6 ms·

The IRS has over 74,000 employees and a budget of nearly $14B. What would be the appropriate level of funding and size of staff for this organization?

paxys 4y ago
There's an easy way to go about it – keep increasing funding while every incremental dollar still has a positive ROI, and then stop. There's probably no other government agency where you can measure the value of an employee so objectively. By all measures we are very far from that limit.
- sokoloff 4y ago
  Suppose each new IRS agent costs $100K all-up. If they bring in just $100,001 (across direct enforcement and secondary compliance), is that a good expansion of the federal government?
  - paxys 4y ago
    What if each agent cost $100K and was bringing in $900K? Would that be worth it to you? Because that's what the Congressional Budget Office's estimate actually is. https://www.cbo.gov/publication/57444 https://www.cbo.gov/publication/57444
    - sokoloff 4y ago
      > What if each agent cost $100K and was bringing in $900K? From the first paragraph of your link: CBO estimates that portions of the Administration’s proposal to increase funding for the IRS by $80 billion over the 2022–2031 period would increase revenues by approximately $200 billion over those 10 years. Later in the article is the argument that the IRS says previous multiples were 5-9x at peak and then immediately goes into why those numbers would not hold for future incremental investments in enforcement. So, no, CBO doesn't estimate that it's a 9:1 return.
      - 29083011397778 4y ago
        ~2.5:1 is a far cry from your $100,000 cost with $1 return though, and GPs point still stands: increasing funding, even if just until the return is ~1.5:1 (less than the guesstimated ~2.5:1), is clearly a sound investment.
        sokoloff 4y ago
        That wasn’t my estimate of the actual IRS plans but rather a clarifying/rhetorical question of an up-thread hypothesis that we should expand the IRS until there is de minimus marginal income. Your issue is with their idea, I think.
        mschuster91 4y ago
        Even a 1.1:1 ROI is still a sound investment - non-compliance with taxes usually closely coincides with other criminal activity like wage fraud, employing undocumented people, selling re-labeled expired meat or environmental crimes like improper waste disposal. By regularly auditing especially large employers, the incentive for companies to commit these crimes because they'll get away with it suddenly also goes away. Simply said, if you can make an extra million dollars a year in profit, your company will be fined 100k if caught, and your company has an audit chance of 1 in 20 years, you can expect 19.9 million dollars in extra profit over that timeframe. Now, assuming an audit every five years, you're at only 4.9 million dollars of extra profit and the high chance of getting caught again and getting a harsher sentence as a repeat offender. Regular tax audits have far-reaching side effects.
    - s1artibartfast 4y ago
      I am somewhat sympathetic to your view, but I wonder if these are simply the annualized costs or a more accurate reflection of net present value. Do they take into account pensions and promotions? If we hire 10k novice IRS agents, I imagine their annual cost will be quite low for the the first years and/or decades. however, At some point it will go negative.
      - mschuster91 4y ago
        > If we hire 10k novice IRS agents, I imagine their annual cost will be quite low for the the first years and/or decades. however, At some point it will go negative. We can extrapolate their cost though. Payment rosters and pension contributions for government jobs are public, which means you can do a worst-case estimate and go with the cost basis of the highest-paid position that they can reasonably achieve in their career. Additionally, a lot of them are replacements for people who will go into retirement or quit, so as people get promoted, other people leave, freeing up their salary budget for the new promotee. (Pension budgets are a different beast because they obviously depend on how long the person lives)
        s1artibartfast 4y ago
        I absolutely believe it can be done no question. I was just wondering if the number is actually reflect that. The cynic and me suspects that depends on whose numbers and which side of the opinion they are on. That said, I think for the most part this is a political issue because of the taxpayers. Many if not most people find taxes stressful and don't want more Auditors looking into theirs. You can say that they will only audit corporations or high value returns, but the little guys will always worry that it frees up more resources to go after them
  - xboxnolifes 4y ago
    Yes. That's also not an expansion of the federal government. That's just hiring to fill the responsibilities that the federal government already has.
    - sokoloff 4y ago
      If the IRS hires 87K more people than it currently has, ELI5 how that's not an expansion of the federal government.
      - paxys 4y ago
        Because the responsibilities of the government stay exactly the same.
        sokoloff 4y ago
        And the employment payments increase by close to $10B/yr from this expansion alone plus the number of employees (and therefore families prone to vote for expansion of federal employees and/or pay/benefits also increases). The federal government gets larger as a result of this. That might be OK, but we shouldn’t pretend it’s not the case.
      - mschuster91 4y ago
        A chunk of the 87k are simply replacement for old IRS staff retiring or quitting. And is it really an "expansion" when the IRS historically has had that amount of staff, got intentionally bled dry by Republican presidents to prevent the IRS from investigating their donors and friends, and is now only returning to the prior norm (and even then, likely not enough to actually fulfill the duties)?
        sokoloff 4y ago
        If it was a contraction when the IRS got smaller, it seems logical to see it as an expansion when it gets larger.
mschuster91 4y ago
Whatever is needed to do their job in a fair manner, obviously. Current plans are at 87.000 new employees [1], although a bunch of that is replacement for workers aging out or quitting due to being overworked [1]. The IRS actually used to have 117k employees back in 1992 with fewer taxpayers [2] (both due to population growth and due to women entering the workforce as a result of the breakdown of gender roles / SAHMs). It's estimated that bringing the IRS up to decent staffing levels could bring in 204 billion dollars at a profit of 124 billion dollars - so the ROI on each new agent or staff is about 50%. That's a margin way better than most companies have! [1] https://www.cnbc.com/2022/08/31/irs-is-not-hiring-an-army-of-auditors-whats-really-happening.html https://www.cnbc.com/2022/08/31/irs-is-not-hiring-an-army-of... [2] https://www.politico.com/news/2022/08/16/irs-spending-new-money-00051523 https://www.politico.com/news/2022/08/16/irs-spending-new-mo...
- NovemberWhiskey 4y ago
  The IRS also had to deal with basically all filings being on paper and very basic technology support, which is very definitely not the case today.

paxys 4y ago

There's an easy way to go about it – keep increasing funding while every incremental dollar still has a positive ROI, and then stop. There's probably no other government agency where you can measure the value of an employee so objectively. By all measures we are very far from that limit.

sokoloff 4y ago
Suppose each new IRS agent costs $100K all-up. If they bring in just $100,001 (across direct enforcement and secondary compliance), is that a good expansion of the federal government?
- paxys 4y ago
  What if each agent cost $100K and was bringing in $900K? Would that be worth it to you? Because that's what the Congressional Budget Office's estimate actually is. https://www.cbo.gov/publication/57444 https://www.cbo.gov/publication/57444
  - sokoloff 4y ago
    > What if each agent cost $100K and was bringing in $900K? From the first paragraph of your link: CBO estimates that portions of the Administration’s proposal to increase funding for the IRS by $80 billion over the 2022–2031 period would increase revenues by approximately $200 billion over those 10 years. Later in the article is the argument that the IRS says previous multiples were 5-9x at peak and then immediately goes into why those numbers would not hold for future incremental investments in enforcement. So, no, CBO doesn't estimate that it's a 9:1 return.
    - 29083011397778 4y ago
      ~2.5:1 is a far cry from your $100,000 cost with $1 return though, and GPs point still stands: increasing funding, even if just until the return is ~1.5:1 (less than the guesstimated ~2.5:1), is clearly a sound investment.
      - sokoloff 4y ago
        That wasn’t my estimate of the actual IRS plans but rather a clarifying/rhetorical question of an up-thread hypothesis that we should expand the IRS until there is de minimus marginal income. Your issue is with their idea, I think.
      - mschuster91 4y ago
        Even a 1.1:1 ROI is still a sound investment - non-compliance with taxes usually closely coincides with other criminal activity like wage fraud, employing undocumented people, selling re-labeled expired meat or environmental crimes like improper waste disposal. By regularly auditing especially large employers, the incentive for companies to commit these crimes because they'll get away with it suddenly also goes away. Simply said, if you can make an extra million dollars a year in profit, your company will be fined 100k if caught, and your company has an audit chance of 1 in 20 years, you can expect 19.9 million dollars in extra profit over that timeframe. Now, assuming an audit every five years, you're at only 4.9 million dollars of extra profit and the high chance of getting caught again and getting a harsher sentence as a repeat offender. Regular tax audits have far-reaching side effects.
  - s1artibartfast 4y ago
    I am somewhat sympathetic to your view, but I wonder if these are simply the annualized costs or a more accurate reflection of net present value. Do they take into account pensions and promotions? If we hire 10k novice IRS agents, I imagine their annual cost will be quite low for the the first years and/or decades. however, At some point it will go negative.
    - mschuster91 4y ago
      > If we hire 10k novice IRS agents, I imagine their annual cost will be quite low for the the first years and/or decades. however, At some point it will go negative. We can extrapolate their cost though. Payment rosters and pension contributions for government jobs are public, which means you can do a worst-case estimate and go with the cost basis of the highest-paid position that they can reasonably achieve in their career. Additionally, a lot of them are replacements for people who will go into retirement or quit, so as people get promoted, other people leave, freeing up their salary budget for the new promotee. (Pension budgets are a different beast because they obviously depend on how long the person lives)
      - s1artibartfast 4y ago
        I absolutely believe it can be done no question. I was just wondering if the number is actually reflect that. The cynic and me suspects that depends on whose numbers and which side of the opinion they are on. That said, I think for the most part this is a political issue because of the taxpayers. Many if not most people find taxes stressful and don't want more Auditors looking into theirs. You can say that they will only audit corporations or high value returns, but the little guys will always worry that it frees up more resources to go after them
- xboxnolifes 4y ago
  Yes. That's also not an expansion of the federal government. That's just hiring to fill the responsibilities that the federal government already has.
  - sokoloff 4y ago
    If the IRS hires 87K more people than it currently has, ELI5 how that's not an expansion of the federal government.
    - paxys 4y ago
      Because the responsibilities of the government stay exactly the same.
      - sokoloff 4y ago
        And the employment payments increase by close to $10B/yr from this expansion alone plus the number of employees (and therefore families prone to vote for expansion of federal employees and/or pay/benefits also increases). The federal government gets larger as a result of this. That might be OK, but we shouldn’t pretend it’s not the case.
    - mschuster91 4y ago
      A chunk of the 87k are simply replacement for old IRS staff retiring or quitting. And is it really an "expansion" when the IRS historically has had that amount of staff, got intentionally bled dry by Republican presidents to prevent the IRS from investigating their donors and friends, and is now only returning to the prior norm (and even then, likely not enough to actually fulfill the duties)?
      - sokoloff 4y ago
        If it was a contraction when the IRS got smaller, it seems logical to see it as an expansion when it gets larger.

mschuster91 4y ago

Whatever is needed to do their job in a fair manner, obviously. Current plans are at 87.000 new employees [1], although a bunch of that is replacement for workers aging out or quitting due to being overworked [1]. The IRS actually used to have 117k employees back in 1992 with fewer taxpayers [2] (both due to population growth and due to women entering the workforce as a result of the breakdown of gender roles / SAHMs). It's estimated that bringing the IRS up to decent staffing levels could bring in 204 billion dollars at a profit of 124 billion dollars - so the ROI on each new agent or staff is about 50%. That's a margin way better than most companies have! [1] https://www.cnbc.com/2022/08/31/irs-is-not-hiring-an-army-of-auditors-whats-really-happening.html https://www.cnbc.com/2022/08/31/irs-is-not-hiring-an-army-of... [2] https://www.politico.com/news/2022/08/16/irs-spending-new-money-00051523 https://www.politico.com/news/2022/08/16/irs-spending-new-mo...

NovemberWhiskey 4y ago
The IRS also had to deal with basically all filings being on paper and very basic technology support, which is very definitely not the case today.