Where's the data on the emergencies themselves? Everyone collects ultraprecise data on the performance of financial assets...
"Data from Shiller (2013) was used to measure stock (S&P 500) returns,
while Ibbotson Associates (2011) data, obtained from Morningstar, were used for the bond returns (long-term Treasuries)..."
...and then shrugs and uses anecdotes and handwaving guesses about the contingencies those assets are supposed to meet.
As nearly as I can tell, unemployment is the only specific income shock
they collected data on, and they don't use data on expense shocks at all.
It is true that "Rising unemployment necessitates having an emergency fund in place (Bureau of Labor Statistics 2011)" and yet unemployment is one of the few expense shocks against which many workers are automatically insured. Other income and expense shocks may be less common, but also be exactly the ones for which an emergency reserve is needed.
"Unemployment data was also obtained from the U.S. Bureau of Labor statistics (2011).
The time period for the return measures was 1926 through 2011, whereas unemployment data ranged from 1948 through 2011."
Well, what's wrong with this
"Emergencies may not happen as frequently as imagined....
...Well, yeah, if you assume that 1929-1940 was a figment of the imagination
, than unemployment emergencies may not happen as frequently as "imagined."
It's not like this data is impossible to find, you know:
And it's Wikipedia, and it cites its sources.
(What's particularly odd is that Wikipedia says it was able to find BLS data back to 1940, yet Scott & al. see fit to go back only to 1948).
"...which means there may be an opportunity cost associated with holding too much cash."
Shouldn't the consequences
of not being able to meet an emergency be assessed? Where in this paper do we see the consequences being assessed? I see "consequences of unqualified withdrawals from tax-advantaged accounts," I see "consequences of asset location for all goals," I don't see "consequences of not being able to meet an emergency."
"A noted problem with relying on borrowing in times of financial distress is that even previously established and collateralized credit lines may be called in times of broad economic uncertainty." The thing that really happened becomes a hypothetical.
Why do they say "may" when it was a "were?"
"There may be an opportunity cost associated with holding too much cash." Of course. But so what? There is a cost in building the Brooklyn Bridge with cables six times strong as needed, but Roebling was glad he had when he discovered that a dishonest contractor had supplied cables that were weaker than they were supposed to be. There is a cost to building an airliner whose wings can bend this far without breaking. There is always a cost to safety, and you can always pocket more if you cut out the safety.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.