I’ve been getting a lot of advertisements touting the salary increases from getting an MBA or a Masters in something else business related. But with any adverisment, is it really worth it?
I decided to look at this from two angles:
You want to stay an individual contributor
You want to move into management
Is it worth it if you don’t want to be a manager?
I don’t doubt that a lot of people get a bump from it, but I think that some of the people getting a bump from an MBA are transitioning from being individual contributors like analysts or data scientists to becoming people managers. As a result for people who don’t want to do that, I decided to investigate if it was still worth by using data from the American Communities Survey (2013-2017) and comparing the income distributions for individuals in the same occupation and age group (30-40 years old) by their education either a bachelors degree (any field) or a masters degree (ex MA, MS, MEng, MEd, MSW, MBA).
For analysts of different types, you can see that there are median increases to wages ranging from 10% to 47%. Compare this to the nominal value of this translates to between $7k to $26k per year depending on the occupation.
Compare this to the cost of a MBA’s or other related masters programs which comes to $50k to $150k to complete a full course depending on the quality of the program. Since we’re comparing the median graduates before and after, you’ll have to use your own judgement into your calculations. These increases aren’t huge if the program has a list price of $50k, but you’re going to only expect to bump up $15k in the case of an Operations Research analyst. Some quick math says it will take at least 3.3 years to make back that amount without taking into account any tax implications or opportunity cost. In fact, one important thing to note though is that the 75th percentiles for masters degree holders is much further out than for BS degree holders so your ceiling will probably increase as well.
Is it worth it if you want to move into management?
It’s hard to say if people get MBA’s and become managers or if the people who get MBA’s want to be managers and already have the qualities needed to do that but just need some extra qualifications to make it.
Using the same data table we can see that the portion of managers in these areas who have masters degrees is substantial at 19% compared to 11% for the working population aged 30-40 years old, suggesting that it helps if you get a masters degree but it isn’t a requirement in all management areas.
Most noteworthy, 3 fields stood out with very elevated levels of masters and doctoral degree holders in management and those were:
Architectural and Engineering Managers (36% Masters, 5% Other Advanced Degree) : Supervise engineers and architects.
Natural Science Managers (39% Masters, 17% Other Advanced Degree) : Supervise the work of researchers, scientists, chemists, physicists, and biologists.
Social and Community Service Managers (35% Masters, 5% Other Advanced Degree) : Many social workers get more advanced degrees and coursework that overlaps with health.
Therefore if you work in one of those areas, you may find more benefit than the median person.
Analyst Occupations by Salary and Education Level
Operations Research Analysts
Operations Research Analysts
Proportion of Workers By Education Level
Other Advanced Degree
Less than BS
Agents and Business Managers
of Artists, Performers, and Athletes
Architectural and Engineering
Chief executives and
Computer and Information
Farmers, Ranchers, and Other
Food Service and Lodging
General and Operations
Human Resources Managers
Managers in Marketing,
Advertising, and Public Relations
Managers, nec (including
Medical and Health Services
Natural Science Managers
Other Business Operations and
Property, Real Estate, and
Community Association Managers
Social and Community Service
Transportation, Storage, and
I pulled the data from the raw American Communities Survey data for 2013-2017 and restricted to individuals between the ages of 30-40. The average MBA student is 28, so I decided to focus on the age range after where most of the people would have graduated. After extracting the data, I ran some summaries to calculate the income percentiles and educational distributions.
I don’t normally go into politics, but this topic seems to touch a little more closely to FIRE, since there are specific growth rate assumptions that were made.
Buried under the headlines over how much Jeff Bezos would be taxed under Elizabeth Warrens Medicare for All plan was a side note on an additional tax to target the top 1% of households by net worth for a new tax called “mark to market.” This tax basically would restructure capital gains to take place every year regardless if there was a sale. Warren’s plans referenced a study freely available, if you’d like to read it yourself. The researchers used the Survey of Consumer Finances by the Federal Reserve, which is what powers many of the comparison calculators on this website.
To be in the top 1% by using total net worth, you would have to have $10.350 million in net worth to be in the top 1% of households based on the last survey results available using 2016 data. The authors don’t mention if they adjusted out any assets from their projections, so I’m going to assume not. The paper referenced by Elizabeth Warren referenced uses a rather optimistic 8.33% return on investments (page 30 of the PDF).
Over estimated returns
Vanguard is projecting a average annual returns of only 4-6% for US equities. 86% of the top 1% are over the age of 50 so their asset allocations are most likely more conservative than the overall stock market, and would likely grow much slower than the 8.33%. So with those 2 factors combined, most likely the $2T incremental tax contribution is very over stated especially considering the effects of lower compounding growth.
Shortfalls will need to be filled with other taxes
Using the 8.33% growth over a decade you would end up with 130% cumulative growth, but with 4-6% over the decade the growth would only be 49%-82% so the taxable gains would be 37% to 69% of the budgeted $2T which would cause a short fall in taxes of $0.6T to $1.3T.
Months ago, Warren and others had mentioned only targeting the wealthiest of the wealthy for the incremental taxes to pay for Medicare for All among other plans. The first primaries are still months away and Warren’s tax plans have started to dig further and further down the wealth ranks. Sure, people with $10m+ dollars in net worth are probably going to be fine. But if there are shortfall like the one I mentioned, where is the incremental money going to come from? A tax on $1m+ net worth households (top 10% by net worth) doesn’t look as implausible any more as the standards of who should be taxed keeps falling and falling.
It’s a common refrain that law salaries for new grads are bi-modal, with a large percentage of law graduates going into Big Law. The NALP even publishes data that shows a double humped distribution of salaries every year. But a quick query of freely available American Communities Survey data (collected by the US Government) shows that there is a small hump in the same spot that the NALP survey reports. But NALP shows nearly 21% of graduates with income around $185k / year and it looks like roughly 26%+ are earning at least $160k which looks to be a huge overstatement. This is what I found:
Only 16% Reported Income Near that Range
Mean : $91,798 / year
Median : $75,000 / year
25th Percentile : $51,771 / year
75th Percentile : $119,976 / year
I have read comments questioning some of the data collection methods that NALP uses, and I think that this single slice of data definitely shows that there is an odd deviation from the results I derived from the ACS. If anything my data should skew higher, because I could not separate people with only 1 year of experience out. So there would be associates with in Big Law with multiple years of raises and experience in this data set. An interesting overlap is that the NALP’s adjusted mean is slightly below the mean salary that I have calculated, which is interesting as to how close it is.
ACS 5 Year 2013-2017 Micro-data
Industry : Legal services so this would be limited to law firms specifically.
Ages 25-28 to try to capture new grads specifically with at least one year of work, there was no way to restrict years of experience.
Occupation : Lawyers, and judges, magistrates, and other judicial workers (This was the lowest level, I could separate it at, but I doubt there are a lot of Judges/Magistrates/Other in this age range)
Education: Professional degree beyond a bachelor’s degree (Could not be more specific)
Weeks worked prior year: >40+ weeks / year (to try to exclude people who were reporting income from partial year internships/clerkships)
Hours Worked in a usual Week : >35+ hours / week (to try to get only full time)
Here are the top 20 metro areas with the largest nominal gaps between the median monthly cost of home owners with mortgages versus home owners with out mortgages.
Median Costs by Metro Area
Dark Blue is Median costs of Home-owners without Mortgages.
Light Blue is Median costs of Home-owners with Mortgages.
Dark Line is Median costs of ALL Home-owners.
Grey $ Amount is the difference between a median Home-owner with a mortgage versus one without.
Hover over or click the bars and lines for a tooltip.
Nationally the median monthly costs are $1,516 (with) versus $474 (without) or a 3.19 ratio. But for the cities at the top of this ranking, the ratio is much higher at 4.35. It seems in SF/SJ if you bought property a long time ago it’s very cheap to hold onto, compared to someone buying more recently with debt.
Californian cities rank very prominently on the list, and this may be due to their laws with restrict property tax increases to long-time home owners, who are also the same people who are most likely to have no-mortgages. I did a quick check of who these people are for the San Francisco Metro area, and the median age of a home owner with no mortgage is 67–practically retired. The median age of a home owner with a mortgage is a bit lower at 52. Comparably expensive areas like NYC and Connecticut have smaller cost gaps between home-owners of different mortgage statuses.
I had originally built this list on the ratio of the costs of HO with mortgages vs HO without, and the chart was pretty much all Californian and Coloradan cities.
Want to see more cities?
If you are interested in more city level information, you can find this data on the City Housing Cost Percentile page. It shows a distribution for home-owners by mortgage status, renters, and the overall housing cost distributions by metro area. The national variation is listed on the National Housing Cost Percentile page. The percentiles at the upper end of the distributions are most likely under-estimates, but that was the data that was available to me
Methodology and Calculations
The data is based on the microdata previously mentioned for each metro area. These results are based on microdata from the 2013-2017 American Community Survey. Housing costs include mortgages, utilities, monthly fees like HOA’s, insurance and property taxes; this is to match the Selected Monthly Owner Costs metric that the ACS publishes. This was calculated in R and exported for these graphs. Utilities are included because some homes roll those costs into different buckets (e.g. HOA), so it is more of an apples to apples comparison to include utilities and all normal fees related directly to the home.
If you would like to compare the results by metro area, you can go to the American Fact Finder tool and select specific geographies to compare the results. There may be small variances between the ones reported and the ones I calculated. These variations may be due to the function in R that did the weighted median / percentile groupings as well as data that was not released publicly by the ACS. In the example to the right for the top 3 cities, the comparison to the FactFinder had almost the same numbers, so I am fairly confident that the remainder of the results should be representative. The All Home-owner metric is not available as a pre-calculated number to check against. But, because the other 2 numbers we can verify are good, most likely these are relatively close as well.