A good way to try to think about it, but unfortunately there are a few key conceptual errors here that make any predictions like this deceptively hard to make. Not to say it’s impossible, but it’s a bit more complex than that.
The first problem is that you are inferring too much about the future. Personally, whenever I have made forecasts like this, I found that any predictions that go beyond 20 years into the future are impossible to make. While you could probably answer “where do you see yourself in 5 years?” with a decent degree of accuracy, it is much harder to be accurate about “where do you see yourself in 15 years” and you can very well be completely wrong about “where do you see yourself in 25 years” (unless your answer is “in a mirror” ).
Even as an average case performance (more like “median case” here though), it’s pretty hard to imagine that someone would work and receive a 3% year-on-year raise for 45 years, no more no less. In fact, I’d say that one of the arguments in favor of an elite school is the idea that Elite U credentials would give you a much faster growth curve - that is, not only do you start with a higher salary but your salary will also grow at a faster rate. Those credentials, and perhaps the skills you learned in school, may be more likely to propel you into the high growth positions at the top of the industry, or give you opportunities to found startups that State U would not have. I’d personally try something like giving the Elite U grad a 5% year-on-year raise, while leaving the State U at 3%, to test this assumption, and I’m sure you’d agree that that would skew the results much more in the favor of Elite U over time. It’s not a great assumption, but it’d be a start.
A related issue is considering the bounds of student performance. I’m quite certain that you would find that the low performers (e.g. weakest 10% of students or graduates) of Elite U are going to be much, much better off than the low performers of State U. That would probably look like “worse job than I bargained for” for Elite U, and “no degree or no job, still in debt” for State U. On the high performers, it’s often difficult to separate how much the school does vs how much the person him/herself does to achieve that success. The very top has a tendency to be highly irregular.
The second issue is in how you consider COL to be a simple flat 95% of income. There are two main problems with this:
- There is nothing stopping people from shifting between regions, i.e. moving to a higher or lower COL area, and getting a salary that is similar (with a small percentage boost for Elite U).
- As your salary increases, your COL as a percentage of income is smaller. If you make $50k, most of that goes into living expenses. If you make $500k, well you’d be just fine if you spent a mere $150k on your COL.
That will guarantee to shift your model a fair bit.
The third issue is a very simple one but often implicitly supported by people discussing loans. When you choose not to take a loan to go to school, you do not have that money on hand. The government borrowed money to be able to fund a $700 billion stimulus package; I did not borrow that money, so does that mean I have $700 billion on hand? Of course not. The same applies with the loan you do not take for Elite U. And you certainly can’t earn interest on that money you didn’t take in loans either.
The fourth issue has to do with interest rates.
Student loans have a percent interest associated with them, which I usually estimate to be 6% overall. That’s actual money that has to be repaid on some repayment schedule, and the best way to do that is to actually explicitly flesh out a repayment schedule. You may pay more in absolute terms in the future, but perhaps your effective use of that money is going to offset your cost and make the choice of Elite U worth it.
In principle, modeling savings isn’t the worst way to do it, but it’s also not sufficient. Perhaps at a $5k/yr savings, all you can do is invest in an index fund. At a $50k/yr savings, you have enough money to be able to get a significant share of high-risk, high-reward businesses that, on average, will make higher percentages. The $5k/yr saver won’t be able to take those businesses because they can’t diversify enough to avoid a situation in which they might lose all of their money (one bad risk = say goodbye to your entire savings).
I won’t go into detail because it’s a rather in-depth topic, but you really should do the analysis in terms of the [internal rate of return](Internal rate of return - Wikipedia) and [net present value](Net present value - Wikipedia) of the investment. It would be a better way to judge the two.
Hope that helps.