----- Original Message ----- From: "Erik Reuter" <[EMAIL PROTECTED]> To: "Killer Bs Discussion" <[email protected]> Sent: Sunday, January 16, 2005 8:03 AM Subject: Re: Social Security
> * Dan Minette ([EMAIL PROTECTED]) wrote: > > > But, wouldn't that just be a slope change....dFIT/DE = .15 to dFIT/DE > > = .25? If so, then as long as I'm using the right marginal rate, then > > I'm calculating the Federal Income Tax (FIT) contribution to T and dT > > correctly. > > Sure, I was pointing out that you might NOT be using the right marginal > rate. You said you chose 15% initially (although I see you using 25% > below). (which I corrected back to 15% a bit later.) > If Kotlikoff's and Gokhale's assumptions about the additional > income are different than yours or they are calculating with a different > year tax laws, the marginal rate could easily be different. This is one > thing I am looking to clarify in their calculation. OK....I've looked at recent tax rates. For example, the pivit point from 15% to 25% is at ~57k of taxable income, well above 40k....and taxable income is after at least the standard deduction and allowance for each member of the family....which pushes the 40k to below 30k for a family of 4. > Okay, let's make this a bit easier. Massachusetts has a flat tax of > 5.3%. (There are other states where it could be even higher, so it is > not like I picked the outlier here) Let's use Massachusetts, okay? Agreed. > > For my example, dR/dE between 39k and 40k, they'd be in the 4% > > bracket. But, since state income tax is deductable, and the > > dFIT/dE=0.25 (25% tax > > I think you are assuming that this couple will be itemizing their > deductions. Did you check to see if they really would be better > itemizing than taking the standard deduction? No, since that involves a number of unknowns. That's a fair point, so we can use the 5.3% for the state income tax rate for computations. > > Also, if one does this, it should be added to the income as well as the > > tax. That's not critical when the rest of dR/dT is low, but as it gets > > high, its important. Let us consider the example where the marginal net > > tax rate calculated without considering employer paid tax on income as > > either income or tax is 75%. Let us then consider a 7.7% tax being added > > to this. The result is (.75+.077)/(1+.077)=.77=77%. Since they were > > talking about net tax rates around 80%, this gives some idea of how the two > > ways of calculating the tax would change the answer. It's only another 2%. > > That is debatable -- I can see it going either way. It all depends on > how you define income. If Kotlikoff was talking about pre-tax income > where pre-tax REALLY means before ALL tax, then you don't add that to > the denominator. You are defining the $20K income in a certain way > (after some taxes), but it is not the only reasonable way. That's true. The reason I defined it this way is that wages and salaries are almost always described this way. Income tax tables, minimum wage law, wage limits for various benefits, etc. are all given assuming the definition I used. Anyways, what I was getting at is that, when the marginal net tax rate is high (in the 70% range), the difference between including the employers SS tax payment as both income and tax, and including it as neither income or tax is only a couple percent...because, in transforming from one way of analyzing the data to another, one adds/subtracts that tax from both the numerator and the denomenator. In short instead of "If Kotlikoff was talking about pre-tax income where pre-tax REALLY means before ALL tax, then you don't add that to the denominator." I'd say ""If Kotlikoff was talking about pre-tax income where pre-tax REALLY means before ALL tax, then he already has added that to the denominator." > Anyway, rather than arguing about definitions, I propose we take what > I think is the simplest definition: the income we will discuss in our > calculation is before ALL taxes. So, if we say someone has a $20K > income, then we mean that in a zero tax world, their income would be > $20K. After taxes, their income would be much less than $20K. > If we do this, it may be that we will be calculating for a lower income > family than you had initially assumed. OK, we can go ahead and put that through. But, the only tax on pay that I can think of is the SS and Medicare tax on wages paid by the employer. That would lower 20k to about 18.5k, doing quick mental math. > > But, for income that is already in the 20k range, we shouldn't be > > talking about much in the way of Mediaid or welfare benefits. Take > > But, why do you make this assumption? The tax law looks very complicated > to me in this regard. Have you studied it carefully? I'd say that I've been following this moderately carefully for the last 25 or so years. Part of at the Federal budget, I know that Medicaid is, by far, the largest need based aid expenditure..roughly 120 billion according to http://www.whitehouse.gov/omb/budget/fy2002/bud22_1.html Food stamps are 21 billion, and supplemental child nutriton is about 9 billion. We've talked about those already. Suplementary Security Income is 31 billion, but, from the looks of things, http://www.ssa.gov/notices/supplemental-security-income/ I took the test for benefits for our hypothetical family , and it appears that it was possible for the spouse to get SS spouse benefits...if the spouse was elegible for SS. The spending on housing assistance is about 18 billion/year. > Kotlikoff specifically mentions Medicaid and housing assistance (in > addition to food stamps and EITC that you already included), so I'd > bet it can be significant at certain income levels. The housing assistance income limit for a family of 4 is 18.2k/year. Even with the renormalization to include employer paid SS and Medicare taxes, this is below the range of our example (20k to 40k). from: http://cses.com/rental/section8.htm#Income%20Limits: >Again, it depends > on the income assumptions, obviously. If the family being examined is > lower-income than you are assuming, then more of these benefits could > become important. Right. I have a source for the Medicaid limit at http://www.legalaction.org/medicaid2.htm which places it far below 20k/year, even after the renormalization for employer paid taxes. I'll agree fully that my arguements about the accuracy of the dR/dE that was proposed for the example of one spouse taking a 20k/year job when the other one already had one may not apply for people who, for example, lose welfare and Medicaid when they take a low paying job. Indeed, my guess is that this is where dR/dE is the lowest. > If you have already carefully studied this, then I'd certainly like to > see your notes or references showing what benefits such a family does > and does not receive at what income levels. I've given some income limits. It appears that a far better arguement could be made for lower income examples than the range the authors were discussing. > If you don't know, then I'll just have to continue my reading, and I > will post my notes on it eventually. > > Also, if anyone else can think of federal benefits that go away when family > > income rises from 20k/year to 40k/year, I'd be interested in seeing some > > numbers. If you don't have the inclination to do the numbers, just mention > > it and I'll see what I can do. > > If you have a ready reference, how about checking on Medicaid and > housing assistance? I hope the references I gave above prove useful. Dan M. _______________________________________________ http://www.mccmedia.com/mailman/listinfo/brin-l
