Wednesday, August 17, 2011

Overvaluing a Law Degree

Matt over at The Law School Tuition Bubble does a good job with research and analysis.  In a recent post entitled "Another Day, Another Study Overvaluing A Law Degree" he reviews a publication entitled "The College Payoff" from Georgetown University's Center on Education and Workforce.  Georgetown suggests that with a law degree your lifetime earnings will be 4,032,000 - far in excess of the lifetime earnings of those with a bachelor's degree (2,268,000) and an associate's degree (1,727,000).



Matt rightly points out that this data suffers from survivorship bias AKA , "the graveyard problem" in that the reported $4M is only for people that actually manage to stay employed as a lawyer for their entire 40 year career - which is only a small percentage of those who graduate law school.  This is a really important point to consider because law does not have a lot of job security - as I discuss here.  My best estimate of those who are initially employed as a lawyer (not just "go to law school", but "employed") and make it through a full 40 years without significant unemployment is considerably less than half.

Matt also points out that the real premium for earning your law degree is the difference between the bachelor degree's earnings and the law degree's earnings - which works out to 1,764,000 - not the full $4M.

One additional interesting thing to consider that is not in the analysis is the impact of taxes.  Why?  Well, the really important number is not how much you "make", but how much you take home.  At certain incomes, you basically get to take home 100% of what you earn, but at other incomes the government can take 40%+ of what you earn.

Let's see how that plays out here.  Let's divide the BS and associate's salaries by 40 to figure out their yearly earnings ($56,700 and 43,175).  For the legal salary, we have to subtract off the three years spent earning the law degree - this shows up in the chart - so we divide by 37.  This gives us $108,973, which agrees closely with the Department of Labor's $110K/year number for average lawyer compensation.

Now let's see what happens to the take-home pay using this simple tax estimator.  Let's use a hypothetical family of 4 - 2 parents and 2 kids - with the standard exemptions and deductions and only one parent working.  Old fashioned?  Maybe, but feel free to create whatever scenario you want.  Here's what we get:

                 Yearly Salary   Tax       Take Home     Adjusted Earnings     Difference
Associate's   43,175       -1524       44699     (x40) =  1,788,000          +61,000
Bachelor's    56,700          971        55729     (x40) =  2,229,000         -39,000
Law             108,973      10,306      98667     (x37) =  3,651,000         -381,000

 Some things to note:
  • With the associate's degree, the family pays negative tax due to all the credits in place.  The welfare program embedded in our tax code just gives them an extra $1,524/year.   If the program continues for all 40 years, that's an extra $61,000 over their careers.
  • Even with a bachelor's degree, although the family is now paying some tax, it is really not much.  They are paying 1.7% of their income as tax.
  • However, once we get to the law degree, taxes become significant.  You can see that we lose almost 10% of our lifetime income to taxes.  
  • Consequently, if we want to see the "true take-home advantage" of the law degree relative to the bachelor's, we need to consider the tax loss and it is already down to 1,422,000 - not the 1,764,000 - a difference of $342,000
  • To put it another way, of the putative advantage to getting the law degree that Georgetown proposes, about 20% disappears immediately when you factor in the higher taxes.
  • Some people may see "earnings" and think "take-home".  Nope, you have to think about the impact of taxes.  Alternatively, people may be familiar with a tax code that  that is kind and gentle and either gives you some money or only takes a little.  That won't be the case when you practice law.
Matt mentions some other factors that also eat into the remaining $1,422,000 advantage including loan payments - both principle and interest.  Additionally, the loan repayment can be quite costly.  For example, if $200K in loans is amortized over 30 years at an interest rate of 6.8%, then the total of payments will be $469,387.

Thus, after taking into account increased taxes and loan repayment, even if you are lucky enough to remain employed for all 37 years as a lawyer, the advantage of the law degree is now only about $953K - that's a far cry from the $1,764,000 that Georgetown was initially trying to tell us - about a 46% decrease.  Consequently, I think that it would be legitimate to say that even if you are fortunate to work in law for your entire career, about HALF of the advantage of getting the law degree is eaten up by increased taxes and loan repayment.

People considering law school should really think about that.


13 comments:

  1. Nice work MP,

    Personally, I'd be happy to take home the extra $42,000 after taxes if law degrees were worth that much.

    On the plus side, I believe the Georgetown study overvalues bachelors degrees too, so if it were accurate on J.D.s, it would cut to the law student's favor. The article that everyone should read is Herwig Schlunk's "Mamas Don't Let Your Babies Grow Up To Be...Lawyers," which does justice to the value of law degrees. For the impatient, the takeaway quote is in Footnote 16:

    "Law students may not appreciate how volatile attorney income is, even in the case of established attorneys. Not all associates become partners. Not all partners become senior partners. Law firms blow up, leaving non-rain-making partners in the lurch. In-house attorneys are subject to all the usual vicissitudes of corporate down-sizing. And so on."

    As for loans, I taught myself how to amortize them myself via spreadsheet. If we take the parameters of your example (two parents, two kids, 200k law school debt), put the loans on IBR, and file separately, we get a monthly payment of $945. IBR will start forgiving loan balances in 20 years rather than 25 starting in 2014, that'll be $226,800 off the lifetime earnings. Then the loan is forgiven, even though it's ballooned to nearly $380k because the monthly payments are less than the monthly interest, which gets capitalized. The forgiven loan adds 380k to the 108,973, meaning our lawyer will have to pay $153,000 in income tax according to the TurboTax site. I think the IRS allows people to claim some kind of insolvency deduction if the tax wipes out their assets.

    ReplyDelete
  2. I should add that I assumed the 200k loans were part Staffords and part Grad PLUS loans, making the average interest rate 7.35% (6.8 & 7.9), which is how ED does it.

    ReplyDelete
  3. Thanks LSTB!
    A couple of things:
    1) I am not asserting that anyone will actually make the additional $953K over their lifetimes. Legal employment is too erratic. The number also ignores large expenses like a partnership buy-in which a 37-year veteran likely had to pay. I am just trying to point out that even using the amazingly biased numbers in the report, you still lose about HALF the value of the law degree to increased taxes and loan repayment.
    2) The $953K over 37 years of practice $25,800/year - you mention $42K, but I am not sure where that comes from.
    3) Schlunk's article is perennial favorite of mine. I have a few quibbles and it needs to be updated, but it is a good read for anyone considering the economics of going to law school.
    4) Thanks for the DirectPlus Loan and IBR notes. Your comments really caused me to take a look at how the loan programs have changed over the last few years - and there have been a lot of changes.

    ReplyDelete
  4. With regard to the changes (and this may expose how I have previously been ignorant - thanks for cluing me in and please confirm this)
    1) Now you can just borrow the entire cost of attendance with a Direct Plus federal loan. That's a pretty big difference because in the 90s student loans only paid so much and then you had to get private loans at a much greater interest rate. I had Staffords at 8.5% and private loans at 11.75%. Federal loans would only pay for about as much as the tuition (year 1) or a few thousand less (year three) (about 20K/year) and you had to come up with the rest of the cost of attending through private loans and work (about $15-20K/year). Books were especially painful to put because they were outside the federal loans and so damn expensive.
    2) Holy Moly does being able to just borrow everything in one fell swoop make it even easier to go to law school. The reduction in interest rate is nice, but the process is so easy I wonder if people even think about it. I recall when I got my private loan they made me go to a class/tv program before they would give it to me - it was basically "you better really be sure. If you are not sure, then maybe you better wait until you are sure." It was obviously a loss prevention aspect for the private lender - I wonder how well it worked? I am not aware that there is anything similar for federal loans - are you?
    3) IBR is really a game changer now that it will apply to your entire loan balance, but I am not sure it is a game changer in a good way. It certainly lowers the perceived risk of a poor result in going to law school. As for the actual risk, it seems like people are putting a lot of faith in the program - which is questionable given the government's track record and current lack of funding. However, if it works, then situations like JDPainterGuy can be avoided, which would be good. However, as you correctly note, the loan forgiveness would currently be taxable (although it appears that loan forgiveness for public service loan forginess is NOT taxable and a bill has been introduced to extend that to IBR.) That would be a massive tax liability. For example, if you never made payments and using your combined interest rate of 7.35%, 200K in 25 years becomes about 1,180,000. Even if we consider inflation of 3%, that number becomes about $590,00 in today's dollars. The tax estimator says that they would owe about $170K to the IRS. (The example you work out is much more elegant and detailed, though).
    4) The libertarian side of me really doesn't like the implication of everyone being so at the mercy of the government. We really need to address the cost issues better rather than just throwing more money after it.

    ReplyDelete
  5. There's no way that your associates degree family gets back $1524 in taxes every year for 40 years. I don't know exactly how you calculated the numbers, since you don't give the details, but it seems to me that you took a single year out of the two-parents two-kids scenario and multiplied it by 40. That's not going to give you an accurate picture of the lifetime tax burden of the family. As the kids get older, they will age out of most of the big-ticket tax deductions such as the child tax credit. This means that the tax liability of the family will change over 40 years as the family ages. Over a 40-year span I would expect that the actual tax savings for the associates degree family would be at the very most half of what you have calculated.

    Playing around with the TurboTax tax estimator, it seems that they only calculate federal income tax. FICA taxes (social security and medicare) are notably omitted, but these taxes are not negligible! Especially at the lower income levels (e.g. your associates degree scenario), the FICA taxes would more than wipe out any federal income tax refund.

    I realize you are only giving estimates, but I don't think you did a very good job at it.

    ReplyDelete
  6. DJ - Your criticism is certainly warranted to an extent. You are right that the children will age out of the child tax credit and (assuming that they are born 2 years apart) the family will only be entitled to the credit for 2 kids for 16 of the 40 years. With zero kids, the husband and wife would pay $2034/year in taxes.

    Here's why I initially didn't include it - my initial thought was that it would wash out because all of the families would be the same. However, I forgot that at the higher incomes the child credit is phased out, so the husband and wife at the higher income will be paying even more tax when the kids are gone, thus making the difference even greater ($14131, an increase of $3825 which is $267 more than the lowest income family's increase.) Incuding this in the analysis will change the lowest income family's lifetime next income tax from negative to positive, but will make the difference between the lowest and highest family even greater.

    With regard to FICA, it looks like all three families would be paying pretty much the same percentage of their income to FICA - 6.2% for Social Security and 1.45% for Medicare - medicare applies to all income and the SS cutoff is $106,800, so just about every dollar would be taxed. You are right that I did not include FICA. Here's why - my intent was to compare additional federal income tax burden with increasing income, not detail all taxes paid. The FICA percentage stays the same with the incomes in the chart, so (unlike federal income taxes) the people will pay the same percentage of their income regardless of total income. Here's why that is important - if federal income taxes operated like FICA, then doubling your income would double your takehome and there would be no disproporionate impact at higher income levels. The higher earning family pays more in total tax, but it is the same percentage of earnings, so it drops out when determining veriably impact with increasing earnings.

    I also didn't include sales tax, property tax, or state income tax in the analysis - any of which could have an impact if you are looking at the total tax burden. However, I am not - I'm just looking at the impact of federal income taxes with increasing income. This is by no means an estimate of the total tax liability of any of the scenarios.

    The point that I am trying to make - that I think that you agree with - is that at higher incomes you are going to lose a greater percentage of income to tax, and that increase will be significant. Consequently, just looking at relative "income" without considering tax liability and loan liability would give a misleading picture. What are the actual numbers? Well they are going to vary based on specific situation and possible changes to the tax code (they might do away with the child credit for one). However, the main point that people should think about the increased tax liability and loan liability when making their decision seems pretty sound.

    ReplyDelete
  7. 1). The 42k was based on subtracting the $56,700 Bachelor’s salary from the $108,973 lawyer salary. I may’ve done that in haste, but whatever.

    2). One of my fears is the moral hazard free legal education has on college seniors. The current economy gives students a good reason to spend three years of their lives in law school at taxpayers’ expense. Fortunately, the number of applicants is dropping through the floor. A while back I did the math and fall 2011 is close to seeing a record low number of law applicants per capita going back to the mid-80s.

    3). To answer your question, I didn’t have to take a class for any federal loans, just an online quiz asking ten multiple choice questions. I remember seeing Grad PLUS loans on my financial aid offering in my 2L or 3L year, without explanation. I balked at the 7.9% rate, but I graduated before the IBR law was passed.

    4). My diminutive libertarian side tells me government shouldn’t play bank. There’s too great an incentive for government to simply look at the ROI without caring about accountability in higher education. Worse, last week I learned that the government isn’t even doing the accounting right. In 1990 it correctly switched from cash accounting for loans to accrual accounting, but it requires using Treasury securities (I have no idea which kind) as the discount rate. When the CBO makes projections using fair-value accounting, government student lending programs are a loss to taxpayers because private investors wouldn’t lend this kind of money out at those kinds of terms.

    ReplyDelete
  8. LSTB -
    1) OK, thanks.
    2) Great! Maybe we can get back to equilibrium and people cans top getting hosed.
    3) Wow. There really should be more attention draw to the fact that people are signing their lives away.
    4) I'm a big fan of getting a return on my tax dollars, but it really doesn't seem like it is happening if many people default on their loans. The wacky accounting does not help. We need to deal with our issues honestly. Hiding expenses and lying about the long-term implications are what got our economy into teh present sorry mess.

    ReplyDelete
  9. Student loans are not at "the taxpayers' expense" unless the borrower dies penniless before paying off the loan or gets a hardship discharge in bankruptcy (which is VERY hard to do).

    ReplyDelete
  10. 10:40 - Thanks for the comment! With regard to the "taxpayer's expense", I am thinking about the new IBR's loan forgiveness after 25 years. Whatever is not paid off by then (and it may very well be a pretty big amount) is going to be a taxpayer expense, right? Taxpayer money made the originaly loan and so it would have to be paid back (with interest payments at least as high as th rate of inflation) in order for there to be no net loss, right?

    ReplyDelete
  11. I disagree with Anonymous 10:40. The point of student loans isn't for the government to make money--government does that by taxing, borrowing, or printing money (all the same really)--the point is to create more productive workers who will increase economic output. There's plenty of good evidence out there that when the costs and benefits are monetized, law school is a losing proposition for a majority of students, meaning the government is not getting higher quality workers. The same goes for higher education generally.

    Debt zombies are bad for the economy, which is a far more urgent mandate for the government than making a profit on student loans.

    ReplyDelete
  12. I would like to appreciate your work and would like to tell to my friends.

    ReplyDelete