Is workers’ compensation spending on healthcare significant?...Would a "single-payer" system make a difference?

I’ve received a lot of questions recently on workers’ compensation healthcare spending in Canada, the US and other countries.  These questions appear have arisen as several US states and political analysts have proposed consideration of “single payer” healthcare systems for their jurisdictions. 

Questions include:

• Is healthcare spending similar across nations? 
• Is the workers’ compensation part of that spending similar across nations? 
• Does it matter to workers’ compensation if there is a single payer system in place?   

At the California state Department of Workers’ Compensation (DWC) Educational Conference, my presentation highlighted workers’ compensation healthcare spending.  Time and format restricted the depth of that presentation.  Hopefully the following background will answer in greater depth some the many questions on this topic and help policy makers better understand the scope and context for their deliberations.

Healthcare spending is a big part of the economy

National spending on healthcare is difficult to compare across economies but may be expressed relative to Gross Domestic Product (GDP) or as an average cost per person (see https://www.cihi.ca/en/how-does-canadas-health-spending-compare-internationally

for comparisons based on 2015 data).  

The Organization for Economic Cooperation and Development (OECD) estimates average healthcare spending at about 9% of GDP with the US the top spender at about 17%; Canada, Australia and New Zealand are all slightly above the OECD average with national healthcare expenditures between 9 and 11% of GDP.

Healthcare spending comes from two general sources:  public funds and private sources.  Private funding includes out-of-pocket healthcare spending by individuals on medical supplies and services, co-pays or deductibles.  Spending by your privately-purchased extended health or dental plan is also considered private.  Public spending refers to spending from governments and public agencies.  The healthcare of military vets, grants for medical research and public medical insurance typically fall into the public spending category.  Workers’ compensation spending on healthcare may be in either category or both depending on the jurisdiction.  A state fund or provincial workers’ compensation board would have its healthcare spending grouped in the public category (often under the category of “social insurance” healthcare spending) while healthcare spending by private workers’ compensation insurers would generally be reflected in the private category. 

One might expect more spending will result in better health outcomes for the population.  There are many metrics on health outcome:  life expectancy, child mortality, disability-adjusted years of life, etc.  Many of these show some correlation with national health spending but there are exceptions.  Spending more does not always result in better health outcomes.  This chart compares life expectancy at birth and national healthcare spending [US in blue, Canada in Orange and Australia in white compared with all other OECD countries].

What portion of National Healthcare Expenditures are related to Workers’ Compensation?

Regardless of the workers’ compensation insurance arrangement (private insurer, exclusive public state or provincial workers’ compensation board) the healthcare expenditures by workers’ compensation systems are relatively small compared to the total national spending on healthcare.  Using data from a number of sources, workers’ compensation spending on healthcare accounts for approximately 1% to 2% of total national healthcare spending in the US, Canada and Australia.  [See figure at top of this post.]  The ranges for US and Canadian estimates are primarily related to the apportionment of administration costs.

In jurisdictions where hard currency amounts are actually expended by workers’ compensation systems, tracking and summation by reporting authorities is complex but relatively transparent.  Not all workers’ compensation systems (or other mandated employer liability arrangements covering occupational injury and disease) in the world cover medical costs; instead, healthcare spending arising from work-related injury and disease may be paid by the individual, included in mandated public or private health insurance coverage, or covered by a universal health plan.   

Where there is no hard currency audit trail for work-related healthcare spending, jurisdictions have to estimate the proportion of healthcare spending for work-related injuries and disease by other means.  The estimate of total healthcare spending related to work injury and disease in different jurisdictions will depend on what you include or exclude from the calculation but should be approximately equivalent to workers’ compensation spending on healthcare reported in jurisdictions where that model applies (and most workers are covered).

In the UK, the National Health Service (NHS) provides universal healthcare coverage for all injuries including work-related injuries, illness and disease.  The Health and Safety Executive (HSE) in the UK attributes the following health spending amounts [based on self-report]  to work-related injury, illness and disease:

• Individual: 
• Out of pocket health and rehabilitation expenses (including prescription charges, travel expenses, home expenses and funeral expenses      £ 83 Million
• Proportion of individual private health insurance premiums attributable to work related illness/injury  £ 21 Million
• Employer:  Proportion of corporate private health insurance premiums attributable to work related illness/injury  £ 97 Million
• Government/Taxpayer:  NHS treatment and rehabilitation costs (short and long term) £738 Million 

[from Costs to Britain of workplace fatalities and self-reported injuries and ill health, 2016/17,  Annex 2: Detailed breakdown of costs by cost bearer in 2016/17 (2016 prices), [Health and Rehabilitation category] Published 31 October 2018.
http://www.hse.gov.uk/statistics/pdf/cost-to-britain.pdf ]

Total current healthcare expenditure for the UK in 2016 was £191.7 billion, so these amounts represent about 0.5%-- a little less than the proportion estimated above for Canada, the US and Australia.  The difference is likely related to the set of spending amounts included as work-related. 

How significant is the healthcare expenditure component of workers’ compensation costs?

It may be tempting to dismiss workers’ compensation healthcare spending as a rounding error in terms of the national healthcare spending.  In my view, this would be a mistake.  Those of us who study or work in workers’ compensation systems recognize the cost of healthcare for injured workers as a big part of the total cost of workers’ compensation claims.  

Healthcare or medical aid spending accounts for more than half the current year claims expenditures by workers’ compensation insurers in the US.  In some states, the proportion is even greater.  In Canada, the data suggest a lower proportion, from about a quarter to a third of current year benefit expenditures paid in the year for claims from all years (for most jurisdictions).  It should be noted, however, that Canadian workers’ compensation benefits tend to be greater than those provided in the US (typically 85-90% of Net or spendable average earnings vs. the typical 66 2/3rds % of gross, generally higher maximum insurable limits, cost-of-living adjustments, etc. ); Canadian healthcare costs (including for many hospital services and prescriptions) also tend to be lower cost than those in the US thus making the healthcare component of current benefits paid appear somewhat lower.   

Why is this important?

Many jurisdictions are looking at their healthcare expenditures.  Proposals in the US include reforms from single-payer systems, 24-hour coverage, and other arrangements.  Proponents of one position or another may draw on Canada, the UK, Australia, or other national system as examples of what to do or not to do in any reform.  While there is no question that comparisons can highlight important differences and may even suggest opportunities, there are challenges in understanding other systems and the context in which they arise. 

One key point to remember is the relationship between claim cost and premium for workers’ compensation insurance. Under the insurance model used in the US, Canada and Australia, workers’ compensation premiums are designed to reflect indemnity payments for lost wages, other benefits (such as rehabilitation), and the costs of medical care that may be needed for a lifetime.  This structure is based on the premise that the industry (or enterprise) that gives rise to a work-related injury or disease ought to cover the associated diagnostic, treatment and medical/rehabilitation costs. Self-insurance provisions seek to confine costs to the insured entity and experience modification to premiums (x-mod, experience rated assessment, etc.) are intended to distribute costs more directly to the entity giving rise to them.  The workers’ compensation insurance cost (including the medical cost) reflected in the premium forms part of the incentive for increased prevention and improved return-to-work outcomes.  Failure to confine and fund the medical cost to the workers’ compensation system by definition will externalize those costs to someone else (often the tax payer or premium payers for other health insurance programs, the worker, worker’s family or community, other employers, or other employees of group insurance plans). 

Is the healthcare spending by workers’ compensation insurers handled similarly in the US, Australia, and Canada,?

A worker gets injured, goes to the doctor or hospital and gets treatment.  In the US, Canada and Australia, if the injury is work-related, payment for the fee for service by the attending physician or hospital service will usually be made by the workers’ compensation insurer, typically as a direct bill by the provider to the insurer. Identifying that insurer can be complicated by the number of factors.  There are often several potential insurers in most states and the responsible insurer depends on which policy was in force on the date of the injury (not necessarily the date of treatment). 

In states with exclusive state funds or provincial workers’ compensation insurers, the responsible insurer is obvious and clear.  There is one insurer, on question of policy year or other coverage.  In jurisdictions where the definition of worker is independent of the coverage status of the employer, the worker’s medical expenses will be paid by the insurer even if the employer was uninsured by omission or fraud. 

The situation may be more complicated in states with employer “deductibles” or other arrangements.  In the Australian state of Victoria, workers’ compensation is payable but only after the employer has paid for the first 10 days of incapacity and $707 (2019, indexed yearly) of medical costs.  [see,  http://www1.worksafe.vic.gov.au/vwa/claimsmanual/Content/4EmployerObligations/2%204%201%20Employers%20liability.htm ].

Many people assume Canada, with universal health insurance has one big insurance plan that covers everyone for every health concern.  That assumption contains many misconceptions.  Saskatchewan adopted the first “Medicare” plan under then premier, “Tommy” Douglas [trivia: Kiefer Sutherland’s grandfather].  A federal Royal Commission on Health Services in Canada under Chief Justice, Emmett Hall, recommended Canada adopted universal healthcare in the mid-1960s but universal does not mean one plan.  Every province and territory has its own medical services plan—a single payer healthcare insurance plan in each province or territory that covers all necessary insured health services… except workers’ compensation for reasons noted below. 

Why are workers’ compensation medical payments not covered by provincial health plans?

Canada does not have a national, single payer healthcare plan; each province and territory does. The background is rooted in the constitution and in the evolution of healthcare in Canada.  The distribution of powers in the Canada Constitution Act, 1867, defines a relatively narrow set of federal powers [section 91] and gives provinces law-making power [sections 92, 92A) over a broader area including “hospitals”, “property and civil rights”,  and “generally all matters of merely local or private nature in the province”, (making workers’ compensation, occupational safety and health regulation healthcare, labour law clearly in the provincial domain for matters not specifically in the federal powers). [Note:  The Canadian federal government has occupational safety and health authority over industry under its constitutional powers (including inter-provincial transportation, communications, for example, as well as for its own government services and military).  It also has authority over workers' compensation for its own employees but contracts with the provincial workers' compensation boards to administer that for them.  See the Government Employees Compensation Act].  

The federal government has greater taxation authority under the constitution so it had the wherewithal to partially fund provincial healthcare system, if the provinces agreed to develop them in a specific way.  Under what is now known as the Canada Health Act, [CHA] the provinces agreed to set up health plans that adhered to five principles:

(a) public administration [single payer, not for profit, public authority];
(b) comprehensiveness [all necessary health services];
(c) universality [everyone entitled to same level of care];
(d) portability [coverage maintained even if in another province]; and
(e) accessibility [reasonable access to care and reasonable compensation to hospitals and providers for services].

Both the original Saskatchewan medical insurance plan and Hall’s recommended legislation excluded workers’ compensation.  The current CHA defines “insured health service” as follows:

insured health services means hospital services, physician services and surgical-dental services provided to insured persons, but does not include any health services that a person is entitled to and eligible for under any other Act of Parliament or under any Act of the legislature of a province that relates to workers’ or workmen’s compensation; [emphasis and underlining added]

The reasoning was clear:  workers were already insured for work injuries. 

Under the CHA and provincial medical services plans, extra billing or co-pays are not permitted.  In most cases, how a medical service is paid and who pays it happens behind the scenes with the physician or hospital billing the single payer provincial health insurance plan or workers’ compensation (sometimes using the same electronic platform).  Note that there is strong similarity of what is covered and not covered among provincial medial services plans but there are differences.

Healthcare spending is only one dimension

Work-related injury, illness and disease has a huge cost.  The human cost is not reflected in the healthcare or compensation dollars expended by workers’ compensation systems. 

It is clear that healthcare expenditures are a significant, growing and even predominant portion of workers’ compensation costs.  At the same time, workers’ compensation costs barely register as a percentage or two of overall national healthcare spending in many developed countries, regardless of the healthcare funding model. 

This analysis does not include any discussion of offsets.  Healthcare is a big segment of the economy; it employs many people and generates significant economic activity.  Workers’ compensation administration, rehabilitation and many other services are part of that activity.  The human and financial cost of work-related injury can never be justified by these activities but few studies quantify them. 

Does universal, single-payer system have advantages for workers’ compensation?

The universal, single-payer healthcare systems in each Canadian jurisdiction offer certain economies of scale and scope to the populace.  With everyone covered, there is no incentive to make claims against workers’ compensation just to ensure an injury gets necessary medical care.  A population where everyone has health insurance (universal coverage) is generally healthier than a population without such coverage, based on many health metrics (like life expectancy, infant mortality, etc.); a healthier population means fewer work-related injuries that do occur will carry with them to their workers’ compensation claims additional healthcare burdens.   Single-payer systems (that include or exclude workers’ compensation) offer efficiencies that lower costs, reduce duplication and provide rich data for further improvements to population health.   As in the Canadian example, the single-payer system can exclude workers’ compensation yet provide lower cost, higher efficiency benefits to the workers’ compensation system (shared systems, data, fee schedules, etc.).

Work-related injuries are almost entirely preventable.  In the ideal world, their contribution to national healthcare spending should be vanishingly small.  The world, however, is far from ideal.  Regardless of workers’ compensation model, the cost of work-related injury, illness and disease is significant.  Under private, competitive models and exclusive state fund or provincial workers’ compensation, those significant healthcare expenditures are a big part of the premium cost and provide further incentive toward prevention.  Keeping the healthcare cost of work-related injury, illness and disease a direct part of the workers’ compensation premium may hasten progress toward that ideal world.

Are Workers’ Compensation benefits protected against the rising cost of living?

The cost of living (almost always) goes up year over year.  If your earnings aren’t keeping pace, something has to give.  In the short run, you might be able to carpool more, eat out less, switch to generic products, and maybe repair your old car rather than buy that new SUV.  In the long run, if the cost of living continues to outpace your earnings, you might have to take on a second (or third) job, downsize your home, move to a lower cost city, or go back to school so you can pursue a career with higher wages.   Those are some of the options… if you are able to earn an income.   

Workers with permanent disabilities often don’t have those options.  The monthly workers’ compensation amount they receive may have sustained them initially but unless it is adjusted for the cost of living, permanently disabled workers will see the buying power of their workers’ compensation income decline with each passing year.  Over time, savings may be depleted, debts incurred, and their health and welfare diminished—furthering the burden of their original work-related injuries. 

To forestall this eventuality, the majority of North American workers’ compensation jurisdictions adjust periodic payments (sometimes called workers’ compensation pensions or permanent disability payments) to account for increases in the cost of living.  This policy, however, is far from universal among US workers’ compensation systems.  A recent WCRI/IAIABC survey of Workers’ Compensation Laws (2016) recorded 27 US states with no cost-of-living escalator for permanent total disability cases. 

Consumer Price Index:  A common reference with many versions and unique calculation characteristics

The most common approach used by North American workers’ compensation jurisdictions that do adjust their payments for increases in the cost of living is to mirror the increases to federal entitlement plans such as US Social Security (USSS) and Canada Pensions Plan (CPP and parallel Quebec Pension Plan, QPP).   These near universal social insurance plans for retirement and disability benefits for working citizens provide a convenient standard for workers’ compensation policy makers designing cost-of-living adjustments (often abbreviated COLA).   

Both USSS and CPP increase benefits annually based on changes the Consumer Price Index (CPI) for their respective countries.  The method of calculation and exactly which components of the CPI are used differ.  There are technical manuals on CPI calculations; however, for workers’ compensation policy makers there are a couple of general comments that may provide insight into the use of CPI as an adjustment factor.

First, CPI is not one universal thing.  The standard definition of CPI refers to the change over time in the cost of a selected (but arbitrary) “basket of goods” in a base year.  The simple concept is more complicated than it sounds; a lot of detail goes into selecting and weighting items for that “basket of goods”.  Most versions include goods and services such as transportation, education, recreation, communications, and medical care.  Other real expenses that are excluded from the “basket of goods” [in Canada, at last] are real estate and life insurance.  And exactly whose basket we are considering can make a big difference.  What a young, urban couple with two kids in school has in their typical basket probably differs from a rural farm family or retired manager might consider typical.   The definition of what is in that basket and what proportion or weight goes to each category are also subject to change over time.  Think about communications, for example; with internet services and mobile data becoming essential utilities, it makes sense that they be included and their weight increased. 

The second point to remember about CPI is that there are often multiple versions of the CPI even within one country.  Variations include geographic subsets [regions, states, provinces, cities], versions that include all or just core items, and even versions that designed to reflect cost of living impacts on specific populations. In the US, for example, two main indexes are often cited for specific populations:  urban consumers [CPI-U] and urban clerical and wage earners [CPI-W].  Note that CPI-U covers most people including the unemployed and retired whereas the CPI-W is intended to reflect the impact of price changes on those working at least 37 weeks per year.  Both exclude rural consumers.    There are other CPI series including CPI-E for elderly.  Each has its uses and merits (as well as limitations and drawbacks).  Each series will produce different results.

Finally, the monthly CPI change is measured against a base year.  The base year for some CPI series or specific line items may differ from others or be changed over the time series in question.  Specifying which CPI, components, geographic location, and base year may be important to interpreting what a particular CPI value means.  Policy makers should be aware of the potential for such changes when designing a COLA based on CPI.

The following table provides how the selected CPI can yield different results based on geography:

Note that increases in each CPI series vary.  Over time, the differential can become significant.

The following US BLS table illustrates CPI All Items data, not seasonally adjusted

The base period is 1982-1984 so each table entry indicates a value relative to that base. Note that the values monthly almost always increase.  Whether comparing month over month values in a  row,  year over year values for months (quarters, half years or other ranges) in a column, you are likely to find examples where the CPI value declines. 

Using the CPI to adjust social insurance payments may be a common approach but it is far from perfect.  It applies defensible average weights to a range of items to derive an adjustment that may fall short of actual individual need or experience.  The converse may also be true; individuals may actually use a different basket of goods and experience less of an impact than the CPI would suggest. 

The selection of a particular CPI series should be intentional and explicitly justified.  For example, if the data show that virtually all recipients of benefits reside in a particular region, then that may become the policy justification for selecting a regional CPI.

Despite the imperfections and caveats, changes in CPI provide a strong indicator of cost pressure experienced by consumers in the real economy.  From a public policy perspective, CPI provides a useful reference against which to assess how well a social insurance program like workers’ compensation addresses the reality of the (almost always) increasing costs of living. 

CPI and Social Insurance:  How US Social Security and Canada Pension use CPI

For US Social security, the following summarizes how CPI increases are applied to the amount sent to beneficiaries:

Social Security COLAs are based on changes in the Consumer Price Index for Urban Wage Earners and Clerical Workers (CPI-W), updated monthly by the Department of Labor’s Bureau of Labor Statistics (BLS). The COLA equals the growth, if any, in the index from the highest third calendar quarter [July, August, September] average CPI-W recorded (most often, from the previous year) to the average CPI-W for the third calendar quarter of the current year. The COLA becomes effective in December of the current year and is payable in January of the following year. (Social Security payments always reflect the benefits due for the preceding month.)

If there is no percentage increase in the CPI-W between the measuring periods, no COLA is payable. No COLA was payable in January 2010, January 2011, or in January 2016.

- Social Security: Cost-of-Living Adjustments, Congressional Research Service, Updated November 19, 2018.

Note the actual application of USSS COLA evaluates changes in the CPI-W for the most recent third quarter and the previous highest third quarter.

Canada Pension Plan uses CPI data supplied by Statistics Canada to adjust CPP amounts once a year in January; rather than a single quarter of data, the calculation takes into account all monthly increases from the past year compared to the increases from the equivalent period the year prior.   The following chart shows the calculation method (see Government of Canada,  Canada Pension Plan Amounts and the Consumer Price Index):

The CPP increase is the percentage change from one 12-month period (November to October) to the previous 12-month period (November to October).  To calculate the 2019 CPP rates increase, a formula based on the average national CPI for all items for November 2017 to October 2018 is divided by the average CPI for November 2016 to October 2017 yields a 2.68 percent increase effective January 2019. 

Beyond the importance of understanding USSS and CPP use of CPI to develop their respective COLA for their plans, any statute that relies on the USSS or CPP COLA automatically rely on their respective calculation methods.  Workers’ compensation policy makers, for example, need to examine if the underlying assumptions and calculations of these social insurance plans make sense for the population of disabled workers receiving monthly compensation payments.

Workers’ Compensation jurisdictions:  Examples of policy implementations of COLA

Where workers’ compensation cases receive COLA adjustments, policy makers have had to make choices about how payment values should be adjusted.  The following are examples of how some jurisdictions have implemented their COLAs.

The Rhode Island workers’ compensation statute embodies the key elements in a COLA statute reliant on CPI by defining which CPI will be used, what time period will be referenced, the category of cases eligible and when the benefit will be applied:

RI Gen L § 28-33-17 (2017)  (f)(1) Where any employee's incapacity is total and has extended beyond fifty-two (52) weeks, regardless of the date of injury, payments made to all totally incapacitated employees shall be increased as of May 10, 1991, and annually on the tenth of May after that as long as the employee remains totally incapacitated. The increase shall be by an amount equal to the total percentage increase in annual Consumer Price Index, United States City Average for Urban Wage Earners and Clerical Workers, as formulated and computed by the Bureau of Labor Statistics of the United States Department of Labor for the period of March 1 to February 28 each year.

Virginia’s COLA escalator is based on the all items CPI each October 1 but because of the way Social Security disability payments are handled, the impact may be lower than the full CPI. In Virginia, the combined weekly compensation rate and weekly Social Security disability benefit cannot exceed 80% of claimant’s established pre-injury average weekly wage. Consequently, the application of the COLA is not automatic; it is subject to individual application and decision-making annually. 

Ontario’s workers’ compensation system, WSIB, recently improved its cost of living formula to fully reflect the Canadian CPI increase.   In January 2019, people receiving WSIB benefits will receive a cost-of-living adjustment of 2.3 per cent beginning on January 1.  The COLA escalator is automatic but wage loss benefits are offset by disability benefits a worker receives from CPP and QPP at a rate of 50%.  This partial integration or offset is common in workers’ compensation and disability insurance, although the degree of integration (and which benefit is reduced) varies.

British Columbia workers’ compensation pension recipients see indexation of their disability awards based on the following formula outlined in the Workers Compensation Act:

25   (1)For the purposes of this section, the Board must, as of January 1 of each year,

(a)determine the percentage change in the consumer price index for Canada, for all items, for the 12 month period ending on October 31 of the previous year, as published by Statistics Canada, and

(b)subtract 1% from the percentage change determined under paragraph (a).

(2) The percentage resulting from calculations made under subsection (1) must not be greater than 4% or less than 0%.

In practical terms, CPI change of 2.444614% less 1% results in a 1.444614% increase for most compensation recipients effective January 1, 2019. 

From a policy planning perspective, it is important to consider the longer-term impact of reduction policies.  For a 40 year old injured worker with a permanent total disability, the cumulative impact of the “less 1%” obviously increases over time.  Assuming just 2% CPI each year, the 1% reduction means that the purchasing power of each $100 awarded in 2002 will have grown after 10 years to $110.46 far short of the $121.90 necessary to meet the full increase in cost of living.  The difference increases with every passing year.  After 20 applications in this example, the adjusted value will have increased to $122.02 while full CPI would have increased that original $100 to $148.59. 

The geographic qualification to the definition of the CPI is fairly common.  Alberta’s WCB uses the change in the Alberta Consumer Price Index (ACPI) for 12 months, ending September 30   and applies the result in January of the following year.  Until a recent policy change to full ACPI, the ACPI amount was reduced by 0.5%.

While the trend is towards applying a full CPI increase to adjust for the cost-of-living adjustment, limits are often set in legislation.  The Yukon Territory uses the percentage change in the CPI for the geographic location of its capital, Whitehorse, calculated by comparing the 12-month period ending October 31st of the previous year with one year earlier, capped at a maximum of 4% and a minimum of 0%.

Which CPI to use is often an important determinant of the actual cost-of-living increase in other jurisdictions as well.  In Massachusetts, the increase is based on the CPI increase for the Northeast urban region.  In Saskatchewan, it is the percentage change in CPI for Regina and Saskatoon for the 12 months ending on November 30 of the previous year that determines the increase to be applied.  

Prince Edward Island combines a restricted formula that yields less than full CPI, a CPI geographic reference and a cap such that  extended wage loss benefits will be adjusted on July 1 each year by an  amount equal to the lesser of: 80% of the percentage change in the CPI [less than full CPI restriction] for Charlottetown and Summerside for all items [geographic reference] for December of the previous year and December of one year earlier and 4% [cap]. (see WCA s. 49.1(1.1)) 

One challenge with geographic or regional CPI considerations is that it may well under-reflect costs associated with individuals who relocate.  One can imagine a totally disabled worker in a rural setting wanting to relocate to a more urban centre where medical and support services are more appropriate and available to his or her need.  The relevance of a COLA based on a CPI indicator from where the injury occurred may be lost. 

Florida has a unique cost-of-living adjustment method.  According to the Social Security Administration:

Florida Workers’ Compensation does not provide for a traditional cost of living increase. However, individuals that are permanently and totally disabled are potentially eligible for a supplemental yearly increase of 3 percent. The increase is only payable for individuals under age 62 that are not subject to offset due to receipt of Social Security benefits. When the individual attains age 62, if they are eligible for Social Security benefits they lose entitlement to the supplemental benefits if the date of injury is on or after July 1, 1990. For injuries prior to July 1, 1990, the supplemental payments continue.

Not all jurisdictions use a version of CPI to adjust compensation payments.   Washington State’s Department of Labor and Industries uses a different method to calculate the cost of living increase.  In that state, most workers injured on or before July 1, 2017 will see time-loss and pension benefit payments increase by 5 percent based on the change in the state's average wage.  That increase was effective July 1, 2018.

Some jurisdictions apply the indexation to very restricted categories.  For example, in Connecticut, only permanently and totally disabled workers or those who have been totally disabled for a period of 5 years or more are eligible for a cost of living adjustment to their compensation. 

Reference frames, Application dates, Caps and floors

You may have noticed the reference range for calculating and applying a COLA varies.  BC and the Yukon use the year ending October 31, Alberta uses September 30, and PEI uses December 31.  The date of application also varies by jurisdiction:  July 1 in Washington state, and May 10 for Rhode Island.  While a period of time between the COLA reference period and its application is reasonable, the rationale for a lengthy delay should be explained.  When calculations were done by hand, a lengthy lag time was justifiable.  If systems are designed with COLA in mind, their routine application should allow a short period between the reference period for calculation and actual application of the COLA.

Most of the policies outlined in this paper use full year CPI data as the basis for their COLA.  This tends to smooth out seasonal variations.  Transportation costs tend to peak in the summer, fresh vegetable costs are lower in the harvest season.  Some CPI series are smoothed or seasonally adjusted; use of seasonally adjusted data should be justified and specified in a policy relying on such data.

Regardless of the CPI or other standard measure used to adjust workers’ compensation payments, policy makers may include limitations on the extent to which the indexation may be applied.  As noted above, several provinces including BC and Alberta cap the possible increase to a maximum 4%; many policies contain a floor of zero percent to prevent a negative percentage being applied should the COLA formula generate such a result.  Although rare, zero results have occurred. 

Accounting for the cost of living

Workers’ compensation legislators and policy makers have long acknowledged that “protection against the value-eroding power of inflation is necessary” [Burton, John F. Jr. [Chairman], Report of the National Commission on State Workmen’s Compensation Laws, US Government July 1972 chapter 3 page 71] for at least some categories of recipients.  In protracted recoveries, permanent disabilities, and compensation for survivors and dependents, that erosion can be substantial. Consider a disabled worker injured in 2002 and permanently disabled; using the “All Items CPI”, that worker will have seen an increase of more than 40% in costs of goods and services in the US or about 32% in Canada.   In most jurisdictions in North America, this worker will have received some cost-of-living adjustments.  In many jurisdictions, however, workers’ compensation payments will not have kept pace with the full increase in the cost of living.

To the best of my knowledge, there is no detailed study of the cost-of-living-adjustment mechanisms in workers’ compensation.  Aside from the WCRI/IAIABC survey, there are no studies that reflect current or at least recent policies in a comparative way.  Few jurisdictions post historical tables of past COLA increases in a convenient way (although WorkSafeBC and Rhode Island data tables were readily available on line).  

The indexation of workers’ compensation payments particularly for permanent total disability cases adds a significant value to the incurred cost of an injury.  That cost is reflected in premium values.  Workers’ compensation analysis that fail to account for compensation parameters such as COLA provisions may mislead readers. Many existing comparative studies of workers’ compensation premiums and claim costs exclude detailed information and cost implications of “system features” such as compensation rate, maximum insurable earnings, and COLA provisions from their analysis. Policy makers need to understand the cost implications of system features in interpreting comparative results and designing improvements to (or the addition of) their cost-of-living provisions.

Failure to include protection against the rising cost of living under-value the human and financial loss of work-related injury, diminish the value and adequacy of compensation as the years go by, and often externalizing costs to family, community and taxpayers through additional welfare and health costs.  

Work-related injury and death have real human and financial costs.  Workers and their families bear their share of both.  Permanent disability, survivor and dependent compensation payments offset some of the financial costs. To be clear, adding or improving inflation protection or cost-of-living adjustments may increase premiums—costs to employers; failing to do so, however, is an intentional policy choice to place an increasing share of the cost to workers, families and taxpayers.  That policy choice should be acknowledged, explicitly stated and justified… or abandoned.

Is Workers’ Compensation primarily a “medical insurance” plan?

Nearly three quarters of US  workers’ compensation jurisdictions now incur more  healthcare cost than “indemnity” costs (wage replacement for temporary disability workers’ compensation claims) for the most common workplace injury claims.

For many years, NCCI has published charts like this one that was presented at the 2016 Annual Issues Symposium (Kathy Antonello, State of the Line 2016):

The total “ benefits pie” in this case is made up of specific incurred costs of claims from the injury year taken to “full development” (essentially the full lifetime of the claim).    The indemnity side of the pie represents the cost of wage-loss or replacement compensation incurred for temporary claims.  It also apparently includes certain vocational rehabilitation costs such as retraining allowances and tuition. The medical slice of the pie includes hospital, physician, physiotherapy, medications, appliances, diagnostics (including x-rays, MRIs, CT Scans) and other healthcare expenses.  Excluded from this pie are permanent disability, administrations and other costs (such as underwriting expenses, advertising, etc.) not directly associated with the main benefit costs associated with claims.The NCCI chart reflects data from NCCI states and State Fund states. 

NASI publishes a similar analysis covering all US states and the District of Columbia using NCCI data and other data from non-NCCI states (see Workers’ Compensation: Benefits, Coverage and Costs, 2013  Figure 3).  The result is similar:

The NASI study contains individual state-level data [see Table 8] .  For the following chart I have extracted the percent of the combined medical and indemnity pie and ordered the data by percentage medical.

This depiction is useful in showing the range.  Note the median value using NASI’s 2013 injury year data is 54.7% medical.  This is statistically very close to NCCI’s estimate of 58%.  Based on this ranking, three-quarters of US workers’ compensation jurisdiction pay more than 50% of the benefit pie on medical expenses. 

What accounts for the variation in the share of medical costs?  Medical costs do vary from state to state.  Some jurisdictions have medical fee schedules;  others have requirements regarding the use of certain medical networks.  The proportion of more serious injuries may also contribute to greater expense on the medical side. A jurisdiction with large, high-risk industrial sectors such as primary resource extraction (logging, mining) is likely to have more serious injuries that involve greater medical costs than a jurisdiction dominated by low-risk industries such as financial institutions or tourism.

A more significant factor, however, relates to the non-medical side of the pie that covers the level of compensation.  The so called "indemnity" or "cash benefit" side of the pie is determined by the percentage of wage replacement provided, maximum insurable earnings covered (or maximum weekly benefit payable),  the duration of a waiting period,  the length (or absence of) a retroactive period, and the overall duration of temporary disability claims (sometimes capped).  States with low compensation rates and long waiting periods (and correspondingly long or absent retroactive periods) are likely to have higher percentages of the pie going to medical.  

To make this point clearer, I used the current “Maximum Weekly Workers’ Compensation Amounts” [as reported by the SSA, Program Operations Manual System (POMS), DI 52150.045 Chart of States’ Maximum Workers’ Compensation (WC) Benefits]  as a handy proxy  indicator for the overall “comprehensiveness” of the wage-replacement compensation side of the equation.  Seven of the ten states with the highest weekly maximums ($1211 to $1628: Washington, Massachusetts, District of Columbia, Illinois, Connecticut, Vermont, Iowa) have lower-than-median percentage shares on the medical side; eight of the ten states with the lowest weekly maximums ($468.63 – $778.83: Kansas, Delaware, Mississippi, Montana, Idaho, Arkansas, Arizona, South Dakota) have medical percentage shares above the median.  Of course, this is a very rough indicator; most injured workers are unlikely to be paid at the maximum rate and this indicator fails to account for financial losses the worker must bear  in terms of uncompensated waiting periods and low compensation rates that fail to approximate usual spendable income.    

The most common claims in workers’ compensation are for healthcare expenses and temporary wage replacement.  The medical-indemnity split analysis underscores the magnitude of healthcare costs and also provides evidence regarding the wage replacement costs.  A very high percentage of medical costs may indicate  temporary disability compensation rates are exceptionally low.

Workers’ compensation systems may not have started out to be medical insurance systems but medical costs now dominate the claim expense for temporary disability in many states.  This warrants close attention to both the medical cost drivers and the levels of compensation.  

Workers will always bear all the physical, psychological and social consequences of workplace injury.  Workers’ compensation should minimize any externalization of medical or financial losses to workers and their families or the communities (including the community of tax payers) in which they live.  Severing medical costs from workers’ compensation effectively removes half the premium incentive for greater investment in workplace health and safety.  Transferring medical costs to other payments would likely amount to a subsidy to industry overall and increased costs for healthcare funders.  

The financial costs of workplace injuries, illnesses and deaths should be covered by the industries that give rise to them.  That was the basis of the "grand bargain", the "historic compromise" and should continue as the foundation for workers' compensation.