Interesting chart (the second one) posted by Lane Kenworthy today and some interesting debates and here on when the US healthcare system performance began to look different from the rest of the world. Kenworthy’s chart depicts the relationship between health expenditures per capita and life expectancy.

Here’s my contribution from a while back, while I was running SignalHealth.

When did the US begin to look different?

1980, give or take a few years, is about the time that the proportion of the US population with health insurance stopped growing and started declining. The data prior to that time aren’t great and clearly after stabilizing in the late 1950s and early 1960s, there would have been a large step increase with the advent of Medicare and Medicaid beginning in 1966. But then, with inflation in service costs and thus premiums and tax burden for public programs, the cost of coverage began to out run growth in economic capacity (however you want to measure that on either a micro or macro basis). Then came the inflection point and more people began draining out of the system than entered it.

No doubt, there are a number of factors that contributed. But we do know that coverage makes a difference in access to care.

Here are some posts I wrote back around 2005 about the impending acceleration in the loss of coverage and the crisis that would inevitably lead to political action:

Care and creation of the uninsured. An extremely important, but usually ignored aspect of this issue is that “the worse it gets, the worse it gets.” That’s part of the reason the proportion of uninsured is rising.

As more uncompensated care is provided it is subsidized either informally through price increase based cost-shifting or formally through such mechanisms are state indigent care pools, the costs of which are borne by health insurers and/or employers. Of course, as that happens, premiums rise. As premiums rise – especially in relation to income or other economic resources – coverage falls. That in turn, creates more uncompensated care … take a look at the Self-Reinforcing Coverage Loop – a vicious circle. Insurer/employer based subsidy programs are thus inherently self-defeating.

bdcc_coverage_loop.jpg

Why might coverage loss accelerate … and what then?. In my last post, regarding the loss of coverage and uncompensated care, I mentioned: As premiums rise – especially in relation to income or other economic resources – coverage falls.

The past decade at least, I’ve been convinced that this is the primary underlying cause of health insurance loss. Call it what you may, dress it up as much as you like, blame employers or public policy makers, coverage has simply gotten too expensive relative to the economic resources that pays for it. All you have to do is note that health care costs and therefore health coverage costs (both public and private) have been rising consistently faster than economy overall and thus the rest of the economy.

Think in terms of a ratio of coverage costs to other economic resources. Regardless of the specific measures you use, the more that ratio rises, the more people will drop out or be dropped out of the insurance system.

And as that ratio begins to cut at higher and higher income levels, it will affect an increasing proportions of the population. I don’t know where the inflection point on the upslope (the left side of the peak) is in this graphic, but until we reach it, each increase in that ratio will affect an accelerating number of people. Then the rate of increase will fall, but the numbers of newly uninsured will continue to increase until we hit the peak.Population_insurance_distribution

Think of the Cutting Edge “Coverage Blades” in the graphic moving to the right and slicing off increasing numbers of currently insured.

Of course, when we do hit the peak, we will be deep into the middle class – being uninsured. At some point just to the left of the peak, I expect we will arrive at the political crisis and deal with the issue.

But, like earthquakes in California, while we know this crisis is coming, we don’t know exactly when.

population_insurance_distribution.jpg

It’s only a matter of when. This post is about behavioral change, particularly as it relates to health economics and the healthcare system. The essentials, the mechanics if you will, apply similarly to individuals, organizations, and the place of healthcare in general economy.

The next couple of posts will be based on it and I’ll likely refer to it many times in the future.

Before we get started, answer the following questions:

If you and your family had (or have) to pay for your healthcare coverage directly, at what percentage of your family income would you be forced to seek less expensive coverage, even if it were more restrictive? Answer: __________
If you and your family had (or have) to pay for your healthcare coverage directly, at what percentage of your family income would you be forced to drop your coverage? Answer: __________

Put your answers aside for a moment. We’ll return to them shortly.

Some of this may look fairly complex, but it’s not really. Just follow it step-by-step and you’ll see what’s going on both in the post and, more importantly, you will see how trouble is inevitable or perhaps why we’re already in trouble. So don’t bail out on me quite yet.

The economy grows and healthcare costs grow. When healthcare costs grow, premiums, employer burdens, and taxpayer burdens grow in parallel as well.

Take a look at the following Chart 1, The Impending US Healthcare Crisis: Why it is Only a Matter of When: What do we see?

We see economic growth E (with the random ups and downs taken out for simplicity). As is typically the case, it curves upward, representing exponential or compounding growth. We see D “$ at Some Percentage of the Economy” which parallels the economy itself. For example, this curve might represent 10 or 20 percent of the economy. Because it’s a percentage, it grows at exactly the same rate as the economy. We also see H, Healthcare Spending. Note that, consistent with a lengthening history in the US, the rate of growth is faster than the economy (E) and faster than at $ at Some Percentage of the Economy (D). Because it grows faster, eventually it overtakes and crosses D at a point we have labeled “Decision/Behavioral Change Threshold.” Then something changes.

Now what might that “Decision/Behavioral Change Threshold” be? Let’s return to the questions you answered at the beginning. (You did answer them didn’t you?) Let’s assume that you answered that at 15 percent of your family income you would have to drop coverage. Beyond that, regardless of the risk, it would be just too expensive to pay the monthly premiums. Your family circumstances and decision behavioral change are represented in Chart 1. Here’s how. Your income is represented by Economy (E). Your 15 percent personal decision threshold is represented by “$ at Some Percentage of the Economy” (D). Premium growth is represented by Healthcare Spending (H). You drop coverage where the lines cross and at time T1.

You may use the same chart to think of movement to more restrictive coverage. Assume that as the cost of coverage approached your 15 percent threshold (say at 12 or 13 percent), straining to maintain some form of protection, you might seek less expensive coverage. The chart will work in exactly the same way.

Now what happens if economic growth slows, healthcare costs accelerate, or you ability to devote a set percentage of your economic capacity to healthcare diminishes?

hc_as__pct_of_economy_base_c1.jpg

Chart 2 shows slower economic growth combined with healthcare cost growth the same as in the base case in Chart 1. hc_as__pct_of_economy_slow_c2.jpg

Chart 3 shows the base case economic growth and even faster healthcare cost growth. hc_as__pct_of_economy_fast_hc_c3.jpg

Chart 4 shows a lower Decision/Behavior Change Threshold. For example, if you would be forced to drop coverage at five or 10 rather than 15 percent of your family income as would generally be the case for lower income families, then the Threshold would be lower relative to the family Economy. What Charts 2 through 4 have in common is that all produce a change in behavior at an earlier time. Slower economic growth, faster healthcare cost growth, or a lower percentage of your economic resources that can be devoted to healthcare all cause you to drop coverage earlier. hc_as__pct_of_economy_diff_thresh_c4.jpg

Pick your own markets of economic growth. Pick your own measures of economic capacity. Consider this at the most macro-economic level or at the personal level. The same effects result.

If a system consistently grows faster than a larger system in which it resides, eventually an intersection is reached at which the behaviors of one or the other, or both change less the larger system be overwhelmed. But even being overwhelmed will generally yield a behavioral change (think uncontrolled cancerous cells overwhelming an organ or body and eventually killing it).

What has been happening is that healthcare costs consistently outgrow the rest of the economy – which pays for healthcare. Increasingly individuals, families, employers, and governments capacity to carry the cost are being overwhelmed. So their behavior changes. In some cases, they seek increasingly stronger means of maintaining a financial relationship with the healthcare system. But in other instances, their behavioral change is to simply walk away. Thus, increasing numbers of uninsured.

We don’t all reach the intersection at the same time or at the same rate. The same is true for employers and governments. But, sooner or later, given the underlying structural relationships and dynamics, the vast majority of individuals and the economy overall will arrive at that intersection. Tennessee and other states arrived at it earlier this year with their Medicaid programs. General Motors appears to be at the doorstep right now. In contrast, for the time being, it appears that the Federal government has chosen to spend a higher portion of its economic capacity (tax revenue plus debt) on Medicare. Yet, when we consider projections for the Medicare Trust Fund, clearly the day of reckoning is not that far away.

I’m going to make a brazen assertion here: If you want to understand what is happening in healthcare and if you want to anticipate what will be happening, then the single most important measurements to track are overall healthcare spending, the carrying capacity of the economy that’s paying for it (whether family, regional, employer, governmental, or macro-economic), and their relationships to one another. With healthcare cost growth rates exceeding economic growth rates, the larger the gap between growth rates in healthcare spending and economic capacity, the closer the day of reckoning.

These economic forces were hard at work long before 2005. I strongly suspect (believe) that the inflection point at which time the proportion of people with coverage stopped growing and started declining was 1980, give or take a few years. And the inflection point was the result of pretty simple economics. We we spending an increasing proportion of our economic capacity on health care. And we were individual-by-individual, employer-by-employer reducing our participation in the system because we couldn’t afford it any more.

I’ve modeled this dynamic using a technology that I can now put up on the web. Maybe I’ll take a crack at that when I’ve got some other things off my plate.

What Kenworthy has done is raised a question regarding life expectancy and the increasing distance between the US and other advanced economies. It’s well worth further exploration.

{ Comments on this entry are closed }

No budget detail yet, but as part of its budget process, Ulster County has determined to get out of the nursing home business. They’ve established a county-run limited development corporation (LDC) as an intermediary to smooth the process.

More details later.

{ Comments on this entry are closed }

My last post, Yes, Rockland County is in Trouble, was devoted to the release by New York’s Comptroller, Tom DiNapoli, of an Audit (PDF) of Rockland’s financial condition.

Clearly, the Comptroller and his staff have been watching closely enough to start their audit earlier this year. Good for them. Carefully done audits take time and a quick scan suggests this one was. So they are to be commended for this. I strongly suspect that they’ve been watching Rockland for a while and it would be interesting to know what specifically triggered the audit.

However (there’s always a “but,” huh?), as the audit itself notes, this did not emerge instantly. Here’s a telling quote from the audit (emphasis added):

The County’s financial condition has deteriorated over a period of several years. In addition to external factors – namely, a significant tax write-off that caused a deficit in the general fund balance in recent years – the budgeting practices of County officials have left the County with a general fund deficit of over $50 million, significant debt, and no comprehensive plan to ensure sufficient recurring revenues to finance recurring expenditures.

“The County’s financial condition has deteriorated over several years.” This leads me to a few questions:

    Might it have been possible to release this audit just a bit earlier, like a month ago? Perhaps the Comptroller decided it was necessary to release it now because Rockland has just put another straw on his back but continuing on the same path as the past few years. Perhaps, there was no way to get a carefully done audit done earlier. Nevertheless, it’s interesting to speculate on how an earlier release might have influenced the local budget debates, internal and external.

    Might it have been possible for more information to be more available to the public well further in advance?

Clearly the answer to the latter question is yes. Making more information, including visuals, available to the public earlier – early enough to influence decision making – does not have to be in the form of reports, and the Comptroller does put municipal financial data online (more kudos for that), but today’s technologies allow much more rapid and much clearer conversion of raw data into useful information. We’ve already done a bit of that (look here, here, here and here also) and we’ll be doing more in the coming months.

In any case, I’m glad the Comptroller’s on the case. Who’s next?

{ Comments on this entry are closed }

Yesterday, after taking a quick look at the budget numbers and actions for 2012, I asked the question, is Rockland County (NY) in trouble?

The answer is yes.

This morning, the NYS Comptroller released the following (the emphases are mine):

DiNAPOLI: POOR BUDGET PRACTICES HURTING ROCKLAND COUNTY
County Has Incurred Large Debt and General Fund Deficit

Budgeting practices by Rockland County have left it with an accumulated deficit of more than $50 million and hundreds of millions in debt, according to an audit released today by New York State Comptroller Thomas P. DiNapoli.

“Since 2006, Rockland County has seen its deficit skyrocket for several reasons, including overly optimistic revenue projections, failing to repay advances of general fund monies to other county funds and the loss of a major property tax case,” DiNapoli said. “County officials need to make some hard decisions to put together a real budget to get out of this mess. Rockland needs to live within its means for the sake of county taxpayers.”

The county’s general fund total year-end balance dropped from $23.3 million on Dec. 31, 2008 to a deficit of $3.9 million for 2010. Because there was a $52 million deficit in the unappropriated fund balance for 2010, the county’s $48 million in reserves are unfunded.

DiNapoli’s auditors, who reviewed the results of operations for fiscal years 2006 through 2009, found the county’s budgeting practice of overestimating revenue necessitated tens of millions of cash advances between operating funds that were not repaid. Sales and mortgage tax revenues were overestimated by more than $70 million during the audit period.

In addition, the county’s nursing home had operating deficits in all five years covered by the audit, an accumulated deficit that now totals more than $86 million. The county also has had to write-off more than $20 million in unpaid real property taxes and penalties owed by a major taxpayer.

The county was hurt by delayed state aid payments, averaging more than $38 million in each of the fiscal years 2006 through 2010. Because these payments were not received until after 60 days following the county’s fiscal year end, the county could not use those moneys for the budgeted years and had to issue short-term debt.

Rockland issued $343 million in debt in 2007 and $100 million or more each year from 2008 to 2011. The county also prematurely budgeted $17.8 million in revenues for 2011 from the planned sale of its nursing home and medical facility, operated under the home and infirmary fund, to a proposed public benefit corporation (PBC) to be established with state approval.

As of August 2011, the state legislature had not granted the requested approval for establishing a PBC, and the county had to issue a $17.8 million deficiency note for which the taxpayers are liable next fiscal year.

DiNapoli’s auditors called on the county to:

  • Evaluate the operations of its key operating funds and develop a comprehensive plan to address the operating deficits;
    Ensure that advances from the general fund are repaid in a timely manner;
  • Realistically budget for sales and mortgage tax revenues or reduce general fund expenditures to levels that can be financed by recurring revenue sources;
  • Include in the budget only the revenue from state aid that will be available within the fiscal year or within 60 days after the fiscal year end; and
  • Obtain a valuation of the facilities and operations of the home and infirmary fund and develop a realistic financial plan to either divest the facilities or make them financially viable.

County officials generally agreed with the audit’s findings and said they have begun taking steps to correct the county’s financial problems.

A copy of the audit can be found here (PDF).

Specific “tells”? Habits of overestimating revenue and inability to deal realistically with their nursing home.

{ Comments on this entry are closed }

This is not merely clever teaching. It’s brilliant at a deep level because it brings a human story to economics and economics to a human story. It combines reading and analyzing a classic novel about the Great Depression with a technical economics textbook.

Combining dialog with demand curves, Stephen Ziliak, Trustee and Professor of Economics at Roosevelt University-Chicago, teaches introductory microeconomics using John Steinbeck’s The Grapes of Wrath.

Here is Mark Thoma’s summary description and interview of Ziliak. It’s well worth reading.

Read the syllabus (PDF) too. It includes the following:

Mostly, however, the course is an introduction to microeconomic ways of thinking. Our course introduces a new grammar, if you will – an economic grammar of scarcity, competition, relative price, opportunity cost, supply and demand, efficiency, and equilibrium, to name a few.

I first read Grapes of Wrath as a college freshman. It introduced me to life as I’d never imagined it. My parents had seen the Great Depression, but never discussed it. It was compelling, an eye-opener on an individual scale, that the history books I’d read had not captured. It’s time to re-read it with new eyes.

{ Comments on this entry are closed }

Going back to 1991, Genesee County, NY has a nice history of its key overall budget figures here.

In Firefox, the table runs off the side of the browser page, but it works fine in the latest versions of Chrome and Safari. I didn’t check it with Internet Explorer. In any case, it’s readable.

It would be really nice if it were also available as a downloadable file that could be worked with programs such as Excel and it would be good to have revenue figures as well, but it’s great having the history readily available to the public. Links are also available for the 2011 budget and the recommended 2012 budget, including program detail.

{ Comments on this entry are closed }

The results are beginning to come in on New York’s county budgets for 2012. Got a long way to go, but figured to start posting rather than wait for a more complete report.

  • Albany
  • 8.7 percent increase ($6,572,032) to $81,971,723. This is 19.5 percent increase in three years. Set stage for pay increase for legislators and other elected officials.

  • Allegany
  • Broome
  • Cattaraugus
  • Cayuga
  • Chautauqua
  • Chemung
  • Chenango
  • Clinton
  • Columbia
  • Cortland
  • Delaware
  • Dutchess
  • Property tax levy up 3.2 percent to $103.8 million. Total expenditures $411.4 million. Uses $25 million in fund balance. Lays off 22 mental health workers from continuing day treatment program. Controversy with judges over providing only two attorneys to handle Family Court cases. Here’s good coverage by Poughkeepsie Journal.

  • Erie
  • Essex
  • Franklin
  • Property tax levy increase was 1.98%. Not using general fund balance but using $1.3 million of other fund balance. Using $765 thousand in Nursing Home fund balance (which is equal to the budgeted operating loss). The NH will finish 2011 with a positive fund balance because of the receipt of IGT funds. Planning on using $547 thousand in capital fund balance that are funds leftover from capital projects. Layoffs reduced from 15 in tentative budget to 10. Did a program using a Voluntary Separation Incentive (VSI). Eliminated $90 thousand for new highway equipment. Plan on $90 thousand in reductions at the jail related to reduced prisoner lodging costs and a reduction in funding for electronic home monitoring because they have reached programmatic limits – i.e. cannot assign violent offenders or sex offenders. Looking ahead the projected gap for 2013 is about $3.8 million which is made up of the $1.3 million use of fund balances plus about $3.5 million offset by about a $1.0 million savings since 2012 is the final year of debt service payments on the county jail.

  • Fulton
  • Genesee
  • Greene
  • Hamilton
  • Herkimer
  • Jefferson
  • Lewis
  • Livingston
  • Madison
  • 1.996 percent increase in property tax to $29,460,932. Spending down 3.3 percent to $102.6 million. First tax rate increase in five years (0.5 percent). December 6, news story from Post-Standard here.

  • Monroe
  • Montgomery
  • Nassau
  • Niagara
  • Oneida
  • No change in property tax levy. Deferred (amortized under State program) pension payments. Passing community college costs on to municipalities. County contribution to legislator’s health benefits dropped. Services reduced at health clinic. News story from the Utica Observer-Dispatch here.

  • Onondaga
  • Ontario
  • Orange
  • Orleans
  • Oswego
  • Otsego
  • Putnam
  • Rensselaer
  • News story here.

  • Rockland
  • Note: I’ve not yet talked to any official in Rockland and am relying on news reports. Generally, that’s ok, but Rockland’s budget issues appear to rise to a higher than average level. Increase the county property tax by about 30 percent to repay an $18 million loan, which represents about 28 percent of the increase, and to raise about $1.25 million in additional revenue, about 2 percent more than the $61.7 million budgeted in 2011. Decrease spending from $709.1 million to $701.8 million. Here is a news story worth reading. Rockland’s budget depends on a sales tax increase initially to avoid 568 layoffs (though 40 would still lose their jobs). The $18 million loan was evidently taken out in anticipation of selling the hospital and nursing home. Evidently, they are also planning on the sale of bonds to cover part of the deficit. This would require extra State oversight and the risk of a financial control board being imposed on the County. This is one worth watching.

  • St. Lawrence
  • Saratoga
  • Schenectady
  • Schoharie
  • Schuyler
  • Seneca
  • Steuben
  • Suffolk
  • Sullivan
  • Tioga
  • Tompkins
  • Property tax up 3.99 percent. Eliminates 25 jobs due to decision to sell the home health agency. News story from the Cornell Daily Sun here.

  • Ulster
  • Here’s story on Ulster’s leaving the nursing home business.

  • Warren
  • Washington
  • Property tax increase 1.176 percent or $331,272 to a total property tax levy of $28.505 million tax levy. Considering sale of nursing home and certified home health agency, both of which are losing money. News story from Bennington Banner here.

  • Wayne
  • Westchester
  • News story regarding controversial pre-payment of 2012 pension costs in 2011 by County Executive Astorino.

  • Wyoming
  • Yates

If you’ve got information that we can use, please pass it on. Thanks.

{ Comments on this entry are closed }

My friend and colleague, Tim Maniccia, who runs Policy Innovation, noted some interesting connections yesterday.

What triggered Tim’s observations was the bankruptcy of American Airlines.

The first thing he noted was that American’s four pension plans cover almost 130,000 participants, but they have only $8.3 billion in assets to cover $18.5 billion in benefits. Were American to dissolve or allowed through bankruptcy to shed its obligations, the Pension Benefit Guaranty Corporation, already in financial trouble, would be responsible for paying about $17 billion in benefits and at least $1 billion in benefits would be lost.

My observation? For all the noise this past year or so, it’s not only the public sector retirement plans that are underfunded. Talk about moral hazard …

And the especially interesting place Tim took it was the potential ripple to bond financed public projects. As reported in Airwise, Fitch Ratings picked it up as well.

The finances of three major US airports could weaken from the bankruptcy of American Airlines and its parent company, Fitch Ratings said on Friday as it revised rating outlooks on some of their debt.

An estimated USD $4 billion in joint revenue improvement bonds for Dallas-Fort Worth, USD $6 billion of aviation revenue bonds sold on behalf of Miami Airport and USD $6.5 billion of airport revenues bonds issued on behalf of Chicago O’Hare Airport are particularly vulnerable from the Chapter 11 filings made by AMR and its American Airlines subsidiary, Fitch said.

Fitch, one of the three major Wall Street credit rating agencies, affirmed the ratings on the debt. The DFW bonds are rated A-plus, Miami at A, and Chicago O’Hare at AA-plus.

Special facility revenue bonds are solely the responsibility of the airline and have lost popularity in recent years after United Airlines defaulted on USD $1.1 billion of the debt when if filed for bankruptcy protection in 2002.

Fitch said “credit concerns do exist for individual airports, particularly facilities where American operates major connecting hubs. The airline companies filed for bankruptcy on Tuesday, an act that “poses only minor immediate risks to most general airport revenue bonds,” Fitch said.

American has major hubs in Miami, Dallas an at Chicago O’Hare. American and its affiliate account for 85 percent of enplanements at DFW and 68 percent at MIA. American also has a substantial presence in Kansas City and at Midway.

The public and the private spheres of today’s economies are intimately, inextricably intertwined. Rhetoric about keeping government out of business matters are naive and risk missing both hazards and opportunities.

You can follow Tim on Twitter at @tmaniccia.

{ Comments on this entry are closed }

Best Way to Herd Cats

by John W Rodat on December 6, 2011

From Keith W. Boone on Twitter (@motorcycle_guy):

“Best way to herd cats? Move the food bowl.”

{ Comments on this entry are closed }

The Tax Policy Center, a joint effort of the Urban Institute and the Brookings Institution dispenses with the question of how much how much tax revenue is generated at higher levels of tax rates. It also undermines some of the supply-side scripture that high marginal tax rates are counterproductive. The evidence here suggests that the marginal income tax rates could be higher without generating less tax revenue.

Massive federal deficits and rising income inequality make it worth asking: just how important were those higher rates in raising revenue? To answer to that question, we compiled IRS data on the amount of tax generated at each statutory marginal tax rate, dating back to 1958.

This is what we found: Those high rates really did raise revenue. Although only a small fraction of tax returns were affected by very high rates, the taxes paid at those rates accounted for a substantial portion of individual income taxes paid. Between 1958 and 1986, an average of 14% of individual income tax revenues were generated at rates above 39.6 percent, and an average of 6% of revenues were generated at rates above 50 percent. At their peak in 1986, rates above 39.6 percent accounted for an impressive 23% of income tax revenue.

Theirs is not an oversimplified, “tax the rich” perspective:

It’s clear that we can’t solve the deficit problem just by increasing taxes at the top. Unless we are willing to make massive cuts in retirement and health benefits that many American families rely on, our revenue needs will be too great to avoid broader tax increases. And higher taxes should come from a reformed – fairer, simpler and more efficient – tax system. But it may not be unreasonable to ask taxpayers at the very top of the income distribution, who have received most of the income gains over the past 30 years, to pay income taxes at levels comparable to those paid prior to the 1980s.

In the comments, Michael Binder, has an interesting perspective (though perhaps awkwardly worded), I’d never thought of regarding the peak marginal rate of 91 percent that was in effect for decades, namely pay employees more to avoid paying more in taxes:

The chief aim of a 91% rate is not to raise revenue but to prevent the payment of income and encourage higher salaries for the workforce, since most businesses would rather pay their workers than the government. Not taking this income in taxes encourages management to decrease worker wages and keep the proceeds. Of course, making such a point is considered class warfare – especially by people who are being funded by people formerly paying a 91% tax rate.

As always, use the damn data.

{ Comments on this entry are closed }