Case Structuring Contest

I came across the following television interview the other day.

The person being interviewed is a leader of a group of college students demanding that all student loans in the U.S. be forgiven, and that college tuition be free.

She was not prepared, and ended up humiliating herself on live national television.

Specifically, she was not familiar with the relevant facts on the issue, did not do her math, and presented her opinion as a conclusion.

The interviewer asked her a few legitimate questions, which quickly revealed her lack of preparation.

In short, she violated every rule of the CEO Code, and all my guidelines for what NOT to do during a case interview (which is a proxy for interacting with senior level clients).

First, the video is worth watching because it discusses a controversial issue. Such issues are precisely when consultants are called in to help make a decision.

Second, the interview very much illustrates the dynamic between an MBB partner and a first year consultant or intern that isn't prepared. In this case, the TV interviewer plays a role similar to an MBB partner, and the student being interviewed plays the role of the new consultant.

Partners do not like people who waste their time (CEOs hate it too). They absolutely will let you humiliate yourself, in part out of a morbid curiosity to see if you will recognize you screwed up and acknowledge it, or continue to insist on the validity of an argument you haven't thought through yet.

Third, rather than solely using this video as an example of what not to do, I thought it would be more constructive to hold a contest to re-write the student leader's argument.

Because you and I don't have access to factual data, it's not realistic to expect anyone to rewrite her entire argument. However, what is reasonable is to structure this "case" and lay out how you would determine your conclusion, what analysis you would do (if you had the data), and why.

In other words, state your hypothesis and how you'd structure a way to test it. For the purposes of this contest, you must assume the same position as the student leader. Your hypothesis must be that it's a good idea to make college tuition free and to forgive all student debt.

The winner gets a $100 donation made in your honor to the U.S. charity of your choice, and will be publicly acknowledged (with your permission) by me for your excellent case structuring skills.

If you're one of the approximately 10,000 people reading this that currently works or previously worked at MBB, you are not eligible to win the prize (but are welcome to participate).

I will post the winning entry on my blog and include it in my email newsletter.

I've never held a contest like this, so I'm curious to see what happens. To submit an entry, post your answer as a comment below.

I will contact the winner privately first to offer my congratulations and to communicate with him or her before posting the winning answer.

UPDATE as of Tuesday, November 24TH AT 8AM ET: New entries are welcome, but not eligible for the prize, as contest has closed.

79 comments… add one
  • Jd Nov 19, 2015, 10:02 am

    a) Expected Costs

    • Total cost of existing student debt
    • % yearly growth trend
    • Average debt $ per new student going to college. This is vital to understand given the debt will increase even more given we will have additional students attending given college is free now.

    b) Funding is available
    • Net worth after 50 % existing tax code for Ultra rich and
    Rich is significantly higher than existing student debt and
    forecasted additional student debt.

    • % yearly growth of net ultra-worth

    c) Benefits

    • College attendance will increase.
    a) Calculate or Guestimate the increased college attendance %
    (data needed is the current % of students who DO NOT
    attend college solely because of college attendance)

    • College drop outs will decrease
    a) Calculate or Guestimate the decrease in drop out %
    (data needed to calculate this is the current % of students
    who drop out of college because of tuition fees)

    • Alternatives are available to primary funding mechanism – alternatives such as the government or the middle class is available in case funding from rich is not available any year.

    • Increased pricing of tuition cost in the future.

  • Jd Nov 19, 2015, 10:04 am

    apologies I seem to have double replied…

    * My only change is

    a) Calculate or Guestimate the increased college attendance %
    (data needed is the current % of students who DO NOT
    attend college solely because of high tuition fees)

  • Adrian Obleton Nov 19, 2015, 10:13 am

    If my hypothesis is that “college in the United States should be free for all students”, there are three assertions that I would need to be true.

    1. A material number of new students will be able to attend school
    2. The economic benefit will outweigh the costs
    3. The quality of education will not decrease

    To expound on the first point, I would need to believe that there are qualified students who currently do not attend college for economic reasons. I define qualified as having grades and test scores indicate at least 90% confidence that they would successfully complete a degree within six years. I would gather that information through statistical analysis of people who have finished degrees and surveys of students who were not able to attend college.

    I would then use the number of qualified students to estimate the economic impact of this proposal.

    I would want to verify four facts to verify that the economic benefits will outweigh the costs. First, the job market could absorb the new students who now have those degrees without meaningfully devaluing the current degree holders. Second I would want to figure out how much value is created by having students without substantial debt who are able to invest in the economy. Third I would want to verify that the decrease in domestic investment created by a tax increase to pay for this program would not outweigh the benefits listed above.

    Finally, I would verify that the quality of the education would stay same by checking two assertions. First that colleges would not lower their admission standards to let less qualified students in. Second, I would want to make sure that employers would not view degrees as less valuable, which would offset the economic gains I had calculated previously.

    If I am able to prove those three things: that it will have an impact on a material number of people, that it will be a net positive gain for the economy, and that the quality of education in the US will not decline, I would believe that college in the US should be free for all students.

    • Lisa Nov 19, 2015, 8:07 pm

      I think the man was condescending and would not have spoken like this to a guy.

      • Victor Cheng Nov 24, 2015, 7:20 am


        Most of the anchors on that network are condescending to anyone that disagrees with them. He actually wasn’t that bad compared to what I’ve seen on that and other networks around political issues.


    • Victor Cheng Nov 24, 2015, 7:18 am


      This is pretty good. I’ll have to think a bit to see if there any flaws, but nice job.


  • Nadir M. Nov 19, 2015, 10:15 am

    Hypothesis: its a good idea to make college tuition free and to forgive all student debt

    1-what are the overarching goals
    – Social impact/justice
    – Students will be able to perform better in college
    – Graduates will be able to focus on finding better jobs
    – increase productivity

    2-Economics of the proposal
    – Cost
    – 1 time
    – What is the total current student debt
    – On-going
    – Annual tuition fees payable
    – US Population x college going age people x % that
    actually go to college (include raise in this % since
    more will go to college) x annual tuition fee
    – Benefit
    – Increased productivity (More people able to go to college,
    perform better jobs, make more scientific discoveries)
    – Increased taxation from more productive and well
    earning labour force

    3-Viability (Is this even possible)
    – How will this be funded?
    – Increase taxation by x% for top 1% earners/increase
    taxation by x% for upper 50% earners
    – Reduce defense budget by x%
    – Increase corporate tax rate
    – Are there enough jobs for increased # of college
    – Due to increased supply will wages for college grads
    – Will people who don’t even want a college degree now
    get a college degree
    4-Indirect effects
    – Is there enough supply of colleges? will there be
    congestion in colleges?
    – What will happen to wages of unskilled people?
    – Will college education standard be maintained
    5- Alternatives
    – Mandatory college savings account for each family to
    fund college education from their own income for each

    • Victor Cheng Nov 24, 2015, 7:31 am


      Is each question you ask absolutely essential? Pick any one question, would the answer to that question change your recommendation?

      For example, one of your questions is, “is there enough supply of colleges?”

      Is this a VITAL question? If the answer is “yes” would you change your recommendation from free college tuition to not, or vice versa?

      Force every question through the same test of importance.

      Assume each question costs the client $50,000 in consulting fees to answer (because it often does), did you spend the clients money in the most efficient way possible?

      If the final recommendation were yes, tuition should be free. What answers would need to be true in your list of issues? You didn’t identify those possible answers in advance with suffice clarity that a client could easily understand,


  • George Nov 19, 2015, 10:27 am

    I base my arguments on the information shared in the video solely as I am not familiar with the other discussions regarding the problem.

    My hypothesis is that college tuition should be made free and student debt should be forgiven.

    In order to test this hypothesis I would like to look into 3 main areas:
    1) Potential benefits and drawbacks of relocating government income to forgive student debt
    2) Total yearly spent with college tuition and its trends
    3) Government yearly taxation income and its trends

    Firstly, I would look into the potential benefits and drawbacks of relocating government income to forgive student debt. I imagine such benefits to be long-term and difficult to measure: better educated population contributing more to the development of the economy, less crime, etc. I imagine drawbacks to be: other industry sectors will suffer (e.g. healthcare), other sectors will ask for more money from the government and produce strikes reducing economical output, etc. If the drawbacks would prove too difficult to tackle by the government, I would be in favor of dismissing the hypothesis.

    Secondly, I would look into the total yearly spent with college tuition and its trends. I would calculate this based the total number of students and the average yearly spent with one student. Suppose this is $1.3trillion. If trends would show a significant reduction in yearly spend I would consider that as a positive effect for students to continue their education, otherwise I will consider to be more and more difficult for them to bare with the pay.

    Finally, I would look into the government yearly taxation income and its trends. Suppose this number is currently $1trillion. If the trend would overpass the college tuition in future years I would be inclined to allocate more from the government budget to compensate for the college tuition. If, on the contrary, the trend will be opposite I will investigate other ways, such as laws to prohibit excessively high college tuition.

    To sum up, college tuition should not be made free and student debt should not be forgiven. This is based on the two following reasons: 1) in order to make tuition free the government will need to relocate $1.3trillion, which is not available even if they would tax the top 1% rich with 100% (that would bring in only $1trillion; 2) the current drawbacks of the relation overpass the benefits.

    The solution I propose is a compromise and is two-fold. On the one hand, the government should focus its available financial resources to cover the college tuition as long as other sectors of the economy are not affected. On the other hand, the population should contribute more to the education system – a larger tax can be applied to people who benefitted from the educational system and a smaller tax to those who did not.

    • Victor Cheng Nov 24, 2015, 7:26 am


      Points 2 and 3 felt confusing to me. You indicated what you wanted to measure but I didn’t see the linkage to what you would recommend based on those calculations. The rationale didn’t seem clear to me.

      If yearly tuition was increasing as an overall trend, would that be in favor of advocating for free tuition or the exact opposite? Why? What’s the clear obvious logic?


  • Badarinath Grandhe Nov 19, 2015, 10:40 am


    1.) Understand the core issue in implementing the three options proposed – Obviously costs involved and possible sources of money

    2.) Calculate the total extra costs that are going to incur in implementing the three conditions – Based on current figures and estimating a ball-park figure of possible number of extra students who may join the college

    3.) Feasibility analysis of implementation – Analyze the various sources of revenues, understand the current spending of the government on various sectors and a possible squeeze from those for example defense and increase of taxes and finally come up with a few options

    4.) Understand the implications – Do a benchmark analysis if any other country has implemented such thing and see whether they are successful or not. If yes, try to replicate with some changes to adapt to the home country’s requirements and if no, understand what can be done differently

    5.) Evaluate these multiple options based on the feasibility analysis and come up with three final options or combinations of options to propose a solution to the core problem of how to implement these changes

    6.) If its not feasible, think of any other alternatives to these demands – relax them a bit and come up with some viable demands

    • Victor Cheng Nov 24, 2015, 7:32 am


      So if your final recommendation were to recommend free tuition, what answers to your questions would you need to see to feel comfortable with that conclusion?


  • Nadir M. Nov 19, 2015, 10:44 am

    Same commenter as above, just fixing the formatting since the comment box removed some spacing i had:

    Hypothesis: its a good idea to make college tuition free and to forgive all student debt

    1-what are the overarching goals
    — Social impact/justice
    — Students will be able to perform better in college
    — Graduates will be able to focus on finding better jobs
    — increase productivity

    2-Economics of the proposal
    — Cost
    — 1 time
    —– What is the total current student debt
    — On-going
    —- Annual tuition fees payable
    —– US Population x college going age people x % that
    actually go to college (include raise in this % since
    more will go to college) x annual tuition fee
    — Benefit
    — Increased productivity (More people able to go to
    college, perform better jobs, make more scientific
    — Increased taxation from more productive and well
    earning labour force

    3-Viability (Is this even possible)
    — How will this be funded?
    — Increase taxation by x% for top 1% earners/increase
    taxation by x% for upper 50% earners
    — Reduce defense budget by x%
    — Increase corporate tax rate
    — Are there enough jobs for increased # of college
    — Due to increased supply will wages for college grads
    — Will people who don’t even want a college degree now
    get a college degree
    4-Indirect effects
    — Is there enough supply of colleges? will there be
    congestion in colleges?
    — What will happen to wages of unskilled people?
    — Will college education standard be maintained
    5- Alternatives
    — Mandatory college savings account for each family to
    fund college education from their own income for each

  • Nicolas Devriese Nov 19, 2015, 10:49 am

    Hypothesis: it’s a good idea to make college tuition free and to forgive all student debt.

    This case can be broken down into 2 parts:
    In the first part we should analyze if it’s a good idea to make college tuition free and in the second part we should analyze if it’s a good idea to forgive all student debt.

    Firstly, I would like to start looking at the total amount of college tuition spent per year in the US.
    Knowing this number will enable me to know how much additional revenues or decrease in costs we need in order to make college tuition free. At this step I would like to analyze possible new revenue streams (such as increase in taxes, corporate sponsorship etc) and see if it is possible to decrease costs (decreasing professors’ wages , optimizing the colleges’ capacity etc). Then, I would like to see if the additional revenue streams and decrease in costs can counter free tuition by doing the following calculation: (amount of college tuition spent per year in the US) – (new revenue streams + decrease in costs). If this result is equal to 0 or bigger than 0, this means that mathematically we could make college tuition free. On the opposite, if this result is lower than 0 this would mean that it would not be possible to make college tuition free.

    Secondly, I would repeat the same process for the student debts.

    After this analysis I would have a good idea if it’s possible or not to make college tuition free and to forgive all student debt.
    If my analysis is favorable (meaning that we should indeed make college tuition free and forgive all student debt), I should however state some possible risks linked to this decision such as for example: a decrease in quality of education due to lower wages of professors, unsatisfied population due to high taxes etc. that we should take into account.

    • Victor Cheng Nov 24, 2015, 7:39 am


      Your structure in part 1 seeks to determine if it is feasible to make college tuition free. By your logic, if you could raise taxes enough could,you make tuition free.

      However, your structure doesn’t account for whether doing so would be s good idea or beneficial.

      The size of the U.S. Economy is approximately $13 trillion gdp. (This information was not provided in the video, so you were not expected to know it.) Total outstanding college debt is $1.3 trillion according to the video. Assuming student loans have a loan duration of multiple years and the portion that is financed is moderate to significant, it’s a reasonable assumption that annual college tuition is less than $1.3 trillion. Given the U.S. Economy is 10 times the $1.3 trillion figure, mathematically one could raise taxes enough to pay for the tuition.

      But just because it is mathematically possible does not necessarily mean it is a good idea. Your structure doesn’t answer that question.


  • Lou Nov 19, 2015, 10:50 am

    “I’m fighting for free college and full forgiveness on student debt. Given that the burden of college education is bore by either state money or students themselves (assume no scholarship/financial aid), let’s start by looking at state funding. I’d hypothesise that non-ability of states to fund free college is due to less state money as a result of slow economic recovery. I’d test both the numbers of state money and economic trends in the past for information on that, but there are also strong economic recovery evidences that supports an interest rate hike already.

    If there are strong economic recovery data already, then the burden to remain on student selves would only be justified if the lifetime value for these students are lower than the state’s investment. I’d test this by looking into data of average GDP produced by college grad vs. non-college grad, taking inflation into account, and compare with the average tuition fees.

    It would be of able-state’s interest to invest into students if their lifetime value is beyond the investments required to put them through college.”

    • Victor Cheng Nov 24, 2015, 7:44 am


      Your last sentence is the most insightful sentence you wrote. The structuring prior to that sentence seemed loose and not cleanly organized.

      I would suggest starting with your last sentence, making it your first sends, and the explaining what you would have to calculate to determine the two key facets of your argument.


  • Jan Wilczak Nov 19, 2015, 10:52 am

    Let us restate the problem one more time – make college tuition free and forgive all student debt.
    Firstly – why do we need? What is the problem? How can our society benefit from it? We could say e.g. that providing equal access to education and making it dependent on talents and hard work and now on one’s wallet is one of the basic duties of country. And inequality and discrimination of poorer people is a big problem – if you cannot have rich parents you cannot get good education. And yes, if we make more talented people study, then our country will become more innovative and our economy will thrive.
    We need to know all the facts such as: How many students do we have? How many of them come from rich families? How much are the student loans?
    I believe it’s good to find any good, working examples of our thesis which were successfully introduced somewhere in the world. Among many countries you can study without any tuitions in Austria, Denmark, Finland, Norway, Scotland, Sweden and Germany. Furthermore, even overseas students can study free in those countries. Going step further – many of those countries provide big grants and loans for living for students. So yes – it is possible, while still remaining rich country.
    Probable questions (or proactively mentioned):
    1. Financing? I would make thorough analysis about how it works in other countries. We need to know exactly how much it would cost and where to take money from. We could have some professors, economists confirming our analysis. Maybe we could also take money not only from the government but also from private corporations? The conclusion should be that is will pay off as this is some kind of investment in our society. Meaning – we will have better doctors, engineers, and so on…
    2. How long can a person study free? Can I study my whole life? Certainly not. One could be allowed to study one subject, or when abandoning it he could study something else the remaining years.
    3. When would it start? What about students who are in the middle of studies?
    4. What about overseas students?
    The same applies to student loans. We should do our math. Maybe it would be good idea to make student loans forgiven and tuition free based on income (in case the math wouldn’t be on our side). Meaning that poorer people will have free studies and the richest would have to pay.

    • Victor Cheng Nov 25, 2015, 2:34 am


      You narrative asks many interesting questions. However, its unclear to me (and you didn’t explain) how EACH question has the potential to prove or disprove your hypothesis. For example “How long can a person study for free?” if the answer is 2 years does your recommendation switch from free tuition to no free tuition? What if it is 3.5 years? Does your recommendation change?

      If your recommendation would not drastically change due the answer of EACH question you ask, perhaps that question is merely interesting but not essential to disproving the hypothesis.


  • Haci Nov 19, 2015, 11:10 am

    H1: its a good idea to make college tuition free and forgive all student debt.
    I’d like to compare debt world to debtless world and would like to do this in 4 buckets (borrower, bank, public, student). I’d expect that overall benefits in debtless world would outweigh overall benefits in debt world

    1: borrowers
    (Assuming its families) What percentage of student families borrow (a)? How much have they borrowed on average (b)? How many families have borrowed (c)? b*c/a gives me the total size of funding needed in debtless world whereas b*c gives me current burden. I’d like to compare them to have an understanding of the burden on student families (debt world) vs burden on general public (debtless world). I’d like to see similar numbers here.. The larger the gap, the more burden public will have (i.e putting a heavy weight on whole population to fix a much smaller population)

    2: banks
    I would look to see what percentage of bank loans are student loans and if the % is material. I would further dive to see if there any sectors/areas of economy that has lack of bank loan funding. The hypothesis would be that loans would be utilised more productively elsewhere (not as student loans)

    3: public
    In a debtless world, taxpayers bear the cost (b*c/a) which is actually more than the outstanding debt since when education is free govt will be covering costs of all families regardless of whether they would have student debt or not. I would expect to see enough resources readily available or can be easily made available in government budget to finance this. Also would expect to see a high priority of education in government agenda.

    4: students
    I would like to see some form of benefits here in debtless world as well, although probably wouldn’t be easily quantifiable. More comfortable life, less stress and more focus on high priority activities etc..

    • Victor Cheng Nov 24, 2015, 7:50 am


      Too many unnecessary questions. If the % of banks that offer student loans is say 50%, would that support a conclusion that college tuition should be free? Or not free?

      It’s unclear. If it’s unclear, one most consider whether the question is necessary as it doesn’t help to prove or disprove the hypothesis. You had a side hypothesis that maybe loan funds would be better deployed in other parts of the economy, how would you test this? Just knowing the % of banks that offer student loans doesn’t help test the original hypothesis Nor the side hypothesis. Also is disproving the side hypothesis absolutely necessary to disprove the main hypothesis?

      Good structuring involves the minimal absolutely necessary questions needed to disprove the hypothesis.


      • Haci Dec 2, 2015, 11:07 am

        Hi Victor,

        Its not the % of banks, its the % of bank loans. And if %50 of bank loans in USA were student loans I would have difficult time suggesting to make education free, since that would create a large shock on banking system

  • Kirill Nov 19, 2015, 11:23 am

    1. Feasibility

    1.1 How much funding do we need? (USA students’ debt volume + average annual cost of tuition at target colleges * average number of students in a year * number of years for this program)
    1.2 How much funding can we get?
    1.2.1 Part of the budget, that potentially might be available for the program
    1.2.2 Taxes (additional income to the budget at a critical tax rate)
    1.2.3 Anticipated donations

    2. Benefit > Cost

    2.1 Benefit: evaluate the impact on GDP/GNP by getting an increased (projected) number of professionals (don’t forget to subtract % that tends to leave the country or happens to be not economically active)
    2.2 Cost: what are the anticipated loss to budget due to wealthy class leaving the country?

  • Oleg Nov 19, 2015, 12:22 pm

    It seems that it is a good idea to make college tuition free and to forgive all student debt funding the expenses from the government. There are 3 main reasons:
    1. education is public good, that is all population benefit from educated people, not only student himself
    2. efficiency of education increases due to enrollment of smarter rather than wealthier students
    3. students saved money stay in the economy and stimulate demand, which is good for tax payers

    To further analyze the problem we need:
    1. estimate benefits from education to the whole population through better decision making by educted people, less spenditures on jail and police for educated people, higher income of and as a result higher taxes from educated people
    2. estimate number of capable students who don’t apply to college because of money shortage
    3. compare impact on economy of investing in education with for example in war and weapons

    • Victor Cheng Nov 24, 2015, 7:56 am


      You were on the right track with your first three points. If you second three points were intende to support you first three pointe , so point 1 at bottom of page supports point 1 at top of the page, they don’t line up.

      If they were intended to be separate points, they should be labeled points 4 , 5 and 6. If that was the intention, points 4, 5 and 6 were not MECE with points 1, 2, and 3.

      Overall most interviewers would consider your proposed structure “not clear or clean” — too much overlap and non MECE-ness between points.


  • Madhav Banerjee Nov 19, 2015, 12:37 pm

    Hi Victor,

    Hypothesis: It’s a good idea to make college students tuition fee free and waive off all the student debt.

    As it is not mentioned in the case, I assume that by “all students” we mean ” all students studying in public and private schools and colleges”.

    If my understanding is correct, I need to look at 3 aspects:
    1. Present Scenario (i.e Null Hypothesis= all paying)
    2. Hypothetical Scenario (Altr Hypo:None paying)\
    3. Waiving off all student debt

    1. In the present scenario , the formula that could be applied is: Profit= Revenue -Cost
    For government institutions, the profit would be negative, as there are a lot of scholarships and other costs (infrastructure, r&d, professors)that are borne by government to subsidize student’s education. To prove my hypothesis, I would like to look at average cost per student incurred by government and I would also like to look at the revenue that government collects through student fees, to fund their education.

    2. Hypothetical situation: All tuition fee free.
    Profit= Revenue -cost
    As there is no revenue, so no profit, and therefore there would be only cost . I would like to look at total cost incurred to fund a student’s education. I would also like to look at the past cost data and find out year on year growth in government funding on education, as this would add to govt., spending more, year on year to maintain the present quality. I would like to look at the gap between the present funding by the government and complete funding by the government when education is free.

    Eg. Present Situation: Total cost of educating a student= $100 a year. So, student pays $40 and Govt. pays $60.
    Hypothetical Situation: Total Cost: $100
    Govt. Spending =$100, so the gap is $100-60$= $40.

    Now, once I know the gap, i would like to know the revenue sources from where govt. can fund this education. It could be debt, i.e. borrowing money. Borrowing money means paying interest rates on that borrowing. The other could be cutting spending on various sectors and investing that money on education. We can look at the sectors where the spending is maximum. For eg. Defence, Healthcare, Social Security and so on. So, if we look at 10 sectors and the gap is $40. We need to get these $40 dollars from these 10 sectors.

    On the other hand we would also like to look at the impact on cutting on expenses on healthcare, defence, social security and other sectors. I would like to look at present govt spending on each sector and the gap. This would give us an idea how much the gap would widen up and what would be the impact? For eg. If govt is spending $100 to minimize infant mortality rate, and suppose, present infant mortality rate is 5% and it requires $500 to reduce it to 2%. So, if by funding free education this amount widens and presently govt. reduces its spending by $10 on reducing infant mortality and plough it in free education. So, how long it will take to minimize this number?

    I would also like to look at the debt aspect that needs to be waived off for all students.
    The implication:
    Profit: Revenue-Cost
    1.If loans are waived off, Government needs to pay off the loans to the private and public banks from their own kitty. If so, they need to borrow money with an interest rate. So, more borrowing and more interest rates. So, no revenue, neither any profit, and over that, it is only cost.
    2. What is the overall impact on the economy, as we borrow more money, so, what is the total borrowings and total interest rates country is paying and how will cost of education add to it? So, how much more interest rates we will be paying? What will be the new deficit between revenues and cost? How can we increase revenues: Investors investing more in US, thereby adding more jobs and citizens paying more taxes. But, are these all possible?
    3. Will the credit rating agency downgrade USA’s rating? If so, what will be its impact? Will investors invest in US any more or move to invest in Asian countries like India or China. If they move to Indian and China, less jobs in US, so less educated people in US and more brain drain, as they will look for options outside US. In a way, fuelling growth for countries like India and China. To prove my hypothesis, I would like to do a primary research, wherein, I would interview few rating companies, and CEOs of big companies from all sectors to understand their perspective and their future investment plans with regards to free education.

    Recommendation: Without any data it would be foolish to come to a conclusion, however having said that, the proposition looks quite scary. The reason it looks scary, because moment government feels any financial stress in funding education, immediately it would look at opportunities to reduce cost, thereby, getting non-phd professors, part time teachers/professors, less spending on R&D and other ways of reducing cost that would impact on the overall quality of students produced from the colleges.

    Still, it is the data that will tell us the real picture.

    • Victor Cheng Nov 24, 2015, 8:06 am


      Your structure feels like you’re forcing the profitability framework on to this case when that’s not the core issue at play.

      If you apply the profitability framework to schools to students, if tuition is free their costs go down and their profit goes up. That is not a major insight.

      The same applies to schools or the government.

      The core issue in this case is whether they aggregate benefit to society or the economy is hire with free tuition or without free tuition. Your structure didn’t cover this point in a well organized, linear fashion.

      Also in terms of your conclusion, K – 12 the grade education in the United States is offered fo free for all children. Is that “scary” as well? The U.S. Is already offering free education prior to college. Why not add just 4 more years? What would have to be true to justify that conclusion?


  • Dallas Cullen Nov 19, 2015, 1:06 pm

    Hi Victor,

    Here is how I would have structured my argument.

    Hypothesis: it’s a good idea to make college tuition free and to forgive all student debt.

    This statement is, at its core, a matter of economics (long-term profitability). I therefore want to dive deeper into the long-term effects such a policy change would have on revenues and expenses. Additionally, I would like to analyze the qualitative effects this policy would have on society to determine its sustainability and overall impact on the US population.

    As profitability is made up of both revenues and expenses, I’d like to take a look at both branches individually.

    On the expense side, I would like to look at three aspects in more detail:

    1) The current size of aggregate US student debt, to determine the immediate amount of capital required to extinguish it.
    2) Both the current and expected future college education cost per person, to determine how much money needs to be raised to subsidize the average student’s education. I expect the future costs will be a function of inflation and annual costs to maintain college campus infrastructure and personnel salaries.
    3) Both the current and expected future number of college students as a result of this policy change, to estimate the amount of annual subsidies required. I expect the future number of college students will be a function of demographics (age) in the country, as well as the population’s price sensitivity and perception of value of a college education.

    By determining the above three factors, we should understand the current and expected future expenses of the policy change.

    On the revenue side, I see three possible revenue streams/segments that could subsidize this policy change: individuals (through taxes), businesses (through taxes), and the federal government (through debt). After analyzing each of these, I want to analyze the increased capital (GDP) the country would earn as a result of a more educated population. I would like to look deeper into all of these revenue aspects by understanding the following in more detail.

    1) Current annual personal and business income levels, and expected income growth rates, preferably arranged by tax bracket (e.g. one million people earn an average of $500,000 and currently pay 50% in total income taxes; 1,000 businesses generate an average of $250 million in net income and currently pay an average of 39% corporate tax on that income; etc.). This will segment two of the three revenue streams into income brackets and show us how much additional revenue could – feasibly and hypothetically – be generated through these means.
    2) The US government’s ability and cost to source the necessary amount of debt. To analyze this, I would like to look closer at the government’s current borrowing rates, taking into account potential rate increases and lender downgrades that could result from the increased debt burden.
    3) Expected increase in annual GDP with the policy change. By comparing the increase in annual GDP to the increase in annual taxes and/or debt, it will be possible to calculate a discounted rate of return and determine the net economic benefit for the US as a whole.

    By determining the amount of potential additional revenue available, and the expected net benefit as a result of this policy change, we can determine whether or not the policy is feasible, practical, and financially rational.

    Finally, to test the feasibility and potential success of implementing this policy change, I would like to look at two qualitative areas in greater detail.

    1) The US populations’ perception of such a policy change. In order for this to be successful, there must be a desire for a more educated population, and a willingness to pay for the increased subsidies. Without either of these premises, the hypothesis is not feasible. If the majority of the population is against this idea, it is unlikely to generate the necessary level of votes in government. Additionally, if those who would be responsible for generating the additional tax revenue – assuming this method of funding is chosen – are unwilling and able to avoid paying the required amount, the plan is no longer feasible.
    2) Assuming there is a desire and willingness to pay for this policy change, I would like to analyze the past success of similar policies in other countries. By analyzing past implementations of similar policies, we can learn whether or not such policies were successful, what aspects worked well, and what issues arose.

    By analyzing each of the above aspects in detail, we should be able to derive a data-driven conclusion to the initial hypothesis: it’s a good idea to make college tuition free and to forgive all student debt.

    • Victor Cheng Nov 24, 2015, 8:16 am


      It feels like you’re “boiling the ocean” on this one – a McKinsey phrase meaning doing everything humanly possible to cover all of your bases; but often at the expense of being inefficient.

      You mentioned 9 key points. Could any of the 9 key points be removed and not materially alter your final recommendation?

      The most important insight was the second point 3 on net impact of economy wide gdp; but it got buried in your structure.

      Also you cited a lot of analyses. It didn’t clearly explain how those analysis would or would not support your hypothesis. If you and data on the U.S. population segmented by income bracket, so what?

      If 10% of the population earns over $250,000 USD is that a clear argument for free tuition or a clear argument against? From my stand point it seems unclear. If it’s unclear, then it’s not helpful. If it’s not helpful, it’s inefficient to include in your structure.

      In contrast your point of about net gdp being higher for the economy as a whole if tuition were free, THAT is material and relevant

      Only include the most relevant and essential issues.


  • Alexandra Bregman Nov 19, 2015, 1:37 pm

    Keely’s hypothesis is that the 1 percent of wealthy Americans is indebted to the 99 percent, because the burden of academic costs is an immoral injustice. She cites her personal experience and does not address the macro-economy.

    Instead, I think the most proactive course to overhauling infrastructure would be to address the financial burden of massive debt on government benefits to the rich. It would be in the best interest of educational philanthropists and innovators to specifically redirect funds allotted to re-configuring academics to address the cost divides that allow for intellectual access gaps in the first place.

    Citing statistics from DOE implementation plans as well as defaulted loan debts, I would redirect those sums into scholarship payment plans for students. I would also propose that while this should be done on a widespread level available to students from all socioeconomic strata, funding should be additionally focused on meritocracy. The goal would be that instead of inviting the uber-rich to pay for moral reasons, it would be in beneficial to restructure the fiscal model to create a smoother, smarter system.

    The ultimate goal would be that students who graduate debt-free have more opportunities to pursue careers they are passionate about, to reap long term gains that fully benefit the economy overall. Although I am not calling for wealth redistribution per se, I am calling for the chance to make wealth available to as many people as possible to avoid divides that cause problems on a macro level.

    • Victor Cheng Nov 24, 2015, 8:21 am


      You presented a bit of a moral argument and drew a conclusion, but you didn’t cite any specific data or identify what data you would seek and now that data might change your recommendation.

      What I’m looking for here is an analytically derived conclusion or structuring of the issue,


  • Srivats Iyer Nov 19, 2015, 1:57 pm

    Its a good idea to forgive student debt and offer free tuition because investing in education for the entire population will provide a higher net return to the government in the form of higher lifetime tax revenue from higher income due to the improved level of education compared to other items in the general budget.

    • Victor Cheng Nov 24, 2015, 8:18 am


      Is that a conclusion or a hypothesis? If hypothesis how would you test it? One layer more detail would be helpful. If it’s a conclusion, what data are you citing?

      You overall focus is the right one. You grasped the key insight in this case, it need to present it with one layer more detail.


  • Ryan Nov 19, 2015, 2:18 pm

    Hi Victor,

    I have a couple comments about the contest. First, I think entries should be posted in a way so that others cannot see what previous participants have written. Second, in regard to the hypothesis, I believe it would be better to state an ultimate goal with the hypothesis (i.e., how do we define a “good idea”).

    For my case structure, I will assume that the ultimate goal is to strengthen innovation and long-term economic growth in the United States. My hypothesis is that making student tuition free and forgiving all extant student debt is the best way to achieve this goal.

    Three main issues must be examined to test this hypothesis:
    1. Funding feasibility
    2. Will free student tuition and forgiveness of student debt improve innovation and long-term economic growth?
    3. Are there alternative measures that could achieve our goal in a more cost-effective manner than the proposed measures?

    Issue #1
    For funding feasibility, I would examine the options for making student tuition free:
    a) increasing government revenue
    b) cutting other expenditures in the government budget
    c) a combination of (a) and (b)

    Furthermore, in regard to forgiveness of student debt, I would want to know who the major creditors are and how the debt will be forgiven. That is,
    a) Will the creditors be legally obligated to accept the losses?
    b) Will the government cover any of the losses of principal and foregone interest?
    c) If option (b) is taken, then how will the government finance covering the losses? Issuance of more debt? Quantitative easing? Adjustments to the budget?

    For Issue #1, I mainly want to see if the proposed measures have a net positive or a net negative impact on the government budget, and I would want to see what the impact looks like over multiple time spans. (Of course, with more time, we could also analyze this issue at the federal, state, and local levels.)

    Issue #2
    Would these measures strengthen innovation and long-term economic growth? To answer this question, I would want to know the following:
    a) drivers of innovation and long-term economic growth historically in the United States, as well as internationally

    Based on the answer to part (a), I would then structure my following questions too see if the proposed measures create conditions that match what economists and sociologists consider sufficient for innovation and long-term economic growth. I would structure it with sections on risks and rewards of the proposed measures, as well as other issues that could potentially have both positive and negative implications.

    Examples of risks include:
    a) Can we statistically measure the risk and implications of abuse of free privilege?
    b) Would the inflated sense of millennial entitlement run rampant under a tuition-free regime?

    Examples of rewards include:
    a) Without student debt, would graduates have more time to select work that they enjoy and thereby do better work, leading to a more productive economy?
    b) With less urgency to remain entrenched in the workforce to pay off student debt, would more people choose entrepreneurship?

    An example of other issues:
    a) How would the proposed measures change the STEM demographic at U.S. universities? How would those changes affect innovation and long-term economic growth?

    Overall, for Issue #2, I would identify the risks and rewards of the proposed measures and then weigh them against each other.

    Issue #3
    Here I would brainstorm alternatives, ideally in a MECE fashion, and then compare the resulting analyses with the analysis of the proposed measures. After making a final comparison of the feasible options, I would then draw my conclusion.

    Thank you, Victor, for the opportunity to engage in this exercise. It was an enjoyable process.


    • Victor Cheng Nov 24, 2015, 8:29 am


      Feedback on contest format noted.

      On your entry, your initial structure was good. The second layer of detail underneath each point was inefficient. You sought to analyze interesting things, but they seem like exploratory analyses as opposed to analyses to disprove a hypothesis.

      For example, in your section 3 you have an incredibly broad and unfocused approached to trying to determine alternatives to free tuition. It seemed very meandering.

      From a communication standpoint, too many examples of this risk and that risk. The entire structure needs to be communicated in about a minute or two. I had to re read your entry 3 times to grasp what you were trying to say, I kept getting lost. That’s not a good sign because unlike I a real case interview, I can re read the words you typed until I comprehended what you were trying to say. In a live interview, the interviewer just gets confused,


  • John McConville Nov 19, 2015, 3:31 pm

    Hi Victor,

    In order to structure this case, I would follow the directions of the interviewer.

    I would start of as Keely did by saying my hypothesis was that it is a good idea to make college tuition free and forgive outstanding student debt. The first thing I would do is break off the double assumption of forgiving outstanding student debt. That is a separate issue, let’s tackle one problem at a time. My refined hypothesis is that making college free for new students is a good idea.

    As I see it, there are two things that determine if an idea is “good” or “bad”. In order to test the validity of my hypothesis I would need to examine two key areas.

    #1 What is the net benefit of this program/idea? (is this a beneficial idea) and
    #2 Is this idea/program realistic or even possible?

    Starting with #1. The benefits of this program are somewhat obvious: everyone gets free college and the US becomes “more innovative” (or so we assume). If I was able to obtain any data on whether or not this assumption is reasonable, I would consider that data. Perhaps it has been shown that college graduates start fewer successful businesses compared to High School grads. Perhaps if college was free, data shows that the quality would decline and fewer people would attend, etc. My point here is that I would collect any data that might prove/disprove the preliminary assumption that having free college for new students would be beneficial to society. However, for the sake of argument, let’s assume that it would be beneficial (ignoring costs of course).

    Now I would look at #2. Is this financially possible? The first piece of data I need is how much tuition is paid each year. Preferably we would have data that could be used to predict future tuition revenues for all colleges in the US; thereby giving us a reasonable assumption as to how much money we would need to make tuition free for the year 2016 alone. Let’s say we determine that it would cost $10 billion. Then we would need to adjust that number based on the expected increase in the future state where tuition is free. Let’s say that because tuition is free, X million more students choose to attend college so we expect 2016 tuition to be $12 billion.

    Now we know that we need an additional $12 billion in 2016 tax revenue (or other revenue sources) in order to make this possible. Keely proposes the idea that the 1% should pay for this. The interviewer suggests this would not be enough to cover the costs. That is an additional hypothesis that we can test. However, as per the contest rules, this is not the hypothesis for which you requested a structure. The result of that hypothesis test will either be “Yes, it will be enough” or “No, it will not be enough”. Based on the video, it sounds like the interviewer thinks it wouldn’t be enough. He suggests that even if we increased taxes substantially, the 1% would go elsewhere or not work as hard. He suggests Greece might be a good example of what happens when these entitlement programs get out of hand. These are valid points. Ultimately, we would need substantial data to prove or disprove this hypothesis. This is where I would tell the interviewer something like: “let’s assume you’re correct and the reality is that the 1% could not pay for this program”, then I would move to the following:

    If the answer is determined to be: “No, it will not be enough”, I would need to find another source of revenue. Could we tax the %5? the 10%? Maybe we could reduce military or other spending to cover the costs. The problem with this entire case is that it hinges on the assumption that we could find a way to pay for it. Without the facts, we don’t know if that assumption is reasonable. If we assume that we can’t find the money from another revenue source other than the 1%, I would conclude that it is unlikely that this idea would be successful (and is therefore a “bad” idea). Of course I wouldn’t say that on national TV. Instead I would ask the interviewer: “do you think it’s reasonable to assume that there is some way by which we could pay for all or part of this program?” He would, of course, say: “Yes I think we could pay for PART of the program, but not all of it”. Then I would conclude that even if we can’t fully fund this program, it would still be beneficial to fund as much as we can. If we increase taxes on the 1%, and it’s not enough to pay for this program, it’s reasonable to assume that we could come up with an alternative program by which we can improve the higher education system in our country. For example, reducing interest rates on student loans or putting a cap on tuition fees (or reducing it if there is one already).

    If the answer is determined instead to be: “Yes, it will be enough”, we can conclude that based on the assumption that there is an expected benefit from free/public college AND we will have a method by which we can cover the costs, this must be a “good” idea (assuming that “good” ideas are those that either break-even or see net benefits) because this idea will at least break even.

    Now that I’ve determined that making college free for new students is a good or bad idea, I can look at whether or not it’s a good idea to forgive outstanding student loans. If we determined that free college for new students is a bad idea, I’m going to assume that forgiving outstanding student debt would also be a bad idea (although we could test that as well).
    However, if we determine that free college for new students is a good idea, we can look at our option for forgiving outstanding debt. We will either:

    A) not forgive debt and allow it to be grandfathered out
    B) forgive some debt
    C) forgive all debt

    In the interview, it sounds like she wants to forgive all debt (option C), but the structure is the same for all options. The same framework used for the first hypothesis applies here:
    #1 What is the net benefit of this program/idea? (is this a beneficial idea) and
    #2 Is this idea/program realistic or even possible?

    #1 The benefit is that everyone gets their student loans forgiven. I guess you could say the benefit to society is the value of the outstanding debt. So now we can move on to #2.
    #2 There are many people who rely on the income generated from those student loans. Trillions of dollars of loans can’t just be forgotten. Someone has to pay for them. In fact, this part of the analysis is a bit unique because it is possible that the costs for “forgiving” outstanding loans could be more than the benefit to society (but I don’t think that’s likely). That is something we could explore.
    In the video, Keely wants the 1% to pay for them. First we must determine the revenue generated from student loans each year. Then we have figure out if and how that number can be paid.

    To conclude, once we have determined a conclusion for the first hypothesis, then we can perform a similar analysis on hypothesis 2, which will ultimately provide us with a position on the merit of the program.

    John McConville

    • Victor Cheng Nov 24, 2015, 8:34 am


      Way, way too long for a case structuring exercise. A case structure should be concise and tight. In a verbal interview, it would take about 30 seconds to communicate. Yours was probably closer to 5 – 7 minutes or at least it seems that way.

      Too much detail, too many examples, too soon. Just layout the “table of contents” of your approach, and why you choose the approach, and. not a summary of every “chapter” too.


  • Prakhar Singh Nov 19, 2015, 3:31 pm

    Hypothesis : It’s a good idea to make college tuition free and to forgive all student debt.

    The hypothesis has two components, with two different impacts:
    1. Making college tuition free: This would impact all the current and future students (to be) enrolled in public colleges.
    2. Forgiving all student debt: This would impact the current and the past students who had taken a debt to pay for their school, irrespective of the kind (public/private) they went to.

    For this stand to be feasible, both 1 and 2 should be feasible economically.

    Starting with 1.
    Cost incurred = No. of public colleges * No. of students/college * Total tuition fee/student = $X per year

    With 2.
    Cost incurred = Number of students passing out of college every year * fraction of students taking loans * Debt/student * Summation(i/Debt term) = $Y
    where i ranges from 1 to Debt term.

    X is the cost incurred per year and Y is one time cost.
    Assumptions made:
    1. Number of students passing out every year are same.
    2. No student will be taking loans once tuition fee is waived. Or, even if they take, the leader is talking about only one time loan waiver.

    For this hypothesis to be correct, the government should be able to incur these costs.
    The government can be able to increase education expenditure by:
    1. Increasing revenue.
    2. Redistributing expenses.

    Starting with 1.
    – Increasing direct taxes
    – increasing rate
    – increasing incidence by incorporating some more low income populace
    – Increasing indirect taxes
    – increasing rate
    – taxing more goods and services.

    Now 2.
    Listing down the current expenses of the government.
    – Basics
    – Developmental

    We will look into these fields to get an idea whether allocating budget is possible.

    • Victor Cheng Nov 24, 2015, 8:37 am


      Your structure answers the question is free tuition financially and mathematically feasible. However, the question being asked implicitly in the interview is, “Is free college tuition a good idea?

      Just because something is financially feasible doesn’t necessarily mean it’s a good idea.


  • Roman Nov 19, 2015, 5:01 pm

    Dear Victor,

    first and foremost I really want to thank you for all the information you provide on your website!

    I really like your idea to hold a “case structuring contest”. Obviously, all the numbers I use are made up, but in my eyes the point of the challenge is the structure and not to do research and check the numbers. So without any further ado I present you my case-structure:

    I would like to state my hypothesis as a question which is:
    Is it a good idea to make college tuition for free and to forgive all student debt?”

    To answer this question, let´s look at the facts.

    A. How big is the impact of a free college tuition and to forgive all students debt in terms of student numbers.
    – Firstly I would check public data about the number of students in the US who would benefit from a free tuition every year.
    – Secondly I would try to find out, how many students would benefit if all the student debt would be forgiven. As this number is properly not captured in any statistic, I would try to estimate this based on public data using a top-down evaluation

    Result: There are about 15-20 million students who would benefit from a free college tuition. The number of students who would benefit if all the student debt would be forgive is obviously smaller, roughly 25 %, so about 3,75-5 million students. The impact of this would be huge!

    B. How big is the impact of a free college tuition on the U.S. economy in terms of an increase in the GDP?
    – Students from US universities are highly educated and very innovative. Every year new start-ups are found by students coming from college. But there is still more potential because students coming from the university with a big amount of debt are not willing to take even more risk/debt to start their own company. Recent studies show, that most of the companies are found by people who are not already having a big amount of debt. As a result of the free college tuition the amount of companies found right after university would increase from 0.5 % to 5 %, which leads – based on a success-rate of about 50 % of the new found companies – to an increase in the GDP of 0.5 % per year. We have benefits for the economy and society!

    C. Do we have enough money to pay for a free tuition and to forgive all the student debt?
    – First, I would name the amount of money we need every year for a free college tuition and how much money we need to forgive all the student debt and spread this over the next 10 years. Let´s say this leaves us with 1 trillion US Dollar every year for the next ten years.
    – Second, I would calculate the future tax earnings by an increased GDP and substract it from the1 trillion US Dollar. Let´s keep the numbers simple: The GDP increase leads to 100 billion US Dollar tax income which is used to pay the free tuition. This leaves us with 900 billion US Dollar.
    – Third, I would check literature for research results on the maximum amount of taxes, rich people(income higher than 250k US Dollar) are willing to pay without leaving the country. Let´s assume it´s 60 %. As a result of the increased taxes to 60 % the money we need for a free tuition reduces from 900 billion to 400 billion US Dollar.
    – To pay for the final 400 billion US Dollar, a nationwide tax increase of (let´s say) 0,5 % is necessary.
    Result: We are able to pay for the free college tuition and to forgive all the student debt.

    Conclusion: We should vote for a free college tuition and to forgive all the student debt!

    Best wishes from Germany,


    • Victor Cheng Nov 24, 2015, 8:41 am


      Your points A, B and C are reasonable. Your logic in that’s graph below B is weak. You make a number of assumptions that have not been confirmed yet as true. Also I strongly advise against using specific numbers. Better to say what numbers you would need to know and WHY and how to relates to your hypothesis.


  • Vincent M Nov 19, 2015, 5:59 pm

    It’s a good idea to make college tuition free and to forgive all student debt if 1) the costs outweigh the benefits 2) other important considerations.

    1) cost and benefit analysis.
    -Uncle Sam’s bill will be equal to whatever # of students are enrolled in a Public school multiplied by the average annual tuition expense. For this example, we will assume that capacity will remain constant. To find the maximum number of public school students we can think about how many public schools there are in States. Small states will have 2-3 public schools while larger states can have 5-15, with California coming to mind as having over 10. If we assume the 30 least populous states have an average of 3 public schools and the more populous 20 have an average of 10, we can assume the US boasts approximately 300 public schools. From my personal experience of attending a medium sized university in Virginia, a mid-sized school enrolls in the range of 10,000 students. From this we can assume approximately 3,000,000 students could currently be attending public universities annually in the US. So to find an average tuition fee I am quick to look at my own personal experience from my mid-sized school in Virginia. An in-state tuition was in the balpark of about the $15,000 range which was also typical amongst my friends in different states. This number will multiply to provide a $45b tab for the government if it were to cover all tuitions in a 100% capacity scenario.

    Moving to the benefit side, I can think of two major long-term advantages. First is in the form of a reduction of student loan defaults, a cost saving measure. Second is the added value for the economy from any increase in college graduates as well as increase in college graduation rate. The logic to finding this number requires us to know how much loan debt is being defaulted each year and add the added value of a college graduate multiplied by the absolute increase in working college graduates. I happen to know there is over a trillion dollars of debt in the states and if we assume even a conservative 3% default rate, less than a mortgage default rate, that would mean an annual cost of $30b saved. This leaves a required $5b in necessary added economic value to ensure this is a sensible investment. We know the average salary of a college graduate is in the 40k range. A non college graduate will work on average for a slightly higher than minimum wage salary of about 15$/hour or 30k per year. So assuming approximately a 10k in added value we need to see at least an increase of 50,000 college graduates entering the workforce each year. To find out this data we require the current public school graduation number and subtract it from a maximum capacity.

    2) If it does appear the the benefits outweigh the costs I’d still like to think about three other things. I call this the “important considerations”. One is macroeconomic, can the american economy absorb this increase in graduates? Second, Sustainability. It is important to consider the quality of education for long term achievement. Will schools have sufficient funding? What if they need to raise tuition? Finally, how will this affect our current system. Will top students suddenly compete for public school spots depleting private school allure? This could be bad news for our economy.

    Above, my proposal to investigating that Yes, we should indeed make tuition free.

    -Vincent, Amsterdam.

    • Victor Cheng Nov 24, 2015, 8:47 am


      You lost me one one point. You argued that reducing defaults would be a cost savings to the economy. Not really, a default isn’t a cost, it’s a financial transfer. A $100 loan to a student that the student defaults on is a $100 cost to the bank, but a $100 worth of education gain to the student. Under free tuition, the student gets $100 worth of education as a benefit, and some other part of the economy – such as tax payers – incur a cost.

      Once you lost me, I stopped considering the rest of your answer (as would most interviewers).


  • Naman Dwivedi Nov 19, 2015, 6:00 pm

    Taxing top 1% Americans(judged by net worth) with 100% on their annual income(instead of the current 50%) will cover up for the student demands i.e.
    1. Free public college
    2. Cancellation of debt
    3. $15 minimum wage for students working on campus

    -US population=300 million
    -Top 1% people(by wealth) pay 50% taxes on their annual income

    (#)The colleges don’t reduce their average intake of students from the current levels as a countermeasure to reduce costs
    (#)The demand for attending public universities doesn’t increase due to this price reduction
    Other assumptions stated as we go with (#) symbol

    1. Estimating the cost of free public college
    = No of students going to public college*Ave tuition fee per year
    a.) No of students going to public colleges
    Total population= 300 million
    (#) Age group 18-22 =10%= 30 million
    (#) % of them going to college=arnd 80% => ~25 million
    (#) Private:Public college students= 1:4 =>
    => Public college student count= 20 million
    b.)Ave annual fee= $15,000
    ∴Cost of free public college= 15,000*20million
    =$300 billion/year
    [Fact Check: No of public universities=800
    With 20 million students attending public colleges(above estimate), we get ave of 25,000 students per public university, which is a reasonable estimate indicating that the above numbers aren’t very far off.]

    2. Estimating the cost of debt cancellation
    (#) Assuming a population growth rate of 0%(for a conservative estimate & ease of calculation), 25 million students graduating every year
    (#) Approx 80% students graduate with a part of college fee as debt= 0.8*25 =20 million students
    (#) Assuming 30% as the average debt size of total college fee($15k/year*4 years), average debt of 1 student upon graduation=30%*60,000 = $18,000
    (#) Ave payoff period= 3 years
    (#) Assuming 0% discount rate(for simplicity and conservative estimate) => $6,000/year paid off by students
    ∴Pending student debt for any year= (20 million*$18,000)+(20 million*$12,000)+(20 million*$6,000) = $720 billion/year

    3. Cost of setting minimum wage to $15/hour
    (*The universities can have more control over these costs by limiting the employment hours offered for students.)
    (#) Lets assume colleges allow a max of 20% people to work => 5 million students
    (#) Assuming ave at 2 hrs/day=> 2hr/day*20day/mo*12mo/yr
    => 480 hours/year
    ∴At $15/hr, total expense= $15/hr*480hr/yr*5million students
    => $36 billion/year

    Total cost of student demands= $0.3 trillion+$0.72 trillion + $0.036 trillion = $1.056 trillion

    4. If top 1%(by wealth) in USA were charged 100% for their income, they would pay a total of:
    (#)1000 people with ave annual earning of $1billion
    (#)3 million people with ave annual earning of $400k
    These people(~1% of total population) are charged taxes at 50% of their annual income currently
    Total annual earnings=$2.2 trillion
    Taxes paid= $1.1 trillion
    ∴Additional tax revenue generated by taxing 100% income= $1.1 trillion
    =>If these people were to give up all their annual earnings, they could narrowly cover the costs for the demands made by students for free college education.

    Proposition Evaluation:
    -USA spends ~5% of the annual GDP on education which amounts to approximately $1 trillion. Increasing that by over 100%(to $2.05 trillion) would require significant compromise by either the other elements of the budget or by the US taxpayer.
    -The US taxpayer is already paying 7-15% higher taxes as compared with her counterpart in other countries. Loading them further would drive away major taxpayers to other countries leading to a decrease in overall tax revenue which could be detrimental to the economy.
    -Driving away large taxpayers would imply driving away large corporations, this would decrease employment opportunities as a result of which college graduates will suffer, leading to a domino effect and thus a downward spiral.
    -Besides this, the macroeconomic effects of such a policy could be disastrous, this would create an anti-bubble effect leading to sharp deflation rates

    With conservative estimates(assuming 100% price inelasticity both on demand and supply sides), the total additional tax revenue calculated by taxing 100% income(50% increment over current tax rate) of the top 1% Americans by wealth adds up marginally to the demands made by students. In the light of the negative effects of this policy on the economy(loss of employment opportunities and deflation, besides the discouraging impact it might have on people with a high annual income), it seems to be an extremely unfavourable proposition.

    • Victor Cheng Nov 24, 2015, 8:51 am


      Too many numerical assumption in your structure especially since a specific numerical answer was not asked for. The focus of the exercise was to explain your problem solving approach and your rationale for it; not to solve the whole case with assumptions.

      If this were a real case, you would ask the interviewer is data were available and he or she might then give you quantitative data to work with. In this exercise, the focus was what would look at to disprove your hypothesis and why? No more, no less.


  • Brian Nov 20, 2015, 1:26 am

    This is not a Case Analysis but I saw the clip and wanted to add to the discussion:

    The anchor is mistaken. The effective tax rate on the top 1% of Americans is less than half what he claims: in 2012 it was 22.8%. The effective tax rate, remarkably, becomes lower as one moves up the income percentile scale to 0.1%, 0.01%, and so on. This is due to the fact that much of the wealthiest Americans’ income comes from capital gains, not ordinary income; the anchor fails to disaggregate these two substantially different sources.

    Further, he misunderstands the causes of the Greek debt crisis. After Greece joined the Euro, foreign investment poured in, effectively in arbitrage, artificially driving >=4% GDP CAGR for several years. Greek assets became overvalued, particularly given the lesser productivity of Greek labor relative to the rest of the Eurozone. Profligate spending had less to do with it. See Krugman’s writing on this.

    I believe the student is correct that the divergence of educational attainment by income is a major problem. Longitudinal studies — the single most reliable source of causal relationships in analyses such as these — conducted by Bureau of Labor Statistics in 1997 (these take many years to conduct so the data are old) indicate that students whose parents earn in the top quartile are about twice as likely to make it to college as those whose parents do not. Note this does not account for difference in the quality of the institutions the two cohorts attend, which is not measured, and would likely further widen the gap.

    She points out that education drives innovation, and this is an extremely critical point: a more educated society drives economic growth and increases Gross Domestic Product, which increases commercial and personal incomes, which increases tax revenues, which allows us to further subsidize education. Well-implemented, it becomes a virtuous cycle (albeit with diminishing returns).

    Another benefit: diversity is critical for innovation: the way we see the world is strongly a function of our experiences, and we need many different people who see the world in many different ways if we wish to find new solutions to difficult problems.

    Yet, I am not sure free college alone is anything but a stopgap. Higher education is subject to Baumol’s Cost Disease: unlike many other industries, it takes just as many professors to teach a lecture today as it did fifty years ago. The economist Baumol theorizes that in such unscalable industries, we will inherently see costs rise faster than inflation. Coursera and the like are a step in the right direction toward scalability and wide reach, but videotaped lectures do not capture the contagious enthusiasm of a good classroom nor the incredible positive influence of a very bright group of classmates.

    America should focus more on how we “disrupt” (in the Silicon Valley sense) this behemoth industry, than on how we subsidize it.

    • Victor Cheng Nov 24, 2015, 8:54 am


      You should have been interviewed instead.


  • BK Nov 20, 2015, 4:47 am

    – Free education
    – Cancellation loans
    – A minimum wage of $15 for all people working on Campus

    It is possible to grant all students free access to a college degree.

    First of all I would make an estimate how much money would be needed to pay for all of this.

    Then I would setup a list of potential payers, which would most likely be the Government, Corporates, and single rich people.

    Starting with the Government I would analyze each group for their very likely motive to support this idea, the concrete financing means, and how they would benefit from this idea. Also relevant in this context is to evaluate the quality of the graduated student in an education system setup by Government, i.e. by the one of the other groups.

    • Victor Cheng Nov 24, 2015, 8:59 am


      You analysis plan is so broad that it’s not very practical to analyze. In the first sentence of your last paragraph you indicated you would analyze every group in the entire U.S. Government.

      An interviewer would think, that’s way too much work and how does it actually disprove the hypothesis. What’s the relationships between that analysis and the hypothesis?

      What the interviewer would say out loud is, “okay let’s go in that direction, what group would you analyze first and what specifically would you be looking for? Ok, then what group next? And then what group? And why?

      And then the integrity would end due to running out of time.

      See my comments on the other answers for some tips.


  • Siyi Nov 20, 2015, 5:14 am

    I would actually avoid going too deep into the math area If I were the girl been interviewed. My point of view in solving this problem are: 1. To Increase tax income. 2. To rearrange existing expenses allocation to prioritize education, so it can get more funds.

    For the first approach, it would ideal that all people contribute according to their abilities, which I believe people already are. The tax rates to different levels of income should be reasonable and appropriate, because it’s important to realize even some minor change in tax rates may have butterfly effects on the national economy and causing other consequences. The tax rates though may try to vary from states depending on their people’s general wealth and public opinions. Don’t just focus on the top 1%, but all of the population should take their part to make reasonable contributions.

    My main focus is second approach. Try to reallocate the money from existing tax system. In this interview particularly, the conversation would change from do you know the math, to do you know how much long term potential benefits education people in college would have to our country? Some specific steps we can take would be: Strengthen and reinforce the policy of students taking loans and paying back when they are capable. Having promotions about education’s benefits as well as organizing debates to gain more educational funds.

    Specifically relating to the questions in the interview:
    1. Don’t agree and encourage the top 1% to pay all the expenses, this is innocent and inappropriate.
    2. When mentioning about good educational welfare countries such as Greek went broke, she should realize that not all of the good educational welfare countries went broke. This means it is not reasonable to conclude that educational welfare causing the country to be broken. In reality, there are so many more comprehensive reasons that leading Greek to broken. It would be naïve to accuse educational welfare of causing those misery. There are so many other countries that going on so well while providing excellent educational support to their citizens.
    3. When he said getting all money from the top 1% cannot even cover the Medicare system of the country. I would like to ask him how did the country survived then by affording so many other items including Medicare system.
    4. About the possibilities of educational institutions raising fees. It is important to realize it has very little to do with educational support but an enforced law and comprehensive legislations which can provide effective control and regulations towards standardize school fees.

  • Daniele Nov 20, 2015, 8:31 am

    My hypothesis is that the US government will be better off by freeing students from college tuition.

    To confirm my hypothesis I also need to verify the following assumptions:
    1) More citizens get a degree every year if college tuition is free – Information needed: % of people that would go to college only if tuition was free, % of students that get a degree without paying tuitions, data from European countries where college tuition is paid by the government
    2) Graduates are more productive – Information needed: average productivity per graduate vs. non-graduate
    3) Graduates earn higher salaries – Information needed: average salary per graduate vs. non-graduate
    4) Graduates and Students without debts consume more – Information needed: average consumption per graduate vs. non-graduate, average consumption per person with debts vs. without debts
    5) Graduates are more likely to be employed – Information needed: recruitment rate per graduate vs. non-graduate

    I structure my analysis in three areas:
    1) Required investment
    2) Profitability
    3) Feasibility

    Required investment: what kind of investment is required from the US government?
    1) Cost of the initiative – Information needed: total number of college students x average pending tuition fee
    2) US government budget – Information needed: size of the Department of Education budget, % of this budget that can’t be used for this initiative, other Departments’ budgets, % of these budgets that can be moved to the Department of Education

    I therefore calculate the loan needed from foreign countries:
    Loan: Cost of the initiative – US government budget

    Profitability: is this decisions profitable for the US government?
    To analyze profitability, I explore revenues and costs.
    Revenues (per year):
    1) Taxes per year from graduates’ salaries and consumptions – Information needed: number of graduates per year – average salary – average salary taxation – average consumptions’ taxation – Assumption 1, 3,4 and 5
    2) Taxes per year from corporates that will benefit from increased productivity and consumptions – Information needed: average impact of productivity on US corporates’ revenues – volume of US corporates’ revenues – average taxation on corporates’ revenues – Assumption 1, 2 and 4
    3) Money saved per year by reducing unemployment – Information needed: unemployment rate, average government cost for each unemployed citizen, graduates’ employment rate compared to undergraduates, graduates’ crime rate compared to undergraduates’, average government cost for crime – Assumption 1 and 5

    Costs (per year):
    1) College tuition fees – Information needed: average college tuition fee, number of students per year, number of students’ growth rate in case colleges don’t require tuition fees
    2) Loan’s interest – Information needed: Loan size, interest rate per year

    Feasibility: is it realistic?
    Once we know the size of the loan and the potential profitability per year, I need to understand how much is feasible, considering:
    1) Access to the loan: is it possible to loan the desired amount? – Information needed: countries with excess of cashflow, volumes of cashflows, political agreement between US government and these governments, historical precedents, US government credit rating
    2) Time frame: after how many years does this initiative starts being profitable? – Information needed: Loan size, interest rate per year, profitability per year
    3) Political cost: how man years in a loss can a government politically afford before being deposed or not re-elected? – Information needed: years left before next election, % of supporters of the elected government, % of supporters of the initiative

    In case it is possible to get the loan to start this initiative, and it’s profitable in a realistic time frame considering the political dynamics, than my initial hypothesis is confirmed.

    I can therefore consider qualitative benefits and risks of the initative.

    1) Impact on society: benefits of having a more educated population, technological progress, lower level of stress
    2) Reputation of the governing party: benefits of building the reputation of a supportive and caring government
    3) Talents’ attraction: benefits of attracting students from all over the world, attracted by inexpensive opportunities

    1) Immigration: making education cheap might attract significantly more immigrants
    2) Implications on colleges’ brand: colleges become more accessible but less exclusive, and consequently might lose some of the best professors
    3) More competition: a more educated population increases internal competition for specialized jobs

  • Shane Sabine Nov 20, 2015, 2:11 pm

    Just a comment, Victor, that Germany – as Europe’s powerhouse – has already been successfully implementing such a model. In fact, there are even the (country-relevant) “elite” institutions that send a comparatively high percentage of students onto McKinsey, etc. Altogether, in essence, the system is (well) structured so that older, financially stable generations pay taxes towards education for those currently in school; upon graduation, this fresh source of labor is expected to do the same.

    The question is not one of feasibility, but rather of a flawed hyper-individualistic society in the US. Conversely, in Germany (and some other noteworthy European nations with higher standards of living than the US) follow the principle of sacrificing for their fellow countryman, albeit in a less militarisic sense.

    Therefore, presenting numbers and facts will do little to sway any American, especially one with power (and thus money and incentive to act according to self-interest); it is much more a matter of cultural change, openness, worldy knowledge, and introspection.

    If you want to gain some expertise on this matter, feel contact me.

    • Victor Cheng Nov 25, 2015, 2:45 am


      The US does the same thing for education from K – 12the grade. So the logic still applies and is relevant.

      While you may well be correct, consultants often have to convince clients of conclusions that they may not like or be accustomed to. It’s not easy to do.

      If it not a question of feasibility, how would you prove that point to a senior client in the US that would be in a position to implement the German model in the US?

      How would the senior client convince her board to follow the recommendation? It’s the consultants challenge.


  • Namu Nov 20, 2015, 3:58 pm

    The hypothesis of the student leader is that a free public college benefits society at large and that a higher education is a basic right; therefore, a U.S. citizen deserves a choice to attend a free public college.

    To support her argument, she will need to prove her core assumption – that free public colleges actually do benefit society.

    One quantifiable benefit to society is captured by the movements in GDP, per capital income, productivity and unemployment rate.

    She will need to make her case that the overall benefit for the majority of society outweighs the burden borne by the top 1% measured by per capital income and unemployment rate.

    How does free public college results in an increase in economic benefits?
    Her arguments could be that free public college:
    1) increases the size of the labor force
    2) improves the productivity and efficacy of the labor force
    3) which results in higher productivity for corporations at a cheaper cost
    4) which makes U.S. corporations more competitive
    5) which leads to higher revenue & tax income

    • Victor Cheng Nov 25, 2015, 2:49 am


      You opened up your structure quite eloquently. “free public colleges actually do benefit society” is exactly THE key point.

      Your supporting structure didn’t match up precisely with your opening statement. You mention GDP, per capita income, productivity, and unemployment rate as key benefits… but what specifically about them would you want to know? How would the specific answer you anticipate getting disprove your hypothesis?

      Further your opening statement talked about the benefits of free college tuition on society, but you didn’t address the COSTS to society of this proposal either.

      The key point is whether the benefit to society as a whole exceed the costs to society.


  • terbotim Nov 20, 2015, 7:20 pm

    Hypoth: it’s a good idea to make college tuition free and to forgive all student debt in the U.S.
    Structure: 4 items would need to be true to prove the hypoth.
    1) there’s more than enough money throughout the entire US population to cover the total cost of college edu + debt
    1a) what is the total cost per year to pay for every college student’s tuition. eqn1a: tot pop usa (320M) / life exp (80 yrs)= 4M people each age. assume 4 yrs for college = 16M people ages 19-22. Not everyone goes to college. assume 60% attend college = ~10M college students. now segment college cost to get total cost (including all costs: tuition, boarding, staff etc). 3 tiers high, mid, low (2M, 5M, 3M) students in each segment cost/segment/yr is (60k, 30k, 20k). Total all segments costs = 330B USD per year to pay for all US college kids tuition.
    1b) similar market sizing analysis in 1a) to calculate total college debt in the US. Add 1a+1b to get total cost.
    1c) is the total combined income of the US population per year well beyond the total cost in 1b? This would require another market sizing exercise. If total cost << total income then you’re done. Hypothesis fails. (assume both on yearly time basis).
    1d) Now that we know plenty of income is available to pay for the total cost, determine what is the appropriate share of each individuals yearly income (tax) to cover the total cost in 1b). Again, market size analysis segmenting by income tiers (low, mid, high) to determine what tax per segment would = total costs.
    2) every US citizen is happy to contribute a portion of their designated income determined in 1d)
    2a) you would need to gather data to answer the following question: the tax laid out in 1d) how much more or less (%) is that tax then what people in each tax tier pay now?
    2b) if the new tax is old tax, then you have to find out (survey or something) if these people would be willing to pay more to contribute to the cause. Or find additional resources (4).
    3) Students WANT to go to school for free!
    3a) now that college is free, we need to determine if families and students value a college edu? This may come through a nationwide survey for example.
    4) Non-tax resources: a number of successful Company/Corporations/Orgs would be happy to contribute money to cover any differences in the total cost NOT covered by the tax (assuming additional money is necessary to cover total cost).
    4a) determine which companies would be most likely to contribute: this could be answered based on a compiled list analyzing of the most profitable companies per college educated employees ($profit/no. college edu employees/yr).
    4b) take your list from 4a) and go negotiate with those companies to assess if they would like to contribute and how much. Then if total non-tax contribution = (total cost – tax contribution) then check! If non-tax contribution < (total cost – tax contribution) then this hypoth fails. You don’t have enough $$$!

    • Victor Cheng Nov 25, 2015, 2:51 am


      I got lost in the details. Too many details. Start simple and conceptual first. What’s the key issue? Make your argument for why that’s the key issue, then you can circle back later and add more detail.


  • terbotim Nov 20, 2015, 7:28 pm

    error for terbotim above. my 2b portion did not copy correctly

    2b) if the new tax is less than old tax. Check- who doesn’t want to pay less? But if new tax greater than old tax, then you have to find out (survey or something) if these people would be willing to pay more to contribute to the cause. Or find additional resources (4).

  • Lay Monica Nov 21, 2015, 5:10 am

    [Changed format]
    The idea to make college tuition free and to forgive all student debt is desirable and feasible because of several reasons:
    1. Need: Equal education is a right of citizens
    2. There are some viable options to fund them
    a. Reduce budget allocation in some areas and shift it to education (calculate optimum level)
    b. Increasing efficiency of government work
    i. Data: Key Performance Indicators of government departments
    c. Cut the unnecessary spending
    i. Data: government spending
    d. Increase the tax rate if the budget cut has reached optimum level
    i. Should be increased in reasonable way (data: the history of tax revenue and their usage in the government budget)
    (a). How much should we increase? (forecast the needs that should be fulfilled and how much would we fund)
    3. The future benefit of the decision is higher than the cost
    a. Reducing income inequality
    i. Higher social cohesion (data from empirical research)
    ii. Lower crime rate, then lower cost of security (data from country with low income inequality)
    b. Increasing the number of educated population
    i. Higher productivity (data of the correlation between education and productivity of a nation)

    The risk of the policy:
    1. In communication to the people about the increased tax rate (if the decision is to increase tax rate)
    2. Higher tax rate instills some people to go to other countries and cause some loss in tax revenue
    3. The danger of pursuing equality: some students who are capable to pay their own college tuition will be benefited as well

    • Victor Cheng Nov 25, 2015, 2:55 am


      You mention 20 factors to consider. I didn’t read them all, and an interviewer would tune out before you finished all 20. This kind of structure relies on stating everything in hopes that the you’ll cover the right issues in there somewhere.

      Interviewers aren’t looking for the largest list of factors. They are looking for the shortest list of factors that’s minimally needed to disprove the hypothesis.

      Work on being more concise and efficient with your questions. Assume each question costs your client $100,000 to answer (because often times that’s exactly how much it costs) Do you really need every question answered? Could you remove 1 question and not change your overall conclusion? How about 2 or 3?


  • Lawrence Heath Nov 21, 2015, 3:03 pm

    Using a What, Why, How format

    What: The United States relieving must relieve the student debt burden and provide free 4 year education, otherwise the inequality will increase and economic growth will continue to stagnate.

    Why: On a quantitative level 4 year public tuition/room and board costs have inflated by ~130% since 1993 according to BLS Delta Cost Project. This is unsustainable given it’s a top three driver of economic growth (healthcare, infrastructure, education.) and education has the highest correlation with social mobility. Next, the student debt levels, currently at a total of $1.3T and the 2015 class an average debt of ~$35k per student, which is the highest ever. On a qualitative level causes many students, who know they will have to take on the burden themselves to avoid pursuing higher education due to a fear of debt. This is leading to only ~ 28% of the population having a bachelor’s degree and ultimately leading to lower productivity per worker in the US. This is why we need loan forgiveness and free tuition.

    How: The common argument against public schooling is twofold: 1) how will we pay for this, so I will address this in two ways.
    1) Increased tax revenue: slight increase to public revenue through tax reform. Currently, the 1% pay an effective tax rate ~ 23%. Although this is still higher than lower earners 23% leaves room to increase. In terms of income levels, the numbers appear weak to cover entitlements, yet quite strong when analyzing net wealth. Using a wealth threshold of $8.5M from the FED and assuming 630k people in the 1% given a population of 320M and average family size of 2.5 from the US Census, the 1% holds roughly 30% of Q2 2015 GDP or $5.3T. This is a sizable sum and can be tapped into slightly to offset cost.
    2) Growth: According to the US Census Bureau 4 year degree holders earn on average 66% more than high school graduates and 42% more than 2 year degree holders. If we are able to increase the number of bachelor degree holders by ~12% to 40% from 28% this will equal ~ 38.4M more Americans with a four years degree and therefore increase their income closer to 66%. To address for the increased supply of 4 year degree holders in the labor force we can cut the 66% down by 12% (the projected increase in supply) to 27% income increase. If we set a goal of achieving a 40% 4 years education target over 10 years this conservative estimate will lead to an increase in GDP of roughly 27% or roughly $4.8T using Q2 2015 figures, again providing increases dollars to pay for the initiative.

    • Victor Cheng Nov 25, 2015, 3:00 am


      For purposes of this exercise, one was not supposed to use any data other than that what was mentioned in the video. When we structure a case, we’re looking for the critical factors that change the recommendation from one answer to another.

      What your structure is missing is it relies on actual data, but doesn’t point out the tensions between competing factors. For example in point 2 about economic growth, you mention 15 specific numerical facts. But the key insight, “if free college tuition stimulates greater GDP growth, at some level of GDP growth the economy overall if better off with free college tuition that without.”

      You mentioned details but didn’t mention explicitly the RELATIONSHIP between the factors.

      We would want to be able to move your structure and apply it to a different economy, and have the structure still be correct (even if the specific numerical facts are different).


  • Volha Nov 22, 2015, 12:39 am

    Come on!
    This is all not serious and seems to be staged.
    The question reminds of one of GMAT essay topics. Some flawed, unsupported, opportunistic claim.
    The guy is smearing the girl all over the place, which was the purpose from the beginning. There was no need to keep that conversation going nowhere, beyond the limits of common sense.
    To be very honest – I do not like that we all the time talk about the fear of being embarrassed, humiliated, looking stupid, etc. Such topics only exaggerate this fear.
    The happiness lies within the acceptance of yourself. At the end – if you are really stupid, this will show ultimately, even after passing the case interview after 1000 hours of practice.
    Some people just haven’t got the talent, that’s it.
    Let’s say that this girl on the video now starts intensely working on her case interview skills and ultimately she would master how to answer case questions. I would still not hire her as my consultant, because real life is far more complicated than just a case maths.

  • Alex Nov 22, 2015, 9:39 pm

    It’s a cost/benefit analysis – if the benefits outweigh the costs, then it’s a good idea; if not, then it’s not.

    A – Costs

    (1) Total Cost = Cost of tuition fees + Cost of student debt

    (1 -1)Cost of tuition fees = Population attending college * Average annual tuition fee

    – Population attending college =Year 01 + Year 02 + Year 03 + Year 04.
    – Year 01 = Population USA * % 20-25 year old * % college attendance * # years of college = 320 * 8% * 80% = ~ 20m students attend year 01 college every year.
    – Year 02 = Year 01 * (1- dropout rate (year 01)) = 20 * (1- 30%) = 14m
    – Year 04 = Year 03 = Year 02 (assume dropout rate 0% between those years)
    – Population attending college = 63M in any given year.
    – Average annual tuition fee (public & private) ~ $20K per year
    – Total cost of tuition fees (per year) = $1.3 trillion

    (1 -2) Cost of student debt = Year 02 debt + Year 03 debt + Year 04 debt = $280B + $280B * 2 + $280B * 3 ~ $1.7T

    So the total cost of free college tuition fees and student debt erased is $1.7T (once off) + $1.3T annually.

    (2) Economic cost

    (2-1) With a budget deficit of $440B and a Debt to GDP ratio of more than 100%, it is unlikely the government would find a way to fund an additional $1.3T or ~8% of GDP in education without seriously harming its debt profile and rating.

    (2-2) It’d have necessarily to be funded by the tax payer. If all of the 130 millions of tax payers in the US would equally contribute to the financing of this measure, it’d mean each tax payer would have to pay an additional $10,000 in taxes every year when 50% of tax payers earn less than $50K a year. It is unlikely to happen.

    B – Benefits

    (1) Benefit of added time – No debt / tuition fees means no need to work for it which leads to more time available for other things. This may lead to an increase in college take-up rate and /or a decrease in college dropout rate. However, this benefit is currently mitigated by student debt not having to be repaid until after the end of college, so the benefit would only be incremental.

    (2) Benefit of added disposable income – No debt / tuition fees means added disposable income when coming out of college. No benefit other than at the individual level since tuition fees & debt payment are already incorporated in GDB.

    Conclusion – Making tuition fees free and erasing debt for college students is not a viable measure economically: The economy does not have the strength (budget position, debt ratio, GDB growth) to finance the cost of the measure.

  • Colin McWatters Nov 23, 2015, 5:44 am

    Hi Victor,

    Thank you for putting this together! Here’s my take:

    Hypothesis: It is a good idea to make college tuition free and to forgive all student debt.

    To make things clearer, I drew out my structure in a tree diagram/flow chart format, viewable here:

    Key Assumptions for my analysis:
    – Government will cover tuition of both private and public universities, making all universities and colleges effectively public
    – The funding will cover tuition costs incurred for any student seeking a 4 year bachelor’s degree
    – All universities are of equal quality and students are just as likely to choose one over the other
    – Housing, transportation, books, and living expenses will not be covered
    – The federal government will repay federal student loan debt but not private student loans

    I break my analysis into two sections, with Part B providing the final conclusion to prove or disprove the hypothesis.

    A. Is eliminating tuition and covering federal student loans financially feasible?
    B. Does the U.S. have the capacity to handle an increased demand for college and university education?

    Part A: Is eliminating tuition from all private and public universities financially feasible?
    This question spans several perspectives, the most important are the federal government and the universities themselves.

    To prove feasibility, I must on the government side (1) ensure there is enough money earned by the government to cover its responsibilities while (2) leaving room for increased education spending; while on the university side (3) explore if additional funding sources for universities can supplement this increased government support to eliminate tuition to (4) cover university costs.

    (1.) “Revenues” to government: What is the annual revenue currently generated in each of these segments? In the case of income tax, how are taxes segmented across income brackets? Can any of these tax rates be increased?
    – Individual Income Tax
    – Corporate/Business Income/Financial Transaction Tax
    – Excise Tax/Tariffs/Import Fees
    – Estate Tax
    – Any additional sources of government funding?

    (2.) “Costs” to government: Paying off existing student loans will be an additional cost. What is the overall yearly budget of the U.S.? How is the budget currently segmented? Can any of the existing federal budget items be decreased? (This will leave more room for education spending)
    – Centers for Medicare and Medicaid Services
    – Military
    – Social Security
    – Transportation
    – Other
    – Any additional responsibilities to the government?
    – Education
    – Repayment of existing federal student debt
    – Additional finances potentially needed to build additional university capacity (Part B.)

    (3.) Revenues to our example university: can any of these be increased? (Will require less education spending from government)
    – Private Donors
    – Non-Traditional Paths (Facility Rentals, Branding, Academic publishing, Merchandise)
    – Any additional funding sources to explore?

    (4.) Costs to our example university: can any of these be decreased? (Will require less education spending from government)
    – Facilities (Dormitories, new construction projects)
    – Staff (Administrative, Maintenance, Professors)
    – Any additional costs to account for?

    Part B: Will there be capacity for this increased demand for education?
    (1.) As tuition is brought to zero, will the demand for universities increase? (Y/N?) N: Good. [Free tuition good idea]
    (2.) Y: If demand increases, by how much is it slated to increase? Is there the capacity to support this increase in students without building new universities and staffing them? (Y/N?) Y: Good. [Free tuition good idea]
    (3.) N: If no capacity, how much will this additional capacity building cost?
    – Plug this number into the federal budget estimates in Part A, Section 2. (Further squeezing additional “revenue” in Section 1, if possible).
    (4.) Is this increased cost to the government less than or equal to the “revenue” it pulls in through taxes and other sources of funding? (Y/N?) Y: Good. [Free tuition good idea] N: Hypothesis disproven, and sadly free tuition is not a good idea.

    If the earlier link to my diagrammed structure was hard to read, here’s a blown up version of Part A:

  • Thomas Nov 23, 2015, 5:51 pm

    (Please approve this comment instead of my previous submission, I have edited it for better readability)

    There are two parts to this hypothesis.

    H1) It is a good idea to forgive all current student loan debt.


    H2) It is also a good idea to make all public university tuition free.

    First we must clarify what a “good idea” means. I suggest that a program where the government takes on a cost (in this case student debt or tuition) that provides a permanent benefit to an individual and will be payed back over time by that individual is a “good idea”. For this example, an individual’s capacity to earn income might be so significantly improved that over their working career they pay back an equally greater share of taxes to replace the funds they used.

    Therefore, the two key tests that a hypothesis must pass are that:

    T1) It provides measurable and permanent benefits to the individual. (Or measure of individuals as a whole)


    T2) The method by which the benefit is funded must be sustainable.

    We can first apply these tests to H1.

    H1T1: Does the forgiveness of all current student loan debt provide a measurable and permanent benefit to each individual? (Or to the measure of individuals as a whole?)

    I suspect that this test will easily be passed by this hypothesis. To make certain of this we would need to analyze data on the effects of debt on an individual (Particularly those who are likely not able to afford an education by other means than debt.) We know for certain that by making payments on debt an individual has less income to put towards other needs such as rent and utilities an disposable income. If the data were to prove this true, and we were to agree that increasing an individual’s available income is a “measurable and permanent benefit” then this test would be passed by the first hypothesis.

    One could make the argument that the increase in income would only be a temporary benefit enjoyed by the beneficiary until such time that under current circumstances they would no longer be paying back their debt. However, since their extra income could be distributed among investment options such as a 401k or other retirement or investment portfolios, the advantage of having the funds available earlier would have cumulative effect on all future savings that could and should be considered “permanent”. Therefore this argument against the “permanence” of an advantage in an increase in available income is fairly weak.

    The second test is more complex than the first.

    H1T2: Can the forgiveness of all current student loan debt be sustainable?

    We are given some basic guidelines to further define this question. The approximate amount of current student debt outstanding is roughly one trillion USD. In addition, since the US government would presumably be the source of this funding and since the replacement of the fund would come from the increased amount paid in taxes by those who received this benefit we can phrase our question as such:

    H1T2: If the one trillion USD (Or whatever portion of that is provided by the US government) of student loans is forgiven, will that result in a net increase in paid taxes over time equal to or greater than that same amount?

    To begin to answer this question we would need a significant amount of research data. First, we would want to discover whether the elimination of debt has an effect on income. Since this is part of the first test and we considered that likely we can safely assume the same for this test. We know that elimination of debt will certainly have a positive effect on income; however, we must then prove that forgiving student loan debt of a certain size will always (or at least in aggregate) correspond to an increase in paid taxes of equal or greater size.

    Without data to prove this it would be impossible to determine that such a policy could be sustainable and the hypothesis would fail. However, if there were data to prove to the contrary, and that such a policy was in fact sustainable, then the first part of the hypothesis would have to be true.

    We must then subject H2 to the same tests.

    H2T1: Would making all public tuition free create a measureable and permanent benefit for individuals? (Or individuals measured as a whole?)

    Similarly to when testing H1, we can quickly conclude that this also will have at least a positive effect on individuals though with better data we could put it into measurable terms.

    H2T2: If the US government were to pay for all public tuition, would that result in a net increase in paid taxes over time equal to or greater than that amount?
    Like before, this is the critical test to determining the feasibility of H2. While H1 focuses on a current lump sum of student debt, this hypothesis concerns a dynamic cash flow which will likely grow into the future. The first and most pressing matter would be to define the current amount, which could be done by taking the sum of all public tuition paid by US students in a year. Once this were done, estimates should also be made for what anticipated costs will be in the future and at what rate they may grow, or potentially shrink.

    (It is likely that were this policy to be enacted, laws would need to be reviewed to regulate the rate at which public universities may increase tuition. This is because without regulation public universities would be given an incentive to artificially inflate tuition costs knowing that burden would be assumed by taxpayers.)

    After this determination, analysis would need to be done to determine the economic benefits likely to result from the policy. Particularly, into how much wages could be expected to increase and how much that will translate into an increase in paid taxes.

    (It would also be worth consideration that this policy may allow for more students to achieve a college education which may in turn result in a higher supply of college educated workers in the labor force. As best as possible, these effects should be taken into consideration when projecting the change in lifetime income and paid taxes)

    In conclusion, if BOTH hypotheses were to pass BOTH tests, then it would be clear that the argument as a whole must be true.

    My analysis of this argument focuses only on determining the truth of these hypotheses in abstraction and though it may be possible for the argument as a whole to be considered true, it may not always be wise. For example, if the immediate cost of implementing the plan were to be one trillion USD but such funding was not available, the costs of acquiring such funding may render the payback period to be longer or out of reach. In addition, there may be a plan of investment for the same one trillion USD that could potentially be proven to provide even greater benefits. Therefore it is critical to bear in mind that though the argument may be proven true by these tests, they do not determine if it is prudent or even feasible.

    It is also worth noting I do not clearly define what an “equal” repayment would be in my sustainability tests. As some student loan payment plans are zero interest it may be appropriate that “equal” repayment be without consideration of the time value of money. If that were not the case, the test could be interpreted in a more stringent manner, potentially with an expected return or an inflation estimate for the future value to be paid back. As these tests were developed in abstraction I will leave this determination up for interpretation.

    • Victor Cheng Nov 25, 2015, 2:41 am


      You noticed the key issues in this case. However, your structure communication is about 400% too long in terms of word length. The structure is to be conveyed in about 30 seconds — at most 60 seconds. Your was several minutes in length.

      For your own practice, you might consider re writing your answer in approximately 10 sentences. See if you can do it.


  • Lars Nov 23, 2015, 7:55 pm

    Well, I imagine this as a typical case interview…

    Interviewer: Should studing in the US be for free?
    Me: Hmm, that sounds interesting… Do you mind, if I ask some clarifing questions at first?
    (In my head I ask myself: Wow! Well, would Victor do…?)
    I: Sure. What would you like to know?
    Me: So, our costumer is the US goverment and they want advice if studing should be free of charge?
    I: Basically yes.
    Me: Do they just want it to be free of universtiy fees or shall also the living costs be included?
    I: No, only free of fees. The students will still need to pay for their living on their own.
    Me: Great. So, to sum this up, the US goverment does want to know, if it is possible to enable studing in the US free of tuition fees. If I remember correctly there are roughly 20 million students in the US, right? Does the goverment want to enhance these numbers?
    I: Exactly. Yes, there are about 20 million and the goverment expects an annual growth of one million. Also they want to double the number of bachelor degrees with this programm, since they estimated a lack of bachelors in the future. The budget is set to a maximum of 2% GDP.
    Me: Hmm, that’s an intersting point. The average tuition fees in the US are about 10,000$ per year per student?
    I: No, due to higher costs of the elite universites the average is like 15,000$.
    Me: Alright, so this programm would sum up to 300 billion dollars. The US GDP is about 17 trillion $, thus 2% would be 340 billion US $. Thus, the programm would be in the budget. Shall the increase of bachelor degrees be in specific field, like IT or natural science?
    I: No, the increase shall be in all fields. Okay, we will meet with the ministery delegates in 5 minutes. What do you recommend?
    Me: Alright, the goverment wants to start a programm for 2% of the GDP to enhance the future number of bachelors in all fields. To solve this issue, I would recommend, that the goverment will pay the future tuition fees for all students. This will cost about 300 billion $, which is in range of the programm’s budget. However, to motivate the students to complete their studies, the fees should only be paid in full when at least a bachelors degree is reached. Also the remaining 40 billion $ could be placed in some kind of trust deposit, since in 3 years the 2% GDP budget might no longer be sufficient due to the anticipated rise of student numbers.
    I: Thank you.

  • Madhav Goyal Nov 23, 2015, 11:45 pm

    Keely could have made her argument strong by following a structure such as below. It includes only the most important facts associated with this topic.
    It’s a good idea to make college tuition free and to forgive all student debt.
    Goals (assumed):
    1. More students attending colleges (universities included)
    2. Students finding higher paying jobs (average) out of college
    3. Higher standard of living/quality of life (for average citizens)
    We need to dig deeper into the following three things to test our hypothesis:
    1. Economics
    2. Social Impact
    3. Risks
    1. Economics: the most important component of our analysis. There are three main parts to this: Revenue, Costs and Funding:

    a) Revenue
    How much money is expected to be generated in the economy by doing this? What are the sources of this money (innovation, more business development, students investing their tuition/loan money)?

    b) Costs: This can be broken down into tuition and loans:
    How much will it cost to provide free college education nationally? We need both current and future numbers (assuming higher future college enrolments).
    What is the cost of forgiving all the current and future student debt (present value)? What is the opportunity cost of forgiving student debt?

    c) Funding:
    What are the sources of funding college education and forgiving student debt? Since we know the cost (from above), is this money available? If we change taxation of the top 1%, will they still make enough money to fund the education system? What is the opportunity cost of using this funding on education (can it be reinvested in creating more jobs, healthcare, etc.)?

    Key findings: We are looking for two main things here: Is Rev-Cost positive? Is there funding available to make education free/forgive student debt?
    2. Social Impact: Here we are asking questions to understand the social impact this change will have on the overall society:

    -Will more students attend college?
    -Is there currently a shortage of supply of college grads in the job market? Will this shortage be met through a free education system?
    -How will other publicly funded programs be affected by this?
    -Will the society’s standard of living/quality of life increase as a whole?
    -Will the gap between the rich and poor decrease? This can be a measure of annual income.

    Key findings: Answers to the above questions will show if our social goals are being met and whether the entire society is benefiting from our quest, or just the student population.
    3. Risks: How can we forget risks?

    -What are the long term implications of this change? Is this only viable in the short term? Eg. will we not have enough tax money in the future to fund free education?
    -Will this lead to economic instability in any way?
    -Will this create a supply/demand problem, i.e. too many degree holders and fewer jobs?
    -What are the political implications of this? (this one can be tricky)
    If we find that providing free education and forgiving student debt will make our economy more profitable, lead to a higher standard of living for everyone and have little negative long term impact (therefore, it’s sustainable), then it will be a good idea to make college tuition free and to forgive all student debt.

    • Victor Cheng Nov 25, 2015, 2:38 am


      Your conclusion statement was excellent. The only problem is you said it at the end, not more towards the beginning. In reading your structure, I got buried in all the detailed questions and I really didn’t grasp the big picture of what you were trying to convey until the very end.

      Most CEO clients will have tuned out by then — so your thinking is correct, but how you communicated could be improved by stating the key message earlier THEN share the detail.


      • Madhav Goyal Nov 28, 2015, 11:31 am


        Thanks for the feedback. I agree that it would be more useful for the listener/reader to see a brief summary of what we are trying to find more towards the beginning. I will add that portion and re-submit, if that’s okay! Would love to get your feedback on the new format.


  • Pallav Nov 24, 2015, 3:57 am

    Yes, the college tuition in USA should be free & also all current student debt should be waived off. I think there are 3 reasons it would be a win-win for all the stakeholders (students, college & state) :-

    1. Strong social & economic benefits that would accrue due to this decision – a. National productivity
    b. Larger skilled / semi-skilled workforce
    c. Encourage entrepreneurship
    d. More disposable income

    2. Innovative financing can take care of the state / private financial burden – a. Donor encouragement through legislative reforms / sops
    b. Coverage through tax-direct / indirect (corporate / income)
    c. Alternative revenue streams – research revenue share; asset utilization; sponsorship, etc.

    3. Size of the problem is not huge + careful implementation (possibly in phases) can help achieve objectives –
    a. Actually how many students want to avail this? – affluent Vs indebted ; private Vs public?
    b. Can the provision be partial ? for ex- x% to be waived off depending on the academic record & affordability

    4. Challenges can be overcome by addressing the following –
    a. Dedicated task-force / team comprising of all stakeholders
    b. Analyzing & learning from case example / previous success stories
    c. Studying impact & possible mitigation on building sufficient college enrollment capacity
    d. Studying impact & possible mitigation on tackling large influx of foreigners
    e. Mitigation of political risks through consensus building & on the ground campaign / communication

  • Sofiane Nov 24, 2015, 7:44 pm

    I would break down this analysis into 2 parts:
    (1) Should we make college tuition free
    (2) Should we eliminate student debt

    For the 1st part, the hypothesis is: we should make college tuition free. I order to validate this hypothesis, I would want to see that there is (a) a net positive economic impact, (b) a net positive social impact, and (c) that it can be executed in a timely manner.

    Let’s explore these issues together.

    (a) In order to assess the net economic impact, I would want to understand:
    (i)what is the cost of college tuition/student times the # of students (note that this number will increase drastically once we make college free – this calculation may be approximated by looking to see if this has happened historically or in other geographies).
    Then I would want to see (ii)how we will offset this cost. So
    (1.1) What will be the impact on current companies (e.g. higher productivity) and on
    (1.2) creation of new companies. To do this we can translate this impact into the taxes generated from these companies’ higher productivity and from new companies.
    (1.3) We should also consider the impact from having people with higher wages (note that this is an assumption and we should test to see if the value of a degree to companies will decrease and thus affect the wages of the newly-degreed companies).
    (1.4) Finally, I would want to see if we can benefit from the larger economies of scale to reduce college costs (and calculate this impact).

    If (ii) are higher than cost of making this tuition free (i), then we can validate this branch of the issue tree.

    (b) In order to assess social impact, I would want to understand and assess :
    (i)the benefits of having a better educated workforce (leading for example to better voting/political discourse, improvements to human condition, etc…
    (ii) benefit of having more cohesion in the population (as most people will have the shared collegiate experience).
    (iii)One key fact to notice is whether or not having more people go to college decrease the quality of the education

    If (i) and (ii) are higher than (iii), than we can validate this branch of the issue tree. Note that this is a very subjective category and thus probably very difficult to objectively measure.

    (c) Finally we can see if this course of execution is actionable. Can we pass the laws/regulations to make this happen? We should also try and better understand the number of people that will actually take this opportunity, and how to accomodate this number in our current schools. We can either (i) expand current schools, (ii) create new schools, or (iii) do it via an online program. Can this be accomplished (capital available, number of teachers available, real estate available, etc…). If any of these options work, then we can validate this branch as well.

    For the 2nd part, the hypothesis is that we should eliminate student debt. This can be validated via (a) a cost benefit analysis and (b) making sure this is executable in a timely manner.
    For (a), we can compare (i) the cost of eliminating current student debt vs. (ii) the increase in disposable income from those that would have had student debt (and translate this into taxes generated directly from purchase as well as indirectly through the increase in materials/products/services purchased).
    If (i) > (ii), we can validate this hypothesis if this is executable (i.e. can we pass the law to accomplish this).

  • Lukasz Nov 25, 2015, 8:35 am

    Hypothesis: it’s a good idea to make college tuition free and to forgive all student debt.

    1) Subhypotheses – this idea is good for the whole society (i.e. somebody in society wins and nobody in society loses):
    – It is good for all the students
    – It is good for all the lenders
    – It is good for all the colleges
    – It is good for all the rest of society

    2) Main assumptions:
    – As this problem regards financial area, we are looking at the financial dimension of the problem, i.e. the idea is good for somebody if this somebody does not lose any money on this idea
    – Everybody who wants to go to college in U.S. can go to every college that she or he can manage to be accepted by (thanks to the tuition loans market) – so being accepted by any particular college is not influenced by tuitions

    3) Verification of hypothesis, that it would be good for all the students

    3.1) We have the following four segments of students (segmentation):
    – Students (past or current) who are directly influenced by student debt
    — (1)The ones who already have student debt
    — (2)The ones who will have to take student debt to be able to study (nowadays or in the future)
    – Students (past or current) who are not directly influenced by student debt:
    — (3)Students who do not need student debt, because they come from rich families
    — (4)Students who do not need student debt, because they are good enough to have scholarships

    3.2) Subhypothesis: It is good idea for all the segments of students
    – To prove that we need to prove that each segment of students is not worse financially with student debt cancellation

    3.2.1) Verification of the hypothesis, that it is good for all the students:
    – Students with student debt and the students who will have to take debt – their debt disappears, so obviously, it is good financially for them,
    – Students who will have student debt or who do not need to have student debt – they do not care, probably.

    4) Verification of hypothesis, that it would be good for all the lenders

    4.1) Let’s divide lenders’ interests into two groups (segmentation):
    – Existing loans,
    – Future loans.

    4.2) Subhypothesis: It is good for all the lenders in both in scope of existing loans and in scope of future loans

    4.2.1) Regarding existing loans
    – Hypothesis: If the lenders are paid off at the fair value of the loans, then they do not care if the loans are paid or not
    – To prove that we need to find a way of paying fair value of loans to lenders.
    – Let us assume that there are two main sources of money which could be used:
    — Government – more on that below, the hypothesis is that it is beneficial in the long run for the society (economy/government) to pay off the debt of students
    — Private corporations – let us treat this source as not possible

    4.2.2) Regarding future loans
    – Hypothesis: Although the tuition loans business will disappear, the lenders will not lose profits because of new, more (or, at least, not less) profitable loans to the students
    – To prove that we need to identify possible new loan needs of the students
    — Why the students will have additional loans is described in the other part of this answer

    5) Verification of hypothesis, that it would be good for all the colleges

    5.1) Subhypothesis:
    It is good for all the colleges financially (i.e. it does not lower the income of colleges, because the lower cost from tuition can be offset by other income sources)

    5.2) Working assumptions:

    5.2.1) We have the following 5 segments of college income (segmentation):
    — (1)Tuitions
    — (2)Endowments and foundations
    — (3)Public support
    — (4)Contributions from graduates
    — (5)Contributions from families of students with poor results from high school exams
    5.2.2) The cost connected with tuition management is neglible

    5.3) Verification of the hypothesis, that it does not lower significantly the income of colleges, because the lower cost from tuition can be offset by other income sources:
    – Let us look at each of the income sources:
    — (1)Tuitions – this income source will disappear
    — Need to verify total value of the lost income from this source – yearly
    — (2)Endowments and foundations – this income source is not influenced by lack of tuitions
    — (3)Public support – this income source could probably compensate for lower tuitions
    — More on that below, the hypothesis is that it is beneficial in the long run for the society (economy/government) to pay off the debt of students
    — (4)Contributions from graduates – this income source could be positively influenced by lack of tuitions
    — More on that below, the hypothesis is that it is beneficial in the long run for the society (economy/government) to pay off the debt of students
    — (5)Contributions from families of students with poor results from high school exams
    — This income source is not influenced by lack of tuitions

    6) Verification of the hypothesis, that it is good for all the rest of society financially:

    6.1) Subhypothesis: Society (economy) can benefit financially from the fact, that students do not have to pay off their debt after graduation, so the society (i.e. government) will pay the existing student loans

    6.2.1) To prove that we need to show that the students with no debt after college add significantly more in taxes to the budget and more in interest to the loan companies – this would be a reason why the government should pay the student loans and why the government should support universities more
    – This could result from the reason, that the students with no debt after studies can possibly be more risk-oriented than students with high debt, so they start companies more often and they go to less safe jobs

    6.2.2) Moreover, we can also prove that the students with no debt after college cause less cost to the budget
    – This could e.g. result from the fact, that the students with debt after studies could possibly be less happy and more stressed, what can lead to illnesses, addictions or wrong-doings

    7) So, to prove the beginning hypothesis, we need to test:

    7.1) What is the income from taxes from the people who:
    – Were students with debt after studies
    – Were students with no debt after studies

    7.1) What is the income for lenders from businesses started by people who:
    – Were students with debt after studies
    – Were students with no debt after studies

    7.3) What is the average cost of support of government for the people who:
    – Were students with debt after studies
    – Were students with no debt after studies

    7.4) What is the yearly income from tuitions for the university

    7.5) What is the yearly income from student loans for the lenders

    7.6) Let’s say, that we should use 20 years of data, and also possibly include trends

    7.7) The beginning hypothesis will be proven if:
    7.7.1) From perspective of government:
    The difference of income from taxes of the students for whom the tuition will be paid (in comparison to the scenario that the government does not pay off the tuitions) for the government is greater than cost of all the existing tuition debt and yearly cost of support to the universities for 20 next years
    – Possibly, also other financial factors could be included, like cost of treatment of students with high debt – if this cost is material
    7.7.2) From perspective of tuition lenders businesses:
    The difference of income from loans to businesses / initiatives of the students for whom the tuition will be paid (in comparison to the scenario that the government does not pay off the tuitions) is greater than profits of the tuition debt business
    7.7.3.) From perspective of colleges:
    The financial support from the government (and, also, possibly from higher contributions from graduates) has to be not less than expected income from tuitions

  • hyder Nov 25, 2015, 9:15 am

    Hi Victor,

    It is a great idea to make a contest, but this is a bad application for three reasons;
    First, it is not time efficient. You have to spend a lot of time. The total time required is equal to time spent per review times the #contributors. I am expecting 5 min for each review and %2-3 contribution of 10K. 5min * 300 reviews are 25 hours.
    I am considering you are a busy person so you won’t spend that much of time for this contest.
    Second, it is not cost efficient, because it’s free. You need motivation for evaluation of cases. Also freeness boosts the # contributors and decreases the quality of the writings. As the # contributor increases, the time you spend is also increases.
    Third, it is not a real contest, because contributors see others’ answers. They should be hidden from contributors.
    On the other hand, I guess main goal of is to help people to get better in case interviews and to make money. For this reason, I suggest you to start a new branch about case interview reviews for three reasons;
    Firstly, as your tips and hints are not very useful without practice; also practice is not very helpful without evaluation. People need reviews & feedbacks and I think they will be willing to pay and that is convenient with the two main goal of this website.
    Secondly, it is both time and cost efficient. By charging money, you would reduce the # contributors and the time you spent. Also it increase the quality of writings this will reduce time wasting.
    Thirdly, contest means competition and it pushes people to do better. So members would like to compete to do better in case interviews which is also convenient with the goal of
    Because of these reasons I suggest you to start a new business about case interview reviews. Furthermore you may consider hiring people for evaluation, and set prizes like live case interview with winner etc. The first contest subject may be, “should I enter in Case interview review contest business” :).

    Great fan of your work, and really appreciate.

  • Ningning Zuo Nov 27, 2015, 5:01 am

    The Hypothesis is: it is a good idea to make college tuition free and to forgive all student debt.

    There are three groups of people who will be impacted by this policy.
    1) Students who have graduated from college and are still repaying the loan back;
    2) College students who are taking loans and still need to pay their tuition;
    3) Prospect college students who are yet to choose which college to enroll in.

    Here is the structure and factual data I will need to test sub-hypothesis that students without loans outperform their peers with loans in every phase of their lives.

    1) Students who have graduated from college with and without loans
    a. Average salary level comparison
    b. Career progression comparison
    c. Drop-out rate from workforce comparison
    d. Achievement at work comparison
    2) Students who are still in college with and without loans
    a. Academic achievement comparison e.g. GPA
    b. Replacement rate comparison for fourth-year students
    c. Drop-out rate comparison
    d. Suicide rate comparison
    3) Prospect college students with and without financial support
    a. Numbers of students who chose not to go to college because of cost
    b. Number of students who chose college of less quality because of cost
    c. Average income level of students who went to top-tier colleges
    d. Average income level of student who went to second-tier colleges
    e. Average income level of students who didn’t go to colleges
    f. Impact on GDP because of the delta above

  • AJ Nov 27, 2015, 12:10 pm

    I will tackle her key demand of making college education free:

    I will do this assessment from the point of the college as a business problem.

    From the point of view of the college, my hypothesis is that this is a revenue problem, ie a significant decline in revenue for the college which impacts the current profit of the college.

    A college typically earns “Revenues” from “student-sourced revenues” (SSR) Eg: fees paid by the students(tuition, boarding, extra curriculars,etc), or “Non-student sourced revenues” (NSSR) consultancy fees, return on investments made, patent fees, licensing fees, donations, grants/loans from the government, dividends on investments and so on.

    A key data point need here would be to understand the ratio of NSSR to SSR. I would also look a the % of colleges where the above ratio is greater than 0.5 ie. where NSSR is significant. Where the ratio is high, there may be an opportunity to further increase NSSR to offset the loss of SSR.

    A college incurs a certain “Cost” to provide the service of education at the required quality & standard. There will be both a fixed component and variable component to the Costs. These include salary of teachers & administrative staff, land/building rental/maintenance, equipment, water/electricity, loan interest fees, etc.

    In order to solve the problem of making college education free for students while assuming the college wants to maintain its existing profitability, the college needs to offset the loss of SSR by:

    1. increasing NSSR
    2. identifying new NSSR options
    3. decreasing its “Cost”.

    Lets look at Revenue:
    Since revenue is Volume x Price, I would check if there is viable option for the college to :
    1. Increase the volume on the existing NSSR.
    2. Increase the “price charged” on the existing NSSR
    3. Add more sources of NSSR
    4. Increase the frequency of receiving NSSR options.

    Data required for further analysis would include:
    Look at Colleges with a high NSSR to SSR ratio and identify whether:
    1. Are are opportunities for additional revenue in a college (or re-application opportunities from other colleges) – either on volume or pricing.
    2. Can the frequency of revenue generation be increased for any of the current “non-student revenue” sources

    Assuming there are still a majority of colleges, where NSSR to SSR ratio cannot be increased to meaningfully offset the loss of SSR, I would look at Cost reduction options.

    Considering fixed costs like rentals, maintenance, etc. I would look for opportunities on these costs vs other colleges in the same state/district in order identify opportunities to reduce costs at parity to more efficiently run colleges. I would also explore options to move courses online – to minimize fixed costs spends on infrastructure.

    On variable costs, I would look to optimize student:teacher ratio, restructure teacher salaries and optimize administrative staff.

    I would also check if there are options to consolidate certain services across colleges in a state/district to get some cost benefits. Example: Having a consolidated college administration centre which acts a single back office to large number of colleges or having a maintenance contract with a single service provider for multiple colleges.

    There will always be a set of colleges where the NSSR cannot be increased or cost sufficiently reduced to maintain the existing profitability of these colleges. For this category of colleges, I would look at alternate strategies like increased Government grants/interest free loans/full funding. The assumption that’s made here is that the medium to long term benefit to society of having well educated citizens exceeds the costs incurred by the Government to provide free education.
    I would seek data to understand whether there are proven models that support this option.

  • Madhav Goyal Nov 28, 2015, 11:41 am

    Keely could have made her argument strong by following a structure such as below. It includes only the most important facts associated with this topic.
    It’s a good idea to make college tuition free and to forgive all student debt.
    In order to test our hypothesis, we need to dig deeper into the following three things to test our hypothesis:
    1. Economics
    2. Social Impact
    3. Risks
    What are we trying to find?
    By testing the above three points, we are looking to prove that the net long-term benefit to society from making college tuition free and forgiving all student debt is positive. Specifically, this change will make our economy more profitable, lead to a higher average standard of living and is a sustainable change that will not negatively affect other social programs.
    Now that we have our major test points and goals in mind, let’s start asking questions:

    1. Economics: the most important component of our analysis. There are three main parts to this: Revenue, Costs and Funding:

    a) Revenue
    How much money is expected to be generated in the economy by doing this? What are the sources of this money (innovation, more business development, students investing their tuition/loan money)?

    b) Costs: This can be broken down into tuition and loans:
    How much will it cost to provide free college education nationally? We need both current and future numbers (assuming higher future college enrollments).
    What is the cost of forgiving all the current and future student debt (present value)? What is the opportunity cost of forgiving student debt?

    c) Funding:
    What are the sources of funding college education and forgiving student debt? Since we know the cost (from above), is this money available? If we change taxation of the top 1%, will they still make enough money to fund the education system? What is the opportunity cost of using this funding on education (can it be reinvested in creating more jobs, healthcare, etc.)?

    Key findings: We are looking for two main things here: Is Rev-Cost positive? Is there funding available to make education free/forgive student debt?
    2. Social Impact: Here we are asking questions to understand the social impact this change will have on the overall society:

    -Will more students attend college?
    -Is there currently a shortage of supply of college grads in the job market? Will this shortage be met through a free education system?
    -How will other publicly funded programs be affected by this?
    -Will the society’s standard of living/quality of life increase as a whole?
    -Will the gap between the rich and poor decrease? This can be a measure of annual income.

    Key findings: Answers to the above questions will show if our social goals are being met and whether the entire society is benefiting from our quest, or just the student population.
    3. Risks: How can we forget risks?

    -What are the long term implications of this change? Is this only viable in the short term? Eg. will we not have enough tax money in the future to fund free education?
    -Will this lead to economic instability in any way?
    -Will this create a supply/demand problem, i.e. too many degree holders and fewer jobs?
    -What are the political implications of this? (this one can be tricky)
    Depending on the answers to the questions above, we can prove or disprove our hypothesis. If we find that providing free education and forgiving student debt will make our economy more profitable, lead to a higher standard of living for everyone and have little negative long term impact (therefore, it’s sustainable), then it will be a good idea to make college tuition free and to forgive all student debt.

  • John Dec 2, 2015, 10:01 am

    Hi Victor,

    I sent in a response to, as I didn’t realize that you wanted answers to be posted as comments.

    I sent the email November 19. Did you evaluate my structure as well?


Leave a Comment