Projected $312 Billion Cost of Lost Earnings of Virginia K-12 Students due to Pandemic School Closures

Courtesy Eric Hanushek

by James C. Sherlock

Over $312 billion in present value.

That is the estimate published by Stanford’s Eric A. Hanushek of expected economic losses attributable to Virginia’s pandemic school closures.

Virginia students in the COVID cohort can expect on average 5.5 percent lower lifetime earnings.

History indicates that the economic losses will be permanent unless the schools get better. Just returning schools to their pre-pandemic performance levels will not erase the lost learning.

Recovering from the pandemic requires swift and decisive improvements to the schools.

That number is an estimate, but a scientific one. Dr. Hanushek has better credentials with which to make that estimate than perhaps any other economist in the world.

For perspective, the $312 billion present value of projected losses to the Virginia economy due to COVID school shutdowns represents about seven years of state revenue realized from Virginia sources in the 2021 and 2022 biennium.

Even the money pales by comparison to the developmental and emotional damage done to the children.

But feel better.

His estimate of similar losses to California exceeds $1.2 trillion.

Dr. Hanushek is the Paul and Jean Hanna Senior Fellow at the Hoover Institution of Stanford University. He is internationally recognized for his economic analysis of educational issues, and his research has had broad influence on education policy in both developed and developing countries. He received the Yidan Prize for Education Research in 2021.

He has authored or edited twenty-four books along with over 250 articles. He completed his Ph.D. in economics at MIT.

He is a strong supporter of closing learning gaps in schools.

He predicted economic losses in September of 2020 if the schools remained closed. Now he measures them.

Shutdown effects. The arguments for and against school shutdowns were exhaustively proclaimed starting almost immediately.

In his broader nationwide assessment presented in the rest of this article, Dr. Hanushek notes that the longer each school division kept their schools closed, the more damage was done.

Dr. Hanushek projects the economic damages to states, but he did so by assessing two factors.

First, there is now consistent, state-by-state data on how learning patterns have changed.

Second, there is substantial economic analysis related to the US labor market that allows direct estimation of the impact of the pandemic.

His assessments are, considering the work presented and the qualifications of the analyst, compelling.

NAEP scores. One of his sources of data and of differentiations among states is NAEP reading and math losses between 2019 and 2022.

The NAEP samples are large enough that reliable results of these tests are available for all of the states plus the District of Columbia. NAEP tests were administered to 4th and 8th graders.

His analysis

considers just the eighth-grade performance. It does so because this performance links directly to the economic analyses that provide information on the future costs to individuals and the economy of the observed differences in student performance.

 

To paraphrase:

The average score for eighth-grade math fell for every state, with a national average decline of eight NAEP scale score points (see figure 1). This was enough to erase all of the gains that had occurred since 2000.

The losses in reading and math were correlated across states: states losing less in math also tended to lose less in reading (see figure 2).

There were exceptions. Washington, DC, New Jersey and Arizona students had relatively small learning losses in reading compared to large math losses.

He finds it also clear that

the magnitude of learning loss was essentially unrelated to the level of scores in 2019. As seen in figure 3, while some low-scoring states also had large losses (e.g., New Mexico and Washington, DC), Alabama had relatively less loss. However, the three highest-scoring states in 2019 (Massachusetts, New Jersey, and Minnesota) had some of the largest losses.

Lost Earnings

Dr. Hanushek notes that

Extensive research demonstrates a simple fact: those with higher achievement and greater cognitive skills earn more (Hanushek et al. 2015). The evidence suggests that the value of higher achievement persists across a student’s entire work life.

The United States rewards skills more than almost all other developed countries. The high value of skills in the US simply reflects the dynamic, technologically driven economy where workers are continually adjusting to new jobs and new ways of doing things (Hanushek et al. 2017).

But the equivalent way of viewing the high rewards of skills is that the United States punishes those without skills more than other countries. In other words, those with lower achievement see larger negative impacts on their lifetime earnings than found elsewhere.

The evidence on the labor market value of skills implies that the average student during the pandemic will have 5.6 percent lower lifetime earnings. Figure 4 compares the expected earnings given the eight-point loss in math achievement to what could have been expected without the pandemic.

The future-income impact of the pandemic also differs dramatically by which state the student was in during the pandemic. Figure 4 displays the percentage drop in income by state based on the varying learning losses seen in the NAEP scores. (These calculations consider just average state skill levels; they value skills the same across the country, reflecting the fact that many people move away from where they were educated.)

He notes that the average losses obscure large variations in losses to individuals. But that too hides bad news.

The existing data points to significantly larger impacts on disadvantaged students who tended to fare worse during the pandemic. The exact magnitude of this differential is, however, not known.

State-by-State Aggregate Impacts.

The author goes on to note that economic growth of states is highly dependent on the quality of the state’s labor force. And

The pandemic’s effects imply that the future workforce will be less prepared to contribute to economic growth. Even if education returns to its pre-pandemic quality, there is a cohort of students that will move through the future labor force with lower skills and achievement.

Figure 5 shows his estimate of GDP losses by states over the rest of this century.

He then calculates those losses as present total value of those losses. See figure 6 with which I opened this review.

Do These Losses Go Away? The calculations of economic costs assumed that all schools have returned to their January 2020 level of performance and that the NAEP losses measured in spring of 2022 are permanent.

He notes that some argue, however, that either kids are resilient enough or parents and schools will adjust to make up for those losses. He cautions that

The existing evidence, although largely international, suggests that just returning to the previous normal will leave the losses as permanent.

That evidence, of current interest, has been accumulated from long-term strikes by overseas teachers unions.

He also notes that research also shows that school breaks in the U.S. have a larger impact on math than reading — exactly what appears in the NAEP estimates of learning losses due to the pandemic.

Conclusions. He states that

There is overwhelming evidence that students in school during the closure period and during the subsequent adjustments to the pandemic are achieving at significantly lower levels than would have been expected without the pandemic.

He concludes the schools were not necessarily responsible for all the losses but that recovery from these losses necessarily falls on them. And that efforts to date won’t get it done.

If the schools are not made better, there will be continuing economic impacts as individuals and the nation will suffer from a society with lower skills. And, while some states have less learning loss to make up, none can avoid taking on the task of improving the schools.

Dr. Hanusek offers that

Without action, not only will individuals in the COVID cohort of students suffer long-term income losses, but also the individual states will see shrunken economic activity.

How to avoid the losses.

He does not in this essay make recommendations on how to improve the schools, but we can glean that from his massive contributions to existing research on that subject.

His just-released work with Danielle Handle U.S. SCHOOL FINANCE: RESOURCES AND OUTCOMES published by the National Bureau of Economic Research will prove foundational.

That article will appear in Eric A. Hanushek, Stephen Machin, and Ludger Woessmann, eds. forthcoming, 2023. Handbook of the Economics of Education. Vol. 7.

Progressives, led by teachers unions, will find much about which to disagree with Dr. Hanushek.

But they will need evidence, lacking at the level of his research, to contradict him effectively.

Endnote added Jan 2 at 16:55

Dr. Hanushek is acknowledged worldwide as expert in the economics of education.

The Yidan Prize Foundation is a $3.9 million education prize awarded to individuals whose work makes profound contributions to education research and development, with the ultimate aim of creating a better world through education.

The largest award prize in the field of education, the Yidan Prize is managed by an international judging committee and global advisory board formed by leading experts in education.

Dr. Hanushek won the Yidan Prize for Education Research in 2019.

I presented in Part 1 above his most up to date assessment of these particular costs.

In Part 2 of this series I will discuss the broader economic effects of school closures in addition to the estimates of personal income declines featured in this very small piece of Dr. Hanushek’s work.

Even in this small study, he stopped the assessment at the end of the 2021-22 school year by stipulating for analytical purposes that in following years schools would return to the efficiency and effectiveness they exhibited in 2019.

In Virginia we anticipate that unless major changes are made that is highly unlikely to happen.

Virginia schools deal with major teacher shortages that are projected to worsen at both the input and retirement ends and, as polled by JLARC, 56% of teachers reported that student behavior was a very serious problem. That speaks to school chaos.


Share this article



ADVERTISEMENT

(comments below)



ADVERTISEMENT

(comments below)


Comments

98 responses to “Projected $312 Billion Cost of Lost Earnings of Virginia K-12 Students due to Pandemic School Closures”

  1. But the ABC Store earnings were up!!!

  2. Eric the half a troll Avatar
    Eric the half a troll

    From the referenced article:

    “The losses in reading and math were correlated across states: states losing less in math also tended to lose less in reading (see figure 2). There were, however, some exceptions. For example, Washington, DC; New Jersey; and Arizona students had relatively small learning losses in reading compared to large math losses.”

    What you wrote (without quotes):

    “The losses in reading and math were correlated across states: states losing less in math also tended to lose less in reading (see figure 2).

    There were exceptions. Washington, DC, New Jersey and Arizona students had relatively small learning losses in reading compared to large math losses.”

    Also you wrote (without quotes):

    “The average score for eighth-grade math fell for every state, with a national average decline of eight NAEP scale score points (see figure 1). This was enough to erase all of the gains that had occurred since 2000.”

    While the author wrote:

    “Looking across the nation, the average score for eighth-grade math fell for every state, with a national average decline of eight NAEP scale score points (see figure 1). This was enough to erase all of the gains that had occurred since 2000”

    1. James C. Sherlock Avatar
      James C. Sherlock

      Sloppy editing by me. Good catch. Fixed.

  3. Nancy Naive Avatar
    Nancy Naive

    For what it’s worth Captain, you should avoid cited non-refereed publications from economists.

  4. I doubt that the so-called “skills” that differentiate lifetime earnings correlate with what one learns in a given year or two in school. My conjecture is that this loss of learning is basically harmless.

    1. LarrytheG Avatar
      LarrytheG

      Good Lord, we AGREE!

    2. I agree. Especially for the younger kids, it will not be that hard for most of them to catch up. Those who were already juniors or seniors in high school may take a little longer to pick up missing knowledge, but I still expect little overall effect on their lifetime earning potential.

      1. I disagree.

        I suspect you do not have extensive experience with marginal students and those already struggling at school.

        I found this to be the most important quote in the article.

        “The existing data points to significantly larger impacts on disadvantaged students who tended to fare worse during the pandemic. The exact magnitude of this differential is, however, not known.”

        The students with highly educated parents who focus on their children’s education will likely suffer the least.

        For example, my wife and I spent considerable time teaching our children. They are grown now, but had they missed significant time during first grade, would they have still be reading at or above grade level? Absolutely! They went to school already reading at a high level.

        There are many students in Virginia and elsewhere who are not as fortunate. The public schools are their only hope, and many didn’t do their jobs during critical years of their development.

        1. LarrytheG Avatar
          LarrytheG

          I don’t think they do their job pandemic or not if you look at the persistent “gap” with economically disadvantaged kids.

          When Mom/Dad do NOT have a good education AND their life is not economically secure , they’re not going to be able to help their kids the same way that educated parents can and will.

          Yet, at most public school districts, there is a SOL “gap” that some perceive as “minorities” but in reality it’s economically disadvantaged kids of all races to include the ones with English as a second language.

          There may be SOME public schools in Virginia that do this right. I know that Region 7 has a program to address the issue but in other districts, even wealthy ones, the “gap” is there and stays that way.

          1. To LarrytheG:
            While I largely agree, it’s a bit more complicated.

            My comments to WayneS relate specifically to the article, and I left it at a high level.

            With respect to non pandemic related education, I agree that there is much room for improvement, but I don’t place it 100% on the schools. I believe students spend about 80% of their time away from school. Parents also have a role, and no amount of government intervention can fully account for a bad home environment.

            Getting back to the article, gaps in instruction are not in the child’s interest at all. That includes summers off. The current school calendar was created at a time when children were needed to work on the farm during the summer. That’s no longer the case, and the issue should be addressed, in my humble opinion.

          2. LarrytheG Avatar
            LarrytheG

            The “gaps” are more or less independent of whether the kids have “good” or “bad” parents. We really don’t know and should not be ascribing the “gaps” to bad parenting. That’s an easy out for the schools.. not their problem.

          3. LarrytheG:
            “When Mom/Dad do NOT have a good education AND their life is not economically secure , they’re not going to be able to help their kids the same way that educated parents can and will.”

            A good education is helpful for parents, but what is most important is their commitment to do what is best for their children, even when doing so conflicts with their own desires.

            Many of the things lacking in a students life are free, and don’t require a high degree of education. Take sleep for example. Many students arrive at school without having had enough sleep the night before. They are at a disadvantage.

            My father dropped out of school before reaching middle school. He grew up during the depression and had to work to feed himself and contribute to the family. Nonetheless, he read to my brother and I EVERY night when we were young. Our education was a high priority for him, even though he worked long hours doing manual labor.

            Schools need to do better, but many parents could as well.

          4. LarrytheG Avatar
            LarrytheG

            Educated parents are also guilty of not doing what is right for their kids. Some folks are good parents
            and some not. Those that are poorly educated and struggling to make a living and do life “right” may not be all the can or should be for their kids. That’s not the primary issue in the schools. They just don’t do a good job with kids who are economically disadvantaged. There are good parents and bad parents in all economic spectrums.

          5. LarrytheG:
            “Educated parents are also guilty of not doing what is right for their kids. Some folks are good parents.”

            True.

            “They just don’t do a good job with kids who are economically disadvantaged.”

            That doesn’t make sense without some explanation. Teachers don’t set out to rob disadvantaged children. Some students have more obstacles to overcome. Schools should try to help, but that’s only one part of the equation.

          6. LarrytheG Avatar
            LarrytheG

            It’s the justification that Conservatives use to advocate for non-public charter schools, no?

            “Economically disadvantaged” is how we identify kids who need help – it’s not decided based on who
            their parents are but rather their economic circumstances and parents will poor educations don’t get
            good jobs not because they’re lazy or “bad” but because they are poorly educated and lack the basic
            skills and knowledge they need to qualify for higher paying jobs.

            How did they get that way? How did they spend 12 yrs in school and finish as functionally illiterate
            and qualify mostly for jobs that don’t require much education?

            Did they have “bad” parents also?

            is it a cycle?

          7. Once again you are straying farther and farther from the topic.

            The point is that public schools have an important responsibility to our youth and many largely abandon their duties during the pandemic. That will have a lasting impact, especially for those most at risk.

          8. LarrytheG:
            “When Mom/Dad do NOT have a good education AND their life is not economically secure , they’re not going to be able to help their kids the same way that educated parents can and will.”

            A good education is helpful for parents, but what is most important is their commitment to do what is best for their children, even when doing so conflicts with their own desires.

            Many of the things lacking in a students life are free, and don’t require a high degree of education. Take sleep for example. Many students arrive at school without having had enough sleep the night before. They are at a disadvantage.

            My father dropped out of school before reaching middle school. He grew up during the depression and had to work to feed himself and contribute to the family. Nonetheless, he read to my brother and I EVERY night when we were young. Our education was a high priority for him, even though he worked long hours doing manual labor.

            Schools need to do better, but many parents could as well.

  5. walter smith Avatar
    walter smith

    Wow. If only someone could have warned us of the cost of lockdowns…
    Without being censored, called an anti-vaxxer, conspiracy theorist, etc.

  6. LarrytheG Avatar
    LarrytheG

    ” The methodology for the projections follows that in Hanushek and Woessmann (2020), which estimated the immediate impact of school closures for the G20 countries. It uses the growth analysis for US states found in Hanushek, Ruhose, and Woessmann (2017a, 2017b).

    3 Growth in state GDP in the absence of the pandemic is assumed to be at the national level of 1.5 percent per year.

    4 For such calculations, it is common to discount future losses at 3 percent. This is the assumed interest rate that the bank pays on the account.”

    I have to say, I’m amazed at the methodology. I probably don’t understand.

      1. LarrytheG Avatar
        LarrytheG

        “The argument made in the new AEI brief is one he and his colleagues have made repeatedly over the years. It goes like this: If we imagine that standardized test scores go up a certain amount, this will add trillions to the GDP. ”

        Okay, that’s what I was getting out of it also but could not understand how that relationship actually “works”.

        One _could_ expand the idea to include other countries test scores and their GDPs I suppose to collaborate the approach but it probably would turn into a rats nest.

    1. Nancy Naive Avatar
      Nancy Naive

      I liked this from the 2020 paper…

      “ Learning losses are portrayed as lower cognitive skills for the cohort of students enrolled in grades 1-12 in January 2020. The learning losses are presented in “school-year equivalents” using the rule of thumb that three years of schooling is equivalent to one standard deviation of test scores. ”

      Rule of thumb? Really?

      http://hanushek.stanford.edu/sites/default/files/publications/The%20Economic%20Impacts%20of%20Learning%20Losses_final_v1.pdf

      I now understand the comment about one-handed economists.

      1. Eric the half a troll Avatar
        Eric the half a troll

        The scatter plot of math vs english showing a “correlation” with no r^2 cited (and I mean in the underlying reference – not Sherlock’s piece) was a dead give away. You can see that the r^2 will be no where near correlation territory. Did not really impact the point of Sherlock’s piece so I did not pick on it outright but it does go to the voracity of the “research” conducted.

        1. Nancy Naive Avatar
          Nancy Naive

          Eric, I’ve linked three papers you might want to review above or below depending on how you sort. All attack the Hoover Institute guys for basically sloppy work. I tried to link one to one of your posts but it was deleted by our new censor.

          Banned

  7. It is amusing to see the usual suspects — quite predictably — questioning Hanushek’s methodology because his conclusions do not suit them.

    Based on Sherlock’s presentation, Hanushek makes three broad arguments:

    (1) In the knowledge economy, a higher level of academic achievement translates into higher lifetime earnings.

    (2) Thanks in significant part to the school shutdowns, American students have suffered a significant loss in educational achievement.

    (3) Absent a significant change in policy or resources, American students are not likely to make up those losses.

    My advice to those who would criticize Hanushek’s findings: Focus on one of those three broad propsitions. The first and second, it seems to me, are unassailable. The third is highly likely to be true but perhaps could be contested.

    1. LarrytheG Avatar
      LarrytheG

      re: ” (3) Absent a significant change in policy or resources, American students are not likely to make up those losses.”

      that’s a pretty broad statement isn’t it?

      How many school districts in the US?

      are we saying that public school failed in general across the US, even the ones that stayed open?

      What would Virginia have to do to make up the losses? Are there any public schools in the US that is successfully making up the losses? Are there private schools that do that especially for the demographic that suffered the biggest losses, the economically disadvantaged?

    2. James C. Sherlock Avatar
      James C. Sherlock

      Watching these guys try to challenge Dr. Hanushek’s scholarship is pretty embarrassing. Read the end note I added.

      1. Eric the half a troll Avatar
        Eric the half a troll

        Sorry to waste a comment here but you asked us to read your end note. You wrote:

        “…as polled by JLARC, 56% of teachers are afraid for their safety at work…”

        What the study said was:

        “More than 56 percent of respondents said that student behavior was a very serious problem”

        Citing student behavior as a problem does not equal teachers being “afraid for their safety at work”.

    3. Nancy Naive Avatar
      Nancy Naive

      Well, I’m not the first.
      The causality claims have been challenged before. Easy reading and easier to see the flaw
      https://www.norrag.org/critical-reflections-2018-world-development-report-learning-important-cant-world-bank-learn-hikaru-komatsu-jeremy-rappleye/

      1. Posting the link once is adequate., hence the deletion of four additional posts of the same link.

    4. Nancy Naive Avatar
      Nancy Naive

      It’s the quantification in question. You do realize he’s projecting 77 years. That alone should give anyone pause.

    5. LarrytheG Avatar
      LarrytheG

      re: ” (2) Thanks in significant part to the school shutdowns, American students have suffered a significant loss in educational achievement.”

      schools that stayed open also suffered losses.

      I thought Florida stayed open?

      https://uploads.disquscdn.com/images/7077c8e226fdc92d7212fc30e5832f3fc5181177ad9fa02bb7f906994cfa7ad4.jpg

      1. Nancy Naive Avatar
        Nancy Naive

        Too many old people?

    6. Eric the half a troll Avatar
      Eric the half a troll

      “Thanks in significant part to the school shutdowns, American students have suffered a significant loss in educational achievement”

      He doesn’t really make that case – I don’t think he even tried to make it – if he did he failed (so to speak). In fact, many of the states that forced in-person schooling earliest (2020-21 school year) are among those with the highest loses.

      1. James C. Sherlock Avatar
        James C. Sherlock

        Clearly the disastrous NAEP results don’t “make the case” to you for significant loss in student achievement. Nothing would. Your job is to deny. You have earned a promotion.

        1. Eric the half a troll Avatar
          Eric the half a troll

          Hope that is not an ad hominem attack there, Sherlock… wouldn’t want to get you censured or anything.

  8. Kathleen Smith Avatar
    Kathleen Smith

    First: Hanushek is a well-respected researcher in the area of economics and finance. The journal that published this is an excellent journal. Two comments:

    Second: The research is similar to that of the findings of Sanders and associates around 2005 described the same findings when a student was exposed to an ineffective vs effective teacher for three consecutive years using results from the Tennessee state assessment.

    Third: The lasting effects may be altered with immediate, and put much stress on immediate, and sound remediation. Teachers are positioned to do this.

    Question for commenters: Do you place those Grade K-5 children who are most at-risk, most behind, with the most effective teacher at that grade level or do you save that most effective teacher for the students who are moderately at-risk, somewhat behind? This is the question for many young principals.

    1. LarrytheG Avatar
      LarrytheG

      hey thanks Kathleen!

      re: ” the same findings when a student was exposed to an ineffective vs effective teacher for three consecutive years using results from the Tennessee state assessment.”

      does the study define what is an effective teacher or an ineffective one?

      is there a link?

      1. Kathleen Smith Avatar
        Kathleen Smith

        Ill find it in ERIC again in the morning.

      2. Kathleen Smith Avatar
        Kathleen Smith

        https://www.edweek.org/leadership/william-sanders-pioneer-of-the-value-added-model-for-evaluating-teachers-dies/2017/03

        The article, his obituary, describes his research definition of an effective teachers- one that could raise scores across all students. The next is the research.

        1. LarrytheG Avatar
          LarrytheG

          Thanks Kathleen… I’m off to read it and will get back!

        2. LarrytheG Avatar
          LarrytheG

          So did read and am somewhat familiar as he is looking at the childls performance to judge the teacher. There are some issues in my mind. 1. class makeup/composition 2. are kids sent out for Title help , is there Title help available? 3. what happens if grades 1, 2 and 3 are not tested with standardized testing and we don’t do that until 4th grade?

      3. Kathleen Smith Avatar
        Kathleen Smith

        https://education-consumers.org/pdf/TVAAS-text-revision-from-web-page.pdf

        https://files.eric.ed.gov/fulltext/ED388697.pdf

        https://comptroller.tn.gov/content/dam/cot/orea/advanced-search/2019/StudentPlacementFullReport.pdf

        The last article is new data analysis, completed w/o Sanders, and shows the same findings as his early stuff.

        He was a really well known researcher. I had an opportunity to spend a day with him in 2005. He was exceptional. Had a prolific belief in high expectations for all kids and was one of the first researchers to show the achievement gap for what it is — low expectations. He grew up poor very close to the Virginia line outside of Bristol.

    2. James C. Sherlock Avatar
      James C. Sherlock

      My endnote added 30 minutes ago addresses your two points directly and agrees with them.

    3. LarrytheG Avatar
      LarrytheG

      ” Do you place those Grade K-5 children who are most at-risk, most behind, with the most effective teacher at that grade level or do you save that most effective teacher for the students who are moderately at-risk, somewhat behind? This is the question for many young principals.”

      I think the kids that were behind prior to the pandemic never got the best teachers to start with. It’s human nature to teach the kids, see progress, and keep on going especially when 2/3 or 3/4 of the class are moving forward.

      Economically disadvantaged kids are just harder to teach, a challenge for many new teachers, and schools don’t have enough specialists to deal with those kids.

      Those kids fall behind more than the others over summer vacation and it’s no shock on remote instruction if there was no or limited internet and the online instruction was not keyed to ED kids but the regular kids.

      These kids won’t catch up (I agree) unless there are a lot more resources and specialists, like Title teacher hired and deployed.

      Your basic question of who to “save” is the way classrooms often run in normal times – you want as many as you can to succeed and you’re not going to abandon them to try to save one or two.

    4. Nancy Naive Avatar
      Nancy Naive

      Yes, yes, world renowned… at the policy institute.

  9. Bubba1855 Avatar
    Bubba1855

    Hey folks…I think we can all agree that lower ‘learning’, aka test scores coming out of the lock downs is not a good thing. How ‘bad’ is ‘bad’? We’ll never know. Especially troubling for me is that the kids that are defined as being in the lower economic quadrants have suffered more which is not good for them nor our society. I prefer to focus on what we can do for the ‘lower quadrant’ kids in order to help them catch up than to try to calculate the impact in the future. Happy new year…

    1. James C. Sherlock Avatar
      James C. Sherlock

      With these commenters, never say “we all know”.

  10. Nancy Naive Avatar
    Nancy Naive

    Well, ya know what Steve says about models. Of course, he usually restricts to just atmospheric models. Nevertheless… Extrapolating is a dangerous game.

    You do realize Captain that those losses are against Hanushek‘s fantastic predicted gains loosely based on his assumption that correlation is causation, right?

    1. James C. Sherlock Avatar
      James C. Sherlock

      Weak try.

      You did not read the references in this period of time. Your comment was posted only 20 minutes after the article was published.

      Compare your credentials to Dr. Hanushek’s.

      Be embarrassed.

      1. Nancy Naive Avatar
        Nancy Naive

        Careful that borders on a personal attack.. Carol! It’s okay this time… let it slide.

        So explain it to me Captain. On what is he basing these losses? How is he extrapolating?

        I’ll wait.

        1. Nancy Naive Avatar
          Nancy Naive

          I’m thoroughly amazed that those, who so willingly dismiss the correlation between atmospheric CO2 and the rise in sea surface temperature, are enamored with a dubious claim of causality between standardized test scores and GDP.

          Every once in awhile, someone comes along and pulls the whole bloody lot out of the mud and into the light. This is not one of those times.

      2. Nancy Naive Avatar
        Nancy Naive

        Don’t have to read them. They’re unrefereed.

        1. James C. Sherlock Avatar
          James C. Sherlock

          “Don’t have to read them”. Got it.

          1. Nancy Naive Avatar
            Nancy Naive

            Those are invited papers from agenda driven institutes. They are not checked for more than a passing reasonableness. BTW, there are some refereed papers refuting the causality claim. You might want to look at the problem Edward or Eric has. Which is it? Look at you first figure acknowledgment.

  11. LarrytheG Avatar
    LarrytheG

    so if someone takes a year (or two) longer to complete their K-12 schooling and graduate is the economic loss the additional year they spent in school rather than the economy?

    Someone who waits a couple of years before going on to college , the economic “loss” is the additional amount they would have made had they not deferred college?

    If a kid never graduates from HS, we compute the economic loss how, on the difference between what he would have made additional if he graduated?

    Just trying to understand.

    I think these things play heavily into the study’s premises.

    If this is the way it works, couldn’t a similar calculation be done for kids that never graduate even without a pandemic?

  12. Nancy Naive Avatar
    Nancy Naive

    Not that this makes much of a difference, but I know Nelly’s father. He’s a who’s who in American mathematicians and has an Erdos number of 0, having published with Paul Erdos. Nelly also takes issue with your author’s conjectures…

    “Abstract
    Regression models employed to help understand, predict, and enhance national economic growth have increasingly come to rely on quality of education as an important variable. Such models, however, often make a number of untenable assumptions not congruent with reality. A case in point is the recent book by E. Hanushek and L. Woessmann, Universal Basic Skills: What Countries Stand to Gain. Unpacking the notion of educational quality – which the book argues is totally captured by PISA and TIMSS scores in math and science – this paper critiques regression models that assume a particular PISA threshold score as quality and stable and linear national development over 80 years, regardless of great variability in countries’ economic production systems, histories, physical resources, and social contexts. Hanushek and Woessmann argue that quality of education makes a substantial contribution to economic growth and that quality contributes 6.3 times more than quantity (i.e., secondary school enrollment). However, the narrow definition of quality and the disregard for complexity in explaining social and economic conditions seriously diminish the validity of the findings. The emphasis on education to the exclusion of other forces in society serves to detract attention from deeper policy measures and makes the book less a valid academic study than an effort to manipulate the soft power of OECD to convince governments of the usefulness of international student testing.”

    http://www.paecon.net/PAEReview/issue76/Stromquist76.pdf

  13. Nancy Naive Avatar
    Nancy Naive

    Not that this makes much of a difference, but I know Nelly’s father. He’s a who’s who in American mathematicians and has an Erdos number of 0, having published with Paul Erdos. Nelly also takes issue with your author’s conjectures…

    “Abstract
    Regression models employed to help understand, predict, and enhance national economic growth have increasingly come to rely on quality of education as an important variable. Such models, however, often make a number of untenable assumptions not congruent with reality. A case in point is the recent book by E. Hanushek and L. Woessmann, Universal Basic Skills: What Countries Stand to Gain. Unpacking the notion of educational quality – which the book argues is totally captured by PISA and TIMSS scores in math and science – this paper critiques regression models that assume a particular PISA threshold score as quality and stable and linear national development over 80 years, regardless of great variability in countries’ economic production systems, histories, physical resources, and social contexts. Hanushek and Woessmann argue that quality of education makes a substantial contribution to economic growth and that quality contributes 6.3 times more than quantity (i.e., secondary school enrollment). However, the narrow definition of quality and the disregard for complexity in explaining social and economic conditions seriously diminish the validity of the findings. The emphasis on education to the exclusion of other forces in society serves to detract attention from deeper policy measures and makes the book less a valid academic study than an effort to manipulate the soft power of OECD to convince governments of the usefulness of international student testing.”

    http://www.paecon.net/PAEReview/issue76/Stromquist76.pdf

  14. Eric the half a troll Avatar
    Eric the half a troll

    Figure 5 shows WV, KS, and OK in the top ten in terms of losses to their GDP. Please note that OK was a trifecta Republican state at the time and importantly KS and WV had mandatory in-person schooling in the 2020-21 school year when most of the rest of the country did not. One would think, therefore, that they would fair among the best – not the worst if the hypothesis put forward were true…. apparently, it is not. Edit: Add Massachusetts to that list in the top ten GDP losses that forced schools open in 2020-21 along with #11 North Carolina and #13 New Mexico. Not much of a correlation let alone causation.

    1. James C. Sherlock Avatar
      James C. Sherlock

      What, exactly was Dr. Hanushek’s “hypothesis”. Quote it.

      1. Eric the half a troll Avatar
        Eric the half a troll

        This is what JAB says it is: “Thanks in significant part to the school shutdowns, American students have suffered a significant loss in educational achievement”

        For the record, I think JAB is mistaken but the author did state:

        “There is overwhelming evidence that students in school during the closure period and during the subsequent adjustments to the pandemic are achieving at significantly lower levels than would have been expected without the pandemic.”

        But his own data shows no such correlation with in many cases those states that forced students to return earlier than others (ie, a shorter “closure period”) are among those showing the highest loses.

        1. LarrytheG Avatar
          LarrytheG

          ” Data from the National Assessment of Educational Progress (NAEP) now show the significant impact of the pandemic on learning. The abstract nature of test score declines, however, often obscures the huge economic impact of these learning losses. NAEP results indicate large differences in learning losses across states, and this analysis provides state-by-state estimates of the economic impacts of the losses. Students on average face 2 to 9 percent lower lifetime income depending on the state in which they attended school. By virtue of the lower-skilled future workforce, the states themselves are estimated to face a gross domestic product (GDP) that is 0.6 to 2.9 percent lower each year for the remainder of the twenty-first century compared to the economic expectations derived from pre-pandemic years. The present value of future losses for states depends directly on the size of each state’s economy. At the extreme, California is estimated to have lost $1.3 trillion because of learning losses during the pandemic. These losses are permanent unless a state’s schools can get better than their pre-pandemic levels.”

          The NAEP data as you pointed out INCLUDES schools that went in-person AND showed losses.

          It appears that he has comingled these schools in the data as if all of them suffered losses because they went remote when in fact, some were in-person.

        2. James C. Sherlock Avatar
          James C. Sherlock

          So. You found no hypothesis. Got it.

          1. Eric the half a troll Avatar
            Eric the half a troll

            Change “hypothesis” to “assertions” if it makes you feel better.

          2. James C. Sherlock Avatar
            James C. Sherlock

            He made no “assertions” either. He explained in detail his analytical methods. The results are called “findings”. But keep at it.

          3. Eric the half a troll Avatar
            Eric the half a troll

            That word works just as well if it keeps you happy… btw, do these comments count…??

  15. James C. Sherlock Avatar
    James C. Sherlock

    50 comments. Time for a “panic on the left” scorecard. 36 comments from Nancy, Eric and Larry, not including deletions.

    They can’t abide the findings that extended school closures may cause lifetime economic damage because those closings were the holy grail of the left. And when a renowned economist puts a price tag on it, they panic rather than apologize. We get it.

    They take the open mike to challenge the scholarship of the most widely honored educational economist alive today.

    Three or four would have registered their disagreement. 36 is panic.

    I find the entire show embarrassing.

    1. LarrytheG Avatar
      LarrytheG

      There is no panic. It’s a bogus study, not peer reviewed for good reason, A prior “study” using the same premise on the connection between education and GDP was refuted by other reputable scientists that clearly documented the flaws in the study.

      NN has posted several studies that refute this study – in response to the false claims in this blog post. Notably, Bacon and Sherlock are not responding to these other studies that refute the premise behind this one as Carol (whose own comments on the issue seem to align with JAB and Sherlock, keeps knocking down the links to these other studies as “duplicates”. Different independent studies all refuting the original premise.

      Basically, this is a “study” in support of a continuing conservative narrative fostered (perhaps paid for) by Conservative organizations like the American Enterprise Institute and the Hoover Institute about the so-called “lock-down” and it’s “harm” to education as if there was no risk in keeping schools open because of a proven false belief that because kids are resistant to the disease, they also can’t spread it. They can and did and do.

      ” So, the presence of a high viral load in infected children increases the likelihood that children, even those without symptoms, could readily spread the infection to others. The bottom line? Public health measures are as important for kids and teens as they are for adults.

      Coronavirus outbreak and kids – Harvard Health”

      https://www.health.harvard.edu/diseases-and-conditions/coronavirus-outbreak-and-kids#:~:text=So%2C%20the%20presence%20of%20a,as%20they%20are%20for%20adults.

      This fact is being ignored by the folks insisting that “in-person” had no risks.

      What the study ALSO clearly showed was that even states that maintained “in-person” instruction also had HIGH learning losses also, especially the economically disadvantaged demographic.

      That point is being lost – that even schools that maintained in-person instruction also had high learning losses. Now THAT, would be something
      worth studying because the basic premise that in-person was/is “better” than remote apparently did not hold true.

      The study also “counts” those schools as being “harmed” also by the pandemic even though they did NOT “lock down”.

      So, it’s just a seriously flawed study that Conservative types claim is done by a guy with “unquestionable” credentials. Other reputable scientists looked at his basic premise and concluded it was fatally flawed.

      So no panic at all, just some simple truth-telling which is necessary on a continuing basis for some of the stuff promoted by some of the usual suspects in BR.

      There is no such problem BTW, with other authors like Matt Hurt and Kathleen Sullivan and others who I am thankful for their informative and very legitimate commentary.

    2. Nancy Naive Avatar
      Nancy Naive

      Yeah, okay. So, uh, how do you address the fact that your author simply ignores temporal in the causality? If you ignore time, all manner of things are possible.

      We needn’t get into the nuts and bolts of the regression models. We only have to see that he correlates scores at time t with GDP at time t, and then declares score caused GDP. Why not vice versa?

    3. I’m not of the left and I am not panicking.

      Neither am I as accepting of Dr. Hanushek’s conclusions as you. It bothers me that I can find no supporting data in his documentation for the dollar values he calculated for his predicted lifetime economic damage.

      How did he arrive at the $312.3 billion figure for Virginia?

      And, Nancy & Eric t 1/2 a T are correct, The scatter plots provided show very little correlation between the variables graphed, and would no doubt have a very low coefficient of determination had the author computed one..

      We all know that correlation does not necessarily equal causation, but causation can never be claimed where correlation does not even exist.

    4. I’m not of the left and I am not panicking.

      Neither am I as accepting of Dr. Hanushek’s conclusions as you. It bothers me that I can find no supporting data in his documentation for the dollar values he calculated for his predicted lifetime economic damage.

      How did he arrive at the $312.3 billion figure for Virginia?

      And, Nancy & Eric t 1/2 a T are correct, The scatter plots provided show very little correlation between the variables graphed, and would no doubt have a very low coefficient of determination had the author computed one..

      We all know that correlation does not necessarily equal causation, but causation can definitely not be claimed where virtually no correlation exists.

      In this case, I think the climate model analogy is apt.

  16. Nancy Naive Avatar
    Nancy Naive

    Larry and I are not alone. The good news for Hanushek is that he needn’t republish in the Journal of Irreproducible Results, nevertheless the challenge to causality shows the claims are suspect…

    “Anxious about all this, we conducted our own study that looked more closely at the causality claims. We utilized the exact same sample of countries, data, and methods utilized by Hanushek and Woessmann. The only adjustment we made was selecting a more reasonable time period for the economic growth calculation.

    Once when we did this, however, we came up with a disturbing result: The relationship becomes so dramatically weak it cannot support the claims.

    Figure 2. Relationships between test scores and economic growth: (a) our replication of Hanushek’s findings and (b) our findings.

    The key point is time mismatch. And one need not be an educational researcher or statistics geek to understand the logic. Hanushek compared test scores and economic growth over the same time period (1960-2000), reporting the strong relationship he found (the left side of Figure 2). However, it logically takes at least a few decades for students to occupy a major portion of the workforce. So we compared test scores for one period with economic growth in a subsequent period (1995-2014). Surprisingly, the relationship, which was once looked so strong, now looked suspect (the right side of Figure 2).

    This finding not only refutes a tight test score-GDP growth link, but also unmasks causality as simply statistical coincidence. And coincidence can certainly not support the notion that aggressively raising test scores will eventually generate enough extra money to “solve our current fiscal and distributional woes.” Nor can it support the claim that helping American children ace PISA-style tests will lead to “20 percent higher paychecks” for every American this century — although we certainly wish it did.”

    https://www.tandfonline.com/doi/full/10.1080/03050068.2017.1300008

    In real layman’s terms, the original authors neglected the time lag between taking the tests and being more than a college new hire. Get it? It takes time for that 4th grader to become a captain of industry, or even a mid-level manager.

      1. Nancy Naive Avatar
        Nancy Naive

        Yep. That’s right. Meaning what? That’s right. They continue to make claims based on claims shown to be highly suspect 5 years ago.

        1. Nancy Naive Avatar
          Nancy Naive

          The claim of causality between test scores and gdp upon which the Captain’s cited paper relies. No causality, no predictability.

          Hey, but the Hoover Institute is willing to publish.

          Economists. Meh.

          1. LarrytheG Avatar
            LarrytheG

            did it have one of those disclaimers saying the researcher had no conflict of interest in the study?

  17. Nancy Naive Avatar
    Nancy Naive

    Here ya go Captain… it may be a purloined copy… enjoy. More than likely the 1st authors copy downloadable copy.

    https://researchmap.jp/7000008095/published_papers/19610621/attachment_file.pdf

  18. Nancy Naive Avatar
    Nancy Naive

    This is yet another challenge to your source’s method Captain.

    Maybe Carol will leave it up this time or at least say how it “violates” the rules.

    https://www.norrag.org/critical-reflections-2018-world-development-report-learning-important-cant-world-bank-learn-hikaru-komatsu-jeremy-rappleye/

    1. walter smith Avatar
      walter smith

      Wow. Sure do seem to have thin skin…
      Guess you don’t like being censored, huh?
      Anyway, rather than do the usual Leftist tactic of correlation does not equal causation or find one thing wrong in some assertion (like the convenient defense medical database having some unexplained error, therefore ignore all the bad vaccine signals and doctor’s testimony), do you assert that the lockdowns did not have any negative effects?
      They clearly did and it is incalculable. Worse than any calculations people can come up with. Lost lives, delayed treatments, lost learning, loss of trust in “experts,” and on and on.
      So are quibbling over price or are you defending the indefensible lockdowns?

      1. James McCarthy Avatar
        James McCarthy

        So, if it’s incalculable, where’s the beef? Since the mythical “lockdowns” saved lives and mitigated medical costs, please offer some calculations for those results.

        1. walter smith Avatar
          walter smith

          You are one of those guys who quotes the Bible with “Judge not” and leaves out “lest you be judged” aren’t you?
          Did you read the next sentence?
          We can never fully quantify the cost. But it was worse than any number that will be calculated. It is like damages in a lawsuit. Absent ridiculous runaway juries, you are never made whole in a normal case.
          Everything our “betters” said to do was wrong. The emperor has no clothes.

          1. James McCarthy Avatar
            James McCarthy

            Y’all must learn to avoid mind reading and/or ESP inferences when commenting.

          2. walter smith Avatar
            walter smith

            I think not. People on the Left do so. Or did I insult you by presuming you knew some Bible to quote…
            Last night, the usual people who hate “thoughts and prayers” were thoughting and prayering a lot…

          3. James McCarthy Avatar
            James McCarthy

            You make it clear you think not most of the time.

          4. walter smith Avatar
            walter smith

            So? My thinking not most of the time is right nearly all of the time.

  19. f/k/a_tmtfairfax Avatar
    f/k/a_tmtfairfax

    Sounds to me more like a climate change study using adjusted datapoints. While I don’t disagree that the long-term, online learning hurt most kids, the idea that someone can project lost earnings due to that event seems to me to be quite a stretch. Call me a skeptic.

    1. LarrytheG Avatar
      LarrytheG

      we agree. And so do a number of other economists who don’t agree either.

    2. f/k/a_tmtfairfax Avatar
      f/k/a_tmtfairfax

      Or climate change studies from entities paid by taxpayers to find evidence of more climate change. Access to other people’s money is a very strong motivator.

      Most of learned in high school science classes not to alter or fill in missing data points to fix the desired result.

      While I believe that human activity has certainly contributed to warming, I question the validity of research by those paid to find specific results.

      The education study seems equally contrived.

      1. I say you are on-topic!

        If both analogies and snark are banned, this place will not be much fun.

    3. You are a skeptic.

      😉

  20. DJRippert Avatar
    DJRippert

    The real issue is that the lockdowns of schools were unnecessary. Or, at least, overdone.

    The theory was that children would get sick and die or that they would carry the disease home and infect their parents who would then get sick and die.

    I don’t hear any studies showing that schools which remained open or opened back up early suffered the predicted disasters.

    Those supporting long lockdowns screamed that those opposed were ignoring “the science”. It seems that “the science” was wrong. Why? Because it wasn’t science. It was conjecture by scientists. Big difference.

    In fairness, the vast majority of people who supported long lockdowns did so with pure motives. They believed the doomsayers who quoted “the science”.

    Unfortunately, it is the children who suffer.

    1. f/k/a_tmtfairfax Avatar
      f/k/a_tmtfairfax

      Agree. But it seems a big stretch to be able to quantify future lost earnings.

      BTW, both my brother and niece taught in the classroom for most of the pandemic period. Public schools too.

      1. LarrytheG Avatar
        LarrytheG

        and those that stayed in person ALSO showed learning losses! How do we understand that?

      2. But it seems a big stretch to be able to quantify future lost earnings.

        Yes. It’s somewhat analogous to trying to prove a negative.

        Fifty years from now, it will not be possible to accurately determine what caused people to not earn an amount of money some statistician fifty years before said they should have earned; never mind being able to accurately predict today what will cause people 50 years from now to not earn an amount of money a statistician says they should be able to earn.

    2. James McCarthy Avatar
      James McCarthy

      However speculative one may regard the info on quarantining, which risk takers and decision makers should have opposed such measures with lives at stake? The death toll was not overdone.

    3. Matt Adams Avatar
      Matt Adams

      It was accurately described in the beginning, yet censored for now towing the status quo. The old, the infirm (known or unknown) and the morbidly obesce we’re those likely to suffer bad outcomes.

      That was inconvenient because it went against body positivity, which while productive is counterproductive to a healthy society.

Leave a Reply