on. But Krugman didn't allege anything like that. He just talked about people trying to refute his arguments or yelling at his editors.
By stc, at
Sun May 29, 01:51:00 AM:
Its not so hard to figure out how Krugman gets his fact. When you consider the average debt load of most Americans you can see how they finance the average American lifestyle. 33% of homeowners have 0 equity loans, lease cars and live within two paychecks of insolvancy. Factor in the looming housing bubble, rising healthcare costs and unstable stock market, making 401k's unreliable, and you'll find the majority of middle Americans living beyond their means and sinking rapidly. The wealthy are getting all the tax breaks and looking forward to an untroubled future, thanks to the actions of the GOP. Krugman may have dodged the question of higher education, but I won't. The only thing "cultivating a trend towards extreme income distribution" is this administration and the lip service they pay towards education.
By Ramesh, at
Mon May 30, 02:12:00 PM:
On lower social mobility in the US (and Britain) than in Western Europe: second article in http://www.suttontrust.com/news.asp
By stc, at
Mon May 30, 02:57:00 PM:
Cassandra - Yes, education should be locally funded and controlled, which is why No Child Left Behind is such a horrible mistake. And while Bush has increased funding for education, the increase falls far short of the monies needed to pay for NCLB testing and resultant federally-mandated consequences, leaving districts with less money and options.
On the other hand, the question was about higher education, a stated Bush priority and yet the number and amount of Pell grants and work-study programs is falling compared to need. If what we deem important is "making younger Americans prepared and competitive for the high-tech demands of the new millinium", we need to encourage emerging industries like stem cell research and stop making the US Army the employer of no-option, low income high school grads. Runaway deficit spending has the potential to lengthen the income divide by raising interest rates, hurting business and consumers alike, and reducing further available government spending. When I said Bush was paying lip service to education, I should have said to the American economy. My bad.
By KJ, at
Mon May 30, 05:00:00 PM:
If districts have less money, like that mattered as they all have more than enough money to teach the real subjects, then maybe they should use a fraction of their money per student to allow school choice. Then they could use that extra money for the students who stayed behind for the crappy, previously underfunded public education.
Raising interest rates? Yeah, since they are at a 40 year low, a higher interest rate would be Bush's fault.
Stem cell research is an industry? And how do you propose we "encourage" it? Oh, of course, government funding. I guess we can take that money from the schools and pell grants.
By stc, at
Tue May 31, 06:14:00 PM:
kj - Many studies have shown that the amount spent per student is not the best indicator of student performance. Schools that work best have motivated students, parents and community involvement, and locally-originated goals and strategies. School choice works best when it is an option not a necessity. We all have a vested interest in education because an uneducated underclass is an economic drain on resources.
Interest rates are at a 40 year low because the fed manipulated them there, after the tech bubble burst, to hold off inflation. They are currently holding on due to cheap Chinese cash fueling a housing boom. But when over 50% of mortages are interest-only or adjusted-rate, just what do you think will happen when the bubble bursts?
Yes, stem cell research is an offshoot of the medical industry, as airplanes are in transportation and conputers are in technology. The government has supplied money for research in all these areas and many others for decades because it is in the country's interest to be in the for front of develpoment worldwide. The idea is for the goverment to provide seed money and for the free market to take over from there, providing educational opportunities, high-paying jobs and long-term financial stability to the nation.
By 
, at
Tue Oct 04, 09:54:00 AM:
Hey, you have a great blog here! I'm definitely going to bookmark you!
I have a making extra income site. It pretty much covers ##KEYWORD## related stuff.
Come and check it out if you get time :-)
By , at
Wed Oct 17, 10:45:00 PM:
A payday loan is a short-term loan that you promise to pay back from your next pay cheque. A payday loan is sometimes also called a payday advance.
Normally, you have to pay back a payday loan on or before your next payday (usually in two weeks or less). The amount you can borrow is usually limited to 30 percent of the net amount of your pay cheque. The net amount of your pay cheque is your total pay, after any deductions such as income taxes. For example, if your pay cheque is $1,000 net every two weeks, your payday loan could be for a maximum of $300 ($1,000 x 30%).
Before giving you a payday loan, lenders will ask for proof that you have a regular income, a permanent address and an active bank account. Some payday lenders also require that you be over the age of 18.
To make sure you pay back the loan, all payday lenders will ask you to provide a postdated cheque or to authorize a direct withdrawal from your bank account for the amount of the loan, plus all the different fees and interest charges that will be added to the original amount of the loan. The combination of multiple fees and interest charges are what make payday loans so expensive (Click here for an explanation of the various fees associated with these types of loans.
The lender should also ask you to sign a loan agreement. If the lender does not offer to give you a copy of the loan agreement, ask for one. Read this document carefully before signing it, and keep a copy for your records
How and when do I pay back the loan?
A payday loan agreement usually says that you must pay the total amount you owe for the loan on or before the date stated in your loan agreement. This includes the amount you borrowed, plus interest and any additional fees and charges.
Some lenders will cash your postdated cheque or process your direct withdrawal on the day the loan is due. However, some lenders may require that you pay the loan in cash, on or before the due date.
If you have not paid the loan in cash by the due date, some lenders may cash your cheque or process the direct withdrawal you signed on the day after your loan's due date, and charge you another fee. Ask the lender what the most inexpensive way is for you to repay your loan.
How does a payday loan affect my credit report?
Credit-reporting agencies collect information on whether or not you make your payments on time. This information, also called your "credit history", is part of your credit report and is used to calculate your credit score.
Making payments on time can help improve your credit score by demonstrating that you are able to manage your debt. Even if you have poor credit, you can rebuild it by using a credit card or other type of credit and paying back the money you owe on time.
This is not the case with payday loans. Since payday lenders are not currently members of the main credit-reporting agencies, getting a payday loan and paying it off on time will not improve your credit score. However, if you do not pay your loan back on time and it is sent to a collection agency, this will likely be reported to a credit-reporting agency and could have a negative impact on your credit report.
How much will a payday loan cost?
A payday loan is much more expensive than most other types of loans offered by financial institutions such as banks or credit unions. Before you apply for a payday loan, find out about all the fees and charges you will have to pay — including the fees you will be charged if you cannot repay the loan on time. The fees may not be easy to see right away, so read the agreement carefully before signing it. If you do not receive an explanation of all of the fees, charges and interest that will apply to the loan, or if you are not satisfied with the explanation you receive, do not sign the loan agreement.
How does the cost of a payday loan compare with other credit products?
Payday loans are much more expensive than other types of loans, including credit cards. But how much are you really paying? How does the cost of a payday loan compare with taking a cash advance on a credit card, using overdraft protection on your bank account or borrowing on a line of credit?
Let's compare the cost of using different types of loans. We'll assume that you borrow $300, for 14 days. Note the considerable difference in the cost of each type of loan.
Things to consider before you apply for a payday loan
Even if you think you may be turned down, ask your bank or credit union for overdraft protection on your bank account, or a line of credit. These are relatively inexpensive ways of obtaining access to extra funds, for short-term use.
If you are turned down for any of these credit options, ask why. If the reason is that you have a poor credit history, contact the three credit-reporting agencies to get a copy of your credit report. Read the reports carefully to make sure that all of the information in it is correct. If you find any errors, contact the credit-reporting agency to find out how you can have the information corrected. The three major credit-reporting agencies in Canada are Equifax Canada, TransUnion Canada and Northern Credit Bureaus. All three of these agencies will give you a copy of your credit report for free if you request that it be sent to you by regular mail.
Ask yourself if you really need to take out a loan, or whether you can get by until your next pay cheque. If you need the money immediately, try to make other arrangements. For example, you may be able to cash in vacation days. Or you might consider getting a short-term loan from a family member or a friend.
If you find that you need to apply for a payday loan because you have no alternative, only borrow an amount that you are 100 percent sure you can repay on the due date of the loan.
Don't borrow more than you need.
Things to consider if you take out a payday loan
Don't be afraid to ask a lot of questions. Read carefully — and take home with you — a copy of the loan agreement that you are being asked to sign. Don't feel pressured to sign the loan agreement right away if you have questions and want more time to read through the agreement on your own. If the lender does not want to give you a copy of the agreement, look for another lender.
Be sure to ask about all the fees, charges and interest that apply when you first get the loan, and what other charges you will owe if you can't pay the loan back on time.
If you are taking out a payday loan at another location to pay back the first payday loan, or you are extending or "rolling over" the loan that you had with the same lender, you could find yourself in serious financial difficulty. The fees, charges and interest will add up quickly on these types of loans, which can put you into serious debt.
How can I figure out the cost of each type of loan?
To estimate the total cost of a loan, including the annual cost of the loan expressed as a percentage of the amount borrowed, follow the steps below.
Step 1:
Determine how much interest you will pay. First, find out the annual interest rate that applies to the loan (if there is one). Figure out the daily interest rate by dividing the annual interest rate of the loan by 365 days. Then, multiply that rate by the length of time you are taking the loan. Finally, multiply the result by the amount you will borrow, in dollars:
Amount of interest
= Annual interest rate
--------------------------------------------------------------------------------
365 days × Length of the loan
(number of days) × Amount of the loan
Step 2:
Determine the total cost of the loan by adding any fees that may apply to the interest you will have to pay. Find out what fees apply to the loan and add them to the cost of the interest, found in Step 1:
Total cost of the loan = Amount of interest + Total fees
Step 3:
Estimate the annual cost of the loan, expressed as a percentage of the amount borrowed. First, divide the total cost of the loan, found in Step 2, by the amount of the loan. Then, divide this rate by the length of time you are taking the loan (in days) and multiply it by 365 (the number of days in the year):
Annual cost of the loan (%)
= Cost of the loan
--------------------------------------------------------------------------------
Amount of the loan ÷ Length of the loan
(number of days) × 365 days
Let's find out the cost of a $300 payday loan, taken for 14 days.
We'll assume that the lender charges you a one-time set-up fee of $10 and a service fee of $40, which includes interest on the loan.
Step 1:
Determine how much interest you will pay. In this case, there is no interest fee. The interest is therefore $0.
Step 2:
Figure out the cost of the loan by adding together any fees that apply and the interest you will have to pay. In this case, you would add the $10 set-up fee and the $40 service fee together:
$10 + $40 = $50
Step 3:
Estimate the total annual cost of the loan, expressed as a percentage of the amount borrowed:
Annual cost of the loan (%)
= Cost of the loan
--------------------------------------------------------------------------------
Amount of the loan ÷ Length of the loan
(number of days) × 365 days
= $50
———— ÷ 14 days × 365 days
$300
= 4.35 or approximately 435%
The total cost of the payday loan would be $50 with an annual cost of 435 percent of the amount borrowed.
Information asymmetries are common in credit market models, but the usual assumption,
at least in commercial lending, is that borrowers are the better informed party and that
lenders have to screen and monitor to assess whether
firms are creditworthy. The opposite
asymmetry, as we assume here, does not seem implausible in the context of consumer lending.
"Fringe" borrowers are less educated than mainstream borrowers (Caskey 2003), and many
are
first-time borrowers (or are rebounding from a failed first foray into credit). Lenders
know from experience with large numbers of borrowers, whereas the borrower may only have
their own experience to guide them. Credit can also be confusing; after marriage, mortgages
are probably the most complicated contract most people ever enter. Given the subtleties
involved with credit, and the supposed lack of sophistication of sub-prime borrowers, our
assumption that lenders know better seems plausible.
While lenders might deceive households about several variables that in
fluence household
loan demand, we focus on income. We suppose that lenders exaggerate household's future
income in order boost loan demand. Our borrowers are gullible, in the sense that they can
be fooled about their future income, but they borrow rationally given their beliefs. Fooling
borrowers is costly to lenders, where the costs could represent conscience, technological costs
(of learning the pitch), or risk of prosecution. The upside to exaggerating borrowers' income
prospects is obvious—they borrow more. As long as the extra borrowing does not increase
default risk too much, and as long as deceiving borrowers is easy enough, income deception
and predatory—welfare reducing—lending may occur.
After de
fining predatory lending, we test whether payday lending fi ts our definition. Payday
lenders make small, short-term loans to mostly lower-middle income households. The
business is booming, but critics condemn payday lending, especially the high fees and frequent
loan rollovers, as predatory. Many states prohibit payday loans outright, or
indirectly,
via
usury limits.
To test whether payday lending quali
fies as predatory, we compared debt and delinquency
rates for households in states that allow payday lending to those in states that do not. We
focus especially on di
fferences across states households that, according to our model, seem
more vulnerable to predation: households with more income uncertainly or less education.
We use smoking as a third, more ambiguous, proxy for households with high, or perhaps
hyperbolic, discount rates. In general, high discounters will pay higher future costs for a
given, immediate, gain in welfare. Smokers' seem to
fit that description. What makes the
smoking proxy ambiguous is that smokers may have hyperbolic, not just high, discount rates.
Hyperbolic discount rates decline over time in a way that leads to procrastination and selfcontrol
problems (Laibson 1997). The hyperbolic discounter postpones quitting smoking,
or repaying credit. Without knowing whether smokers discount rates are merely high, or
hyperbolic, we will not be able to say whether any extra debt for smokers in payday states
is welfare reducing.
2
Given those proxies, we use a di
fference-in-difference approach to test whether payday
lending
fits our definition of predatory. First we look for diff erences in household debt
and delinquency across payday states and non-payday states, then we test whether those
di
fference are higher for potential prey. To ensure that any such differences are not merely
state e
ffects, we difference a third time across time by comparing whether those di fferences
changed after the advent of payday lending circa 1995. That triple di
fference identifies any
di
fference in debt and delinquency for potential prey in payday states after payday lending
was introduced.
Our
findings seem mostly inconsistent with the hypothesis that payday lenders prey on,
i.e., lower the welfare of, households with uncertain income or households with less education.
Those types of households who happen to live in states that allow unlimited payday loans
are less likely to report being turned down for credit, but are
not more likely, by and large,
to report higher debt levels, contrary to the overborrowing prediction of our model. Nor are
such households more likely to have missed a debt payment in the previous year. On the
contrary, households with uncertain income who live in states with unlimited payday loans
are
less likely to have missed a debt payment over the previous year. The latter result is
consistent with claims by defenders of payday lending that some households borrow from
2
Consistent with a high discount rate, Munasinghe and Sicherman (2000) discover that smokers have
fl
atter wage profiles and they are willing to trade more future earnings for a given increase in current earnings.
Gruber and Mulainathan (2002)
find that high cigarette taxes make smokers "happier," consistent with
hypberbolic discount rates (because taxes help smokers commit to quitting). DellaVigna and Malmendier
(2004) show how credit card lenders can manipulate hyperbolic discounters by front-loading bene
fits and
back-loading costs.
payday lenders to avoid missing payments on other debt. On the whole, our results seem
consistent with the hypothesis that payday lending represents a legitimate increase in the
supply of credit, not a contrived increase in credit demand.
We
find some interesting differences for smokers, but those diff erences are harder to
interpret in relation to the predatory hypothesis without knowing
apriori whether smokers
are hyperbolic, or merely high, discounters.
We also
find, using a small set of data from different sources, that payday loan rates
and fees decline signi
ficantly as the number of payday lenders and pawnshops increase.
Reformers often advocate usury limits to lower payday loan fees but our evidence suggests
that competition among payday lenders (and pawnshops) works to lower payday loan prices.
Our paper has several cousins in the academic literature. Ausubel (1991) argues that
credit card lenders exploit their superior information about household credit demand in their
marketing and pricing of credit cards. The predators in our model pro
fit from their information
advantage as well. Our concept of income delusion or deception also has a behavioral
fl
avor, as well, hence our use of smoking as a proxy for self-control problems. Brunnermeier
and Parker (2004), for example, imagine that households
choose what to expect about future
income (or other outcomes). High hopes give households' current "felicity," even if it
distorts borrowing and other income-dependent decisions. Our households have high hopes
for income, and they make bad borrowing decisions, but we do not count the current felicity
from high hopes as an o
ffset to the welfare loss from overborrowing.
Our costly falsi
fication (of household income prospects) and costly verification (by counselors)
resemble Townsend's (1979) costly state veri
fication and Lacker andWeinbergs' (1989)
costly state falsi
fication. The main difference here is that the falsifying and verifying comes
before income is realized, not after.
More importantly, we hope our
findings inform the current, very real-world debate,
around predatory lending. The stakes in that debate are high: millions of lower income
households borrow regularly from thousands of payday loan o
ffices around the country. If
payday lenders raise household welfare by relaxing credit constraints, anti-predatory legislation
may lower it.
Payday lenders make small, short-term loans to households. The typical loan is about $300
for two weeks. The typical fee is $15 per $100 borrowed. Lenders require two recent pay
stubs (as proof of employment), and a recent bank account statement. Borrowers secure
the loan with a post-dated personal check for the loan amount plus fees. When the loan
matures, lenders deposit the check.
Payday lending evolved from check cashing much like bank lending evolved from deposit
taking. For a fee, check cashiers turn personal paychecks into cash. After cashing several
paychecks for the same customer, lending against
f uture paychecks was a natural next step.
High
finance charges is the main criticism against payday lenders. The typical fee of $15
per $100 per two weeks implies an annual interest rate of 15
x365/14, or 390 percent. Payday
lenders are also criticize for overlending, in the sense that borrowers often re
finance their
loans repeatedly, and for "targeting" women making the transition from welfare-to-work
(Fox and Mierzewski 2001) and soldiers (Graves and Peterson 2004).
Despite their critics, payday lending has boomed. The number of payday advance o
ffices
grew from 0 in 1990 to 14
, 000 in 2003 (Stegman and Harris 2003). The industry originated
$8 to $14 billion in loans in 2000, implying 26-47 million individual loans. Rapid entry
suggests the industry is pro
fitable.
Payday lenders present sti
ff competition for pawnshops, even though the internet, namely
E-bay, signi
ficantly foreclosure costs for pawnshops (Caskey 2003). The number of pawn
shops in the U.S. grew about six percent per year between 1986 and 1996, but growth
essentially stalled from 1997 to 2003. Prices of shares in EZCorp, the largest, publicly
traded pawn shop holder, were essentially
flat or declining between 1994 and 2004, while
Ace Cash Express share prices, a retail
financial firm selling check cashing and payday loans,
rose substantially over that period (Figure 4). EZCorp CEO, Joseph Rotunday, blamed
payday lenders for pawnshops' dismal performance:
The company had been progressing very nicely until the late 1990s.... (when)
a new product called payroll advance/payday loans came along and provided our
customer base an alternative choice. Many of them elected the payday loan over
the traditional pawn loan. (Quoted by Caskey (2003) p.14).
Payday lending is heavily regulated (Table 1). As of 2001, eighteen states e
ffectively
prohibited payday loans
via usury limits, and most other states prices, loan size, and loan
frequency per customer (Fox and Mierzwinski 2001). Note that the payday loan limit ranges
from 0 (where payday loans are illegal) to 1250. Nine states allow unlimited payday loans.
Payday lenders have circumvented usury limits by a
ffiliating with national or state
chartered banks, but the Comptroller of the Currency—the overseer of nationally chartered
banks–recently banned such a
ffiliations. The Federal Deposit Insurance Corporation still
permits payday lenders to a
ffiliate with state banks, but recently restricted those partnerships
(Graves and Peterson 2005).
Regulatory risk—the threat of costly or disabling legislation in the future—looms large for
Payday lenders. The Utah legislature is reconsidering its permissive laws governing payday
lending. North Carolina recently drove payday lenders from the state by expressly outlawing
the practice.
Heavy regulation increases the cost of payday lending. High regulatory risk increases limits
entry into the industry and increases the expected return required by industry investors.
Driving up costs and driving away investors may be exactly what regulators intended if they
view payday lending as predatory.
We de
fine predatory lending as a welfare reducing provision of credit. Households can be
made worse o
ff by borrowing if lenders can deceive households into borrowing more than is
optimal. Excess borrowing reduces household welfare, and may increase default risk.
We illustrate our concept of predatory lending in a standard model of household borrowing.
Before we get to predatory lending, we review basic principles about welfare
improving
lending, the type that lets households maintain their consumption despite
fluctuations in
their income.
The model has two periods: today (period zero) and payday (period one. Household income
goes up and down periodically, but not randomly (for now): income equals zero today
and
y on payday. If households consume Ct in period t, their utility is U (Ct) .Household welfare
is the sum of utility over both periods:
U (C0)+ δU (C1), where δ equals the household's
time rate of discount. Households with high
δ value current consumption highly relative to
future consumption. In other words, high discounters are impatient.
A digression here on discount rates serves later discussion. In classical economics
δ is
constant. If
δ changes over time, so does household behavior, even if nothing else changes.
If
δ(t) is hyperbolic, households will postpone unpleasant tasks until current consumption
does not seem so precious relative to future consumption (Laibson 1997). With hyperbolic
discounting, that day never arrives, so hyperbolic discounters have behavioral problems: they
procrastinate. They may never repay debt, much less begin saving. Hyperbolic discounters
who start smoking may never quit.
Returning to the model, if the marginal utility of consumption (
U 0) is diminishing, households
will demand credit to reduce
fluctuations in their standard of living. Households
without credit, however, must fend for themselves (autarky). Welfare under autarky equals
U
(0)+δU (y). The fluctuations in consumption for households without credit make autarky
a possible worst case, and hence, a good benchmark for comparing cases
with credit.
If households borrow
B at interest rate r, welfare equals U (B) + δU (y − (1 + r)B).
Borrowing increases utility in period zero, when the proceeds are consumed, but lowers utility
in period one, when households pay for their borrowing. Rational, informed households trade
o
ff the good and bad side of borrowing; they borrow until the marginal utility of consuming
another unit today just equals the marginal, discounted
disutility of repaying the extra debt
on payday:
U
0(B) = δ(1 + r)U 0(y − (1 + r)B). (1)
Equation (1) determines household loan demand as a function of their income, their
discount rate, and the market interest rate:
B(y, δ, r). For standard utility functions,
household loan demand is increasing in income and decreasing in the discount factor and
interest rate:
By > 0; B δ < 0; Br < 0. Household welfare with optimal borrowing equals
U
(B(y, r, d))+δU (y − (1+r)B( y, r, δ)). As long as households follow (1), their welfare with
positive borrowing must be higher than without (autarky).
The welfare gain from borrowing depends on the cost of credit production. Suppose the
cost of lending $
B to a particular household equals (1 + ρ)B + f, where ρ represents the
opportunity cost per unit loaned and
f is the fixed cost per loan. Think of f as the cost
of record-keeping and credit check required for each loan, however large or small the loan
may be. If the going price for loans is (1+
r) per unit borrowed, the lenders' profits equal
(
r − ρ)B − f.
With perfect competition among lenders, the loan interest rate is competed down until
it just covers the costs of the loan:
r = ρ + f /B. Equilibrium r and B are determined
where that credit supply curve equals demand (1).
Equilibrium in the payday credit market is illustrated in Figure (3). If
fixed costs per loan
are prohibitively high, the market may not exist. Perhaps the payday lending technology
lowered the
fixed cost per loan enough to make the business viable.3 Before the advent of
payday lending, households who applied to banks for a very small, short-term loan may have
been denied.
Fixed costs per loan imply that smaller loans will cost more per dollar borrowed than
larger loans. That means households with low credit demand will pay higher rates than
households with high loan demand. Loan demand is increasing in income, so high income
households who demand larger quantities of credit will enjoy a "quantity" discount, while
lower income households will pay a "small lot" premium, or penalty. That price "discrimination"
is not invidious, however; the higher cost of smaller loans re
flects the fixed costs of
lending. The high price of payday loans may partly re
flect the combination of fixed costs
and small loan amounts (Flannery and Samolyk 2005).
A usury limit lowers household welfare. Suppose the maximum legal interest rate is
r.
At that maximum rate, the minimum loan that lenders' cost is
f /(r− ρ) = B. Low income
households with loan demand less than
B face a beggar's choice: borrow B at r or do not
borrow at all. Such households would be willing to pay more to to avoid going without
credit, so raising the usury limit would raise welfare for those households.
Competition is another key determinant of how much households gains from borrowing.
3
Alternatively, or additionaly, the demand for small, short term loans may have increased in the mid
1990s. The welfare reform then almost certainly increased demand for such credit as households who once
"worked" at home for the government were forced to go to work in the market.
Even with no competition — monopoly—households cannot be worse o
ff than under autarky.
The monopolist raises interest rates until the marginal revenue from higher rates equals the
marginal cost from lower loan demand:
B
(y, r) = −(r − ρ)Br(y, r) . (2)
At that monopoly interest rate,
rm, household loan demand equals B(y, rm).Household welfare
under monopoly equals
U (Br(y, r m))+δU (y −(1+ rm)Br(y, r m)). Welfare is lower under
monopoly because credit costs more and their standard of living
fluctuates more (because
costly credit reduces their demand for credit) If households borrow from the monopolist,
however, they must better o
ff than without credit.
In sum, welfare for rational households is highest if credit is available at competitive
prices. If households choose to borrow, they must be at least as well o
ff as they were
without credit. Limiting loan rates cannot raise household welfare and may reduce it.
Monopoly lenders lower household welfare, but even with a monopolist, households cannot
be worse o
ff than without credit.
The high cost of payday lending may partly re
flect fixed costs per loan. Before payday
lending, those
fixed costs may have been prohibitive; very small, short-term loans may not
have been worthwhile for banks. The payday lending technology may have lowered those
fi
xed costs, thus increasing the supply of credit to low income households demanding small
loans. That version of the genesis of payday lending suggests the innovation was welfare
improving, not predatory.
In the textbook model household welfare cannot be lower than under autarky because households
are fully informed and rational. Here we show households how can be made worse o
ff
than without credit if predatory lenders can delude households about their (households')
future income.
Suppose that by spending
C(τ ), lenders can convince a prospective borrower that her
income on payday will be
y +τ. The cost C can be interpreted variously as the cost of a guilty
