Thread by @michaelstepner

Hi, I'm Michael! You might know me as the author of binscatter and maptile. But these days, I'd prefer you know me as the author of "The Insurance Value of Redistributive Taxes and Transfers".

…yes, I'd love to tell you about my job market paper.👇🏼
michaelstepner.com/jmp/

People face a lot of income risk during their lives. Layoffs and illness at working ages cause long-term earnings declines. So what insures people against those risks?

Unemployment and disability benefits provide social insurance, of course. But they're less than half the story.

Progressive taxes and transfers don't only redistribute from high-income people to low-income people. They also redistribute from high-income years to low-income years.

In other words, redistributive taxes and transfers provide insurance. A lot of insurance, as my JMP shows...

I estimate how the tax and transfer schedule insures 1.6 million layoffs and 1.2 million hospitalizations linked to Canadian tax records.

Matching each person who experienced a shock to someone nearly identical who wasn’t laid off or hospitalized in the same year, I find...

Result 1: Layoffs and hospitalizations do indeed cause long-term earnings losses. 12 years later, average earnings remain thousands of $$ lower and are not recovering. (I cite many papers with similar findings!)

So, how does the government insure those risks?

Result 2: The majority of social insurance after a layoff or a hospitalization comes from redistributive transfers (which go up as income falls) and progressive taxes (which go down as income falls).

Their insurance value becomes more interesting if we dig past the average...

Result 3: Redistributive transfers are especially important for reducing the risk of catastrophic income losses. More than UI or DI, they dampen the probability of losing more than half one's income after a layoff or a hospitalization.

Result 4: People who already have low incomes lose a disproportionate share of their income following a layoff or a hospitalization. But after taxes and transfers, the losses are much more level across the income distribution.

That insurance value is itself progressive!

Result 5: The insurance value of the tax and transfer system against the income risks of layoffs and hospitalizations *alone* exceeds $1000 annually in the bottom income quintile, and exceeds $400 annually at all income levels.

Bonus result! I estimate an *upper bound* on the total insurance value of the tax and transfer system against all income risk, not just layoffs and hospitalizations. It’s about 4 times larger: over $4000/year at low incomes, under $2000/year at high incomes.

🎉Two important implications🎉 First: By insuring income risk, progressive tax and transfer schedules provide more value and a higher degree of progressivity than their statutory rates suggest.

🚨Second implication🚨 The large share of total social insurance provided by redistributive taxes and transfers means that the optimal levels of unemployment and disability insurance benefits are highly sensitive to changes in redistributive policy.

If you're curious about the details in >280 character chunks, you can explore the interactive charts on my website. And if you REALLY enjoy words and details, I have a PDF file just waiting for you 👇🏼

michaelstepner.com/jmp/

And if a live conversation is your favorite way to discuss research, I'd be delighted to tell you all about this paper and my other work at the AEA interviews in San Diego or at the CEEE interviews in Toronto! 😃