Why Boom Times Kill

It seems that the puzzle of why death rates rise in good economic times is nearly solved. There’s an effect of increased driving deaths from increased driving, but the main effect is that in good times nursing homes have to compete more for minimum wage nursing assistants. Apparently a one percentage point cut in the unemployment rate leads to three percent fewer nursing assistants, which increases the national death rate by a half percent (which cuts about three weeks of life per person):

We find that most of the additional deaths that occur during times of economic growth are among the elderly, particularly elderly women. … Cyclicality is especially strong for deaths occurring in nursing homes, and is stronger in states where a higher fraction of the elderly reside in nursing homes. … Staffing in skilled nursing facilities moves counter-cyclically.

A typical estimate suggests that a one-percentage point increase in a state’s unemployment rate leads to a 0.54% reduction in that state’s mortality rate. … Deaths by motor vehicle accidents are associated with the largest coefficient estimate. … It is likely that motor vehicle deaths fluctuate because people drive more during strong economic times. …

Approximately 80 percent of the averted respiratory deaths are among those over age 60. … Virtually all of the additional cardiovascular deaths are among those over age 65. … The correlation between changes in hospital employment and changes in aggregate employment is strongly negative (-0.90). … Nursing homes experience especially severe shortages of nursing aides when the economy is strong. … Between 70 to 90% of home health care agencies and nursing homes indicate shortages of direct care workers. …
Nursing home deaths are associated with an estimated [unemployment rate] coefficient that is an order of magnitude larger than the coefficient that is estimated among deaths taking place elsewhere. …

A one percentage point increase in the unemployment rate raises total full-time employment at skilled nursing facilities by approximately three percent. There is no statistically significant increase in the number of physicians, but there are significant increases in nurses, certified aides, and other occupations. (more)

A quick calculation says the US paid ~$13 billion for nursing assistant salaries in 2004, less than one percent of US medical spending (source). By cutting nursing assistants 3% when unemployment falls by 1% (and cutting three weeks off US lifespan), we save ~$400 million a year, or one part in 5000 of US medical spending.

Given how cheap they are, it seems inexcusable that we don’t raise wages on nursing assistants in boom times, to keep nursing homes fully staffed. Might this be due to a fixed government price that refuses to adapt to the business cycle?

Added 23Dec: See the comment here suggesting that med understaffing is chronic and makes a big difference.

GD Star Rating
Tagged as:
Trackback URL:
  • Robert Koslover

    You say:

    “…it seems inexcusable that we don’t raise wages on nursing assistants in boom times…”

    To whom are you referring by “we”?
    (a) Nursing home owners/managers/stockholders?
    (b) Government officials?
    (c) Customers/relatives of the customers?

  • > Might this be due to a fixed government price that refuses to adapt to the business cycle?

    I am surprised you didn’t jump immediately to the status argument: nursing home assistants are very low-status on the medical hierarchy, and so there is no political will to help them in any economic form, much less a cyclical form. (I recall locally a case where such assistants – most foreign, incidentally – attempted to strike for better wages and were immediately haled into court.)

    • lemmy caution

      I agree about this. There really is a type of “natural wage” theory that makes companies reluctant to raise the wages of low status workers.

      • KPres

        lemmy, if that were the case, then who’s bidding up the price of the nurses? They couldn’t be demanding more money unless some other industry was tempting them with better offers.

      • michael vassar

        Agreed. I think individuals have this problem too. People would rather pay a housekeeper a low hourly wage than pay a higher hourly wage for a more skilled or energetic housekeeper who makes the house equally clean for the same cost.

      • lemmy caution

        “who’s bidding up the price of the nurses? ”

        Certainly not nursing homes.

      • I think it’s a productivity problem. There’s just no way for them to be very productive in this kind of work.

        I like the Japanese solution: robots!

  • I think we should address the point that the death rate doesn’t differentiate between nursing home patients who die one year early because of sub-optimal care and young adults who die decades early because of a car crash.

    • So? You really think looking at QALYs is going to change much? We already spend a ton on automotive safety, the impact of another dollar on the margin is epsilon. On the other hand, it’s very easy to raise nurse/aid salary.

      • Moving from deaths, or persons lost, to person-years of life lost changes things by a factor of, say, 30.

        Moving to quality-adjusted life years could change things by many, many orders of magnitude, depending on how low we rate the quality of life of those who might die a year early from sub-optimal nursing care. That quality could be close to zero or even negative — especially when we take into account the tendency to show that we care, rather than to let people go.

      • Michael Wengler

        QALY seems relevant. The differential deaths are occurring primarily among, essentially, permanently hospitalized people. Now I am amused by the rationalist orthodoxy that living forever for the individual must be a Good Thing ™. But is it really? And is it as good a thing as living when you are not in a nursing home?

        Economic theory would suggest these aides who are taken to other occupations during a good economy are indeed doing something mroe valuable in that economy. Maybe extra weeks in a wheelchair or a bed is not the highest and best return on low wage work.

      • Douglas Knight

        $1 for 3 weeks of life, even in a nursing home? sounds like a bargain to me.

      • Douglas Knight

        $1 for 3 weeks is on par with the best interventions in the developing world, though the QALYs probably aren’t as good.

      • Right, but who pays the $1 and who gets the 3 weeks?

  • KPres

    Wow! Talk about missing the point people.

    As Robert mentioned…

    “Might this be due to a fixed government price that refuses to adapt to the business cycle?”

    In other words, boom times aren’t boom times for the nursing home business because their revenues are largely constrained by government payments, which don’t increase with the cycle. So while the boom raises the price of nurses, the nursing homes themselves haven’t got any more money to pay them.

    • Wonks Anonymous

      Robin: “we don’t raise wages on nursing assistants in boom times”
      KPres: “the boom raises the price of nurses”
      Sounds like the two of you have a disagreement. I can’t read the paper Robin linked, but the quoted bit said nothing about the behavior of nurse wages.

      • How is that a disagreement? Boom times raise the amount employers are willing to pay for scarcer labor, which means nurses face greater opportunity costs than in bust times, which means the wage they will demand (in exchange for eating the opportunity cost) will rise. As nurses’ employers are apparently constrained in what they can offer, the employers will not offer enough to offset that rise, and hence, supply will decrease (fewer nurses will be employed as nurses).

      • Wonks Anonymous

        I guess I’m so used to thinking of employees (particularly of nurse status) as price-takers that I interpret price to refer to an employer’s demand.

  • [i]”It seems that the puzzle of why death rates rise in good economic times is nearly solved.” … “Given how cheap they are, it seems inexcusable that we don’t raise wages on nursing assistants in boom times, to keep nursing homes fully staffed. Might this be due to …”[/i]

    Excuse my ignorance and lack of link-spelunking, but if the puzzle is only just now [nearly] solved, why suprise that policy hasn’t already adapted to this solution?

    • Khoth

      It seems like a continuation of a pattern that’s reasonably common in Robin’s arguments:

      1. Here’s a dubious-looking argument I just invented that A implies not B.
      2. Therefore, everyone who believes A and B is a hypocrite.

  • carl213

    What is the exact cause of the deaths?
    Is it due to acute, but recoverable, emergencies that get overlooked?
    Is it due to slower developing emergencies?

    Is it because a 72 year old woman has a heart attack and there is no nursing aide around to notice?
    Is it because a 72 year old woman gets dehydrated because there are fewer nurse visits per day to remind her to drink some water.

    It’s possible technology could help with reminding the elderly, especially those with some cognitive impairment, to eat and drink.

    It’s also possible technology could monitor and spot acute conditions – heart attacks – quickly.

    Is it sheer # of nursing aides? Is the the quality?

    Is it something provided that is intrinsically human? We are built to sense the presence of other humans. Having another human being enter our room a few times a day and saying hi, maybe smiling, showing some concern, produces a physiological response.

    Just wondering what, if any, of the services provided can be replicated with machines and software.

    To save money we could just have people with friendly manners make skype video calls to nursing homes around the country. 5 minutes calls just to provide a friendly face.

    A single friendly person could handle, say, 5 calls an hour. 40 a day.

    In the near future we could have AI avatars make the calls. For the elderly they might not even realize it’s a computer saying hi and having a basic chat.

  • lemmy caution

    Medicare requires that you spend down your assets before you are eligible for nursing home care benefits:


    These benefits have been recently cut squeezing nursing homes:


  • If the deficit is caused by a boom’s shortfall of government cash, this would also affect staff availability at all levels. Presumably, the nursing homes have to compete against bigger spenders at other levels besides nursing assistants. Why just the lowest level?

    I suspect the supply of nursing assistants in inelastic with respect to price and depends (inversely) entirely on competitor buyers. That is to say, workers don’t become nursing assistants unless it’s their only choice; given alternative employment, a higher wage won’t coax them back. Consider what the work involves—

  • Douglas Knight

    There are three results in this paper of varying levels of reliability. The most reliable is the claim that procyclical mortality is among the elderly. This eliminates most hypotheses that have been put forward for procyclical mortality. The second claim is that it is not all elderly, but those in nursing homes. The third is the causal claim that this is due to quantity of nursing staff. This is much less reliable.

    Also, they claim that procyclical mortality is among elderly women, not elderly men. That’s bizarre. Are elderly women kept alive by talking with the staff, while the men don’t?

  • rapscallion

    If the elderly lives are worth it, why don’t the elderly and their families pay more for better nursing? I don’t see why I should care more about their lives than they do.

  • This might be a matter of signaling. It might be more important to signal that you are saving money (e.g., by keeping wages stable) than to spend it wisely.

  • Sackerson

    And if everyone in nursing homes were to be given a speedy and painless death, how much would we save? Isn’t that, perhaps, at the back of the government’s limp-wristed approach to “mercy-killing” in recent years?

  • Thanks to lemmy caution for sharing some actual facts.

    Unfortunately I don’t have time at the moment: is there enough data in there to (at least roughly) calculate the cost per life year of raising the hourly medicare reimbursement rates for nursing assistants?

  • MPS

    Seems like there are mixed incentives here. It might be better to let those people in nursing homes die, and focus more on increasing the welfare of others. It sounds heartless but think seriously about it: many times this might be exchanging raising some old nearly senile person’s existence from miserable to slightly less miserable, vs raising a young and healthy person’s existence from tolerable to decent.

    • MPS

      I should make clear the other sides of the equation: raising the welfare of the old nearly senile person also has with it keeping the welfare of the young healthy person tolerable, while raising the welfare of the young healthy person to decent has the old senile person dead, which is not unpleasant at all.

  • Pingback: Assorted links — Marginal Revolution()

  • carl213

    The future of medical care seems likely to consist chiefly of two things: morphine and hospice care.
    Medical care for the elderly is bankrupting the country (and the world). What can’t go on forever, won’t.
    Pain meds are cheap. I see large amounts of morphine in the future.

  • JasonK

    I suspect that this is a case of quality, not quantity being the problem. I suspect that the number of and the wages of nursing assistants is fairly stable in good times and in bad. In bad economic times there are relatively higher skilled nursing assistants available because they have few options. In good times those relatively higher skilled nursing assistants are able to leverage their higher ability for better paying jobs in other industries.

    Simply paying more may not be a solution to this issue. If the problem is in fact quality not quantity, then higher pay would only be helpful if it was possible to accurately identify the higher quality assistants

  • Dave

    How many cycles were followed? Less respiratory deaths could have been due to confounding variables such as milder flu season one year or the previous season might have killed off the weaklings so next year the stats looked better. Why the sex difference? Nursing home workers anecdotally say old men go fast but women just gradually get weaker, so any increased stress that happens to come along might accelerate deaths.

    As usual, the explanations for the facts needs to be refined,using hypotheses derived from new data before announcing a demanding for an obvious sounding solution. For example one could experimentally raise the salary of a group of nurse’s aids to see if this helps.

    The frustrating thing about these tantalizing studies is how they seldom seem to go to the next level. In other medical studies, promising leads usually flop when pursued, but there is always hope that some increment of progress will be made.

  • Stuart Armstrong

    >Might this be due to a fixed government price that refuses to adapt to the business cycle?

    Might this be due to small furry creatures from alpha centauri kidnapping our nursing home staff when the party lights during boom times allow then to hide their space-ships?

    i.e. do we have any particular reason to think the fixed government price is to blame?

  • Pingback: Does the Economy Affect Air Pollution and Health for Seniors? « Pollution Free Cities()