6.2.1 The History

To understand why the public option emerged, and why it has struggled to gain traction, it is helpful to recount the development of health insurance in the United States. This well-worn story is worth revisiting because it is still salient.

The United States is unique among OECD nations when it comes to paying for health care, and not in a good way. Most countries’ systems for health-care financing grew up in the early- to mid-twentieth century as medical care became more advanced and more expensive than most people could afford on their own.Footnote ⁶ In Europe, what emerged were public systems of health-care finance in two forms, often characterized broadly as Beveridge and Bismarkian systems.Footnote ⁷ The Beveridge approach was direct provision of health care by the government, as in England, where the government owns hospitals and employs medical professionals – aka “socialized medicine.” Bismarkian systems, or social insurance, are ones where the government finances health care but the providers can be public or private. Over time, countries in Europe and beyond, such as Canada, developed variations on these themes. At the core, however, these systems embraced the basic idea that the government would take a central role in ensuring access to affordable health care for the entire population.

The United States charted a wholly different path. The beginning upsurge of health insurance in the United States was initiated by industry itself. As medical care became both more effective and expensive, people increasingly sought it. Hospitals feared not being paid for their work if they relied on patients to pay cash for services, nor did they want to have to confirm the financial solvency of every patient prior to caring for them.Footnote ⁸ Hospitals thus created prepaid health-care funds, beginning with Baylor University Hospital in the 1920s, that guaranteed people access to medical care up to a certain level, with prepayment.Footnote ⁹ These plans spread and eventually took on the name Blue Cross. Within a short period, Blue Shield followed, offering a similar structure for monthly prepayment of fees to groups of physicians in turn for guaranteed access to outpatient care. Unlike the health insurance of today, the Blues embraced some of the solidaristic characteristics that define systems elsewhere in the world, like charging all members of a community the same rate for membership regardless of their personal characteristics or health status.

A second – and the most defining – major development in the United States was the rise of employer-sponsored health insurance (ESI) during and after World War II (WWII). Increased reliance on employers as a source of health coverage in the United States coincided with the moment that many other countries were doubling down on the government’s role. In England, for example, during WWII the government built health infrastructure to deal with an unmet need for medical services and this infrastructure served as the beginning of the National Health Service, established at the end of the war.Footnote ¹⁰

In the United States, in contrast, in the years during and especially after WWII, ESI surged, bolstered by several public policies. A commonly told story is that the trend began with wage freezes during the war prompting employers to compensate with benefits instead of cash wages, but the growth in these plans was relatively small in this period.Footnote ¹¹ More consistent with the timing of a major upsurge in adoption of ESI were a 1945 federal rule that required employers to leave wartime health benefits in place, a 1949 federal rule allowing unions to bargain for health benefits, and most importantly a 1954 rule by the Internal Revenue Service excluding dollars spent on health benefits by employers and employees from taxes.Footnote ¹² This meant, by one estimate, that a dollar of compensation in cash wages only cost employers an average of $.66 if spent on health benefits. With all of these factors, ESI and the centrality of private insurance took hold. Today, half of all Americans have insurance through an employer.Footnote ¹³

Public financing, however, maintained a key role. It began in small scale in the Veterans Health Administration and Indian Health Services. In 1965, after decades of attempts at universal, public health coverage, Medicare and Medicaid were signed into law by President Johnson. These programs established public programs to pay for medical care for populations seen as vulnerable and also least likely to have access to ESI – the elderly (Medicare) and poor children, pregnant women, and, later, people with disabilities (Medicaid).

People who didn’t qualify for a form of public coverage and who didn’t have private coverage available through an employer had limited options.Footnote ¹⁴ Some people could access charity care. Others paid out-of-pocket for health care, although doing so for anything other than the simplest care was out of the reach of most people. A final pathway was to buy health insurance directly from an insurer in the individual, or nongroup, insurance market, and doing so was thorny. In the individual market, insurers underwrote applicants and deemed many people with prior health problems as risky. Six to seven percent of the non-elderly population (about 15 million people) had individual-market coverage prior to the passage of the ACA.Footnote ¹⁵ Historically, this coverage was relatively more expensive, in part because administrative costs were as high as 15–20 percent of total costs. Individuals deemed risky were declined coverage.Footnote ¹⁶ As many as three in five people who applied for policies before the ACA could not afford the high premium prices or were denied coverage, and many people remained uninsured or underinsured.Footnote ¹⁷

6.2.2 The ACA and the Individual Insurance Market

In the year before the ACA’s passage, 16.3 percent of Americans were uninsured, including 18.4 percent of people under age 65, and many more were underinsured.Footnote ¹⁸ The goal of the ACA was to fill in the gaps between public coverage and ESI through two primary mechanisms: expansion of the Medicaid program to poorer Americans and regulation of the individual market to make it a source of affordable and meaningful coverage for everyone else without ESI or public coverage.

Here is where the public option came into play.

The ACA’s strategy relied heavily on getting more people enrolled in private health insurance, for reasons both political and pragmatic.Footnote ¹⁹ Yet, the individual market was inhospitable to the goals of universal access and affordability, and early architects of the Obama reform had two responses to discipline the private insurance companies in this market: (1) regulate them and (2) create competition through a public option.

The first required a federalization of heath regulation and a complicated, multipronged approach. Prior to the ACA, most health insurance regulation, especially of the individual market, occurred at the state level, and there was little of it.Footnote ²⁰ The ACA created federal rules for individual insurance, drawing lessons from the 2006 health reform in Massachusetts. Insurers were required to issue insurance to any applicant (“guaranteed issue”).Footnote ²¹ Medical underwriting was prohibited, and insurers could not consider preexisting conditions in determining eligibility or price.Footnote ²² In fact, premiums for a policy were allowed to vary based on only four factors: age, geography, family size, and tobacco use status.Footnote ²³ Even if an insurer devised a way to cherry-pick out healthier applicants, the law intended to disgorge any resulting profits through reinsurance and risk-adjustment arrangements.Footnote ²⁴ And the ACA also regulated benefits, requiring that all plans cover preventative care without cost sharing and a set of essential health benefits for individual-market plans, and prohibited limits on these benefits for most plans.Footnote ²⁵

So that these regulations did not exacerbate adverse selection, or the tendency of healthier people to wait to buy coverage until they need it, the law included an individual mandate that required that most Americans carry health insurance that offers “minimum essential coverage,” or else pay a penalty.Footnote ²⁶ As a carrot, the ACA provided for financial support to help lower-income individuals. Anyone who earns from 100 to 400 percent of the federal poverty level ($12,490 to $49,960 for a single person in 2020) and does not have another source of adequate insurance, such as through an employer or Medicaid, is eligible for subsidized premiums and in many cases also cost-sharing reductions to help pay for their out-of-pocket share of costs.Footnote ²⁷

6.2.3 The ACA’s (Foregone) Public Option

In case the nearly 200 pages of the ACA devoted to the endeavor of regulating private insurance to achieve broader policy goals fell short, there was a second strategy: create competition through a public option.

The idea was to develop a public health plan, based either loosely or very closely on Medicare that would compete with private health plans in the exchanges – new marketplaces where people would go to compare and buy health insurance policies. The public option was described by its proponents as simply one option among many, a public health insurance plan that would compete side-by-side with private plans and would win if the private options were not good enough. Presumably, if the public option offered a similar or better product for lower prices, people would choose it. As Jacob Hacker suggested: “public plan choice gives Americans the opportunity to choose for themselves how they value the strengths and weaknesses of a public, Medicare-like plan and competing private health plans.”Footnote ²⁸

But this idea faced staunch resistance. Even advocates of market-based policy fought back, both on the details and on the concept as a whole. The public option was cast aside. Yet imagine for a moment that Congress had included a public option as part of the ACA. Would the world look different than it does today, and how?

6.3.1 The Two Pathways for a Public Option

The first pathway would have led to everyone enrolling in the public option – what Jacob Hacker describes in his chapter as a back door to universal Medicare (Pathway 1). In Pathway 1, the public option would be based on Medicare. It would borrow Medicare’s existing provider network and negotiated rates. This is the version its architects envisioned. Since the plans sold on the ACA’s exchanges are standardized by regulation – they vary little on benefits and cost sharing structure – the way plans distinguish themselves is based on network, pricing, or perhaps name recognition or brand.

If the public option were built on Medicare’s platform, it would easily have bettered private options on these dimensions in most geographies. Medicare has an extensive network of providers, who accept lower rates from Medicare than they do from private insurance, which pays rates to hospitals that are sometimes 50 percent higher than what Medicare pays.Footnote ²⁹ Medicare also controls spending growth better than private plans.Footnote ³⁰ The divergence between what Medicare and private insurers pay is increasing as consolidation among hospitals and other providers has enabled providers to demand significant rate increases from private plans.Footnote ³¹ Plus, Medicare’s administrative rates are lower than those for private insurance, even considering that private insurance administrative rates for the individual market are limited under the ACA. All of this means that even if a public option plan paid providers a cut above Medicare’s standard negotiated rates, it could offer similar benefits at much lower premium prices than private plans with a more comprehensive network of providers. It would simply be the better option.

Assuming people understood that the public option was clearly better – an assumption that should be taken with a grain of salt for reasons described below – everyone would have chosen the cheaper, better public option. Even if people chose only based on premium price, which evidence suggests is exactly what often happens on the ACA exchanges,Footnote ³² the public option would have won out. Over time, it would have displaced the private plans altogether and become the de facto coverage for anyone seeking a health plan on the ACA exchanges.

Pathway 1 would thus deliver on one half of the competitive public option vision. It would have provided guaranteed access at controlled prices, but eventually would not have coexisted with private plans at all, at least not side-by-side in a competitive structure.

Interest groups who opposed the public option – including insurers and medical providers and suppliers (pharmaceutical and medical device companies) whose reimbursement would be lower under Pathway 1 – saw the writing on the wall. During the ACA debates, they answered the mention of a public insurance plan option with cries of socialism.Footnote ³³ They also demanded that if a public option were passed it had to compete without relying on the preexisting advantages of Medicare over private insurance.

The second possible way the public option could have, and likely would have, played out in the current political environment would have been that these interest groups’ demands won out (Pathway 2). Imagine that a public option had to compete on “even ground,” or “break even financially” (the latter of which even President Obama asserted)Footnote ³⁴. Although it’s difficult to know exactly what that would have meant in practice, at the very least it would have prohibited the public option from coasting on the preexisting Medicare network and negotiated prices. If the public option did not have a clear pricing or network advantage – if it were not a clear winner – what would that have meant?

6.3.2 The Problem of Choice

In some of the examples that Sitaraman and Alstott profile in The Public Option, competitive public options are effective because they offer a genuinely meaningful alternative. Sometimes this alternative is one that is less expensive, or even free, as compared to pricier and probably fancier and more exclusive private options. This is the case with public pools – the inspiration for the cover of the print version of the book. Sometimes the alternative is part of a tiered system. With the USPS, they suggest, one can usually get reliable standard mail services, but for higher-end packing and shipping needs, Fed-Ex and UPS are the better go-to options.

The problem with health care’s competitive public option is that in its more politically realistic form (Pathway 2), it would not have been easily differentiated in this way, and end users would have struggled to see its relative benefit. The public option would have offered nearly the same benefits as the private options, likely at a similar price. It might have been marginally better under close examination – maybe a better network of providers, maybe some extra benefits because of the ability of these plans to operate more slimly, maybe less administrative hassle for enrollees (or perhaps not), or maybe it would have provided an easier transition to Medicare once eligible. But none of these attributes would have been easy to detect, nor would they likely have driven someone’s health plan selection.

The main problem, thus, with any politically realistic version of a competitive public option in health care is the problem of choice. Consumers are notoriously bad at deciphering differences among health plans and choosing wisely among them. This should be unsurprising when considering the nature of health plan choice. Fundamentally, buying health insurance demands having preferences about things that most people have never experienced before, like hospitalization or cancer care. Then they have to weigh this risk against spending on other goods and services.

Then someone must understand how such preferences translate into health insurance policy terms. Most people do not understand the basic features of health insurance plans that should shape their decisions – such as how much a plan costs and what benefits are covered.Footnote ³⁵ In a survey of insured adults, only 14 percent correctly answered four simple multiple-choice questions about cost-sharing features, such as a deductible or copayment.Footnote ³⁶ However, people overestimated their understanding of these insurance concepts, which suggests many would not seek help or education even if available.

Furthermore, choosing a health plan requires making calculations regarding deductibles, cost-sharing, and premiums that exceed many Americans’ literacy and numeracy skills.Footnote ³⁷ Even college-educated people show surprisingly high levels of error on simple arithmetic tests.Footnote ³⁸

Even putting aside these challenges, choosing health insurance has all of the telltale characteristics that impair rational decision-making, sometimes referred to as generating cognitive biases. People are overly optimistic about their own health,Footnote ³⁹ which could prompt them to underinvest in health insurance. People also struggle to factor risk into decision-making – an element central, of course, to health insurance choices.Footnote ⁴⁰ This is why young, healthy people often forgo buying health insurance, even if it’s cheap.

Documenting the end result of all of these barriers, a volume of empirical work illuminates the many ways and reasons that we – regardless of education, income, or smarts – make poor choices among health plans. As described later, these poor choices persist when there is a clear, superior option. These poor choices persist even when options are simplified. And they persist even in the face of efforts to help people make better choices through decision-making supports, or nudges. As one set of researchers behind several studies that looked at how to help people make better decisions by simplifying health plan options or helping consumers through options concluded: “[T]he main barrier to financially efficient choice was not the number of options confronting employees, nor the transparency of their presentation, but rather the … lack of basic understanding of health insurance.”Footnote ⁴¹

A few select examples from the many studies showing the extent that we struggle when selecting a health plan can illuminate this problem. For example, one study simulated the purchase of an ACA plan, using participants who passed a screening test for basic insurance literacy.Footnote ⁴² Even these more-literate-than-average respondents selected the best choice only about half of the time, and Wharton business school participants got it wrong over one-quarter of the time.Footnote ⁴³

Among enrollees in the University of Michigan’s employee health plan, over one-third of workers selected a plan that was identical to another in every way except that it had a more restricted provider network.Footnote ⁴⁴ No one would be better off enrolled in this plan. Importantly, this kind of network size variation is exactly what might differentiate a private and public option on a health insurance exchange. Another study of a large US firm found that a majority of employees chose a worse option and, as a result, they paid on average 24 percent more than they should have on premiums.Footnote ⁴⁵ Lower-income employees were more likely to make a bad choice.

Similar results occur in Medicare when beneficiaries choose among private prescription drug plans. One study found that 73 percent of enrollees could have chosen a plan with lower premiums with no risk of spending more on prescription drugs over the course of the year.Footnote ⁴⁶ Another estimated that only about 10 percent of enrollees choose their least-expensive option.Footnote ⁴⁷

On the ACA marketplaces, a significant share of people choose plans with the lowest monthly premiums but that make them ineligible for cost-sharing reductions, which reduce their deductibles and copayments when they use medical care.Footnote ⁴⁸ People who select these plans will likely spend more in the long run.

Others choose health plans that are not aligned with their own stated medical needs and preferences.Footnote ⁴⁹ In a study simulating the purchase experience on ACA exchanges, only one-third of respondents chose the cost-minimizing plan, based on their own anticipated medical care need.Footnote ⁵⁰ Forty-three percent over insured, on average overspending by 24 percent or $1,324 on premiums, and nearly a quarter underinsured.Footnote ⁵¹ The authors of this study estimated that if all people buying plans on the ACA exchanges had similar error rates as the study population, “the result would be roughly $7.1 billion of excess spending each year, borne by a population with low to moderate incomes.”Footnote ⁵²

This quick and only partial yet representative review of research on health plan selection is simply meant to illustrate that if the public option were not an obvious best alternative – and probably even if it were –people would not necessarily select it. If, in turn, the public option failed to gain significant market share, it would not exert pressure on the private insurers to offer better quality or lower-priced plans.

Thus, at the end of the day, the public option would coexist with private plans. Yet, it would not serve the other promise that Sitaraman and Alstott see in competitive public options, that of ensuring guaranteed universal access at controlled prices.

That’s not to say that having an undifferentiated public option would have no benefit. In 2018, there were 37 percent of counties where only one private insurer participates on the ACA exchange, which results in 17 percent of enrollees having only one choice of insurer.Footnote ⁵³ In these regions, evidence suggests health plans have higher premiums.Footnote ⁵⁴ In these areas, the mere presence of a public option might hold down premiums and premium growth. Such benefits are laudable, but far short of the transformative vision that the public option’s architects had for it.

Further, the marginal gains from a competitive public option would have come at a cost. The public option would have further justified preserving the existing system and problems with it. Injecting this option into the existing ACA exchanges would perpetuate, and perhaps even validate, this structure that is causing fundamental problems of inequity and regulatory bloat in health care.

6.3.3 The Problem of Market-Based Bureaucracy

This regulatory bloat is the beginning of the more structural problems with a competitive individual health insurance public option. While the promise of markets is that they are nimble and less bureaucratic than direct regulation, market-based approaches to health insurance have produced exactly the opposite: massive regulatory scaffolding to establish choice infrastructure and ongoing technocratic tinkering to try to fix the market’s flaws and poor decision-making. Elsewhere, I’ve called this reality health care’s market bureaucracy.Footnote ⁵⁵

The ACA’s exchanges have cost billions of dollars and have demanded extensive regulatory investments and, at the end of the day, only provide coverage to 10 million people, less than 3 percent of the population.Footnote ⁵⁶ The federal government spent nearly 5 billion dollars on state grants to establish exchanges and continues to spend 1–2 billion dollars annually to operate healthcare.gov, the federally funded exchange.Footnote ⁵⁷ The effort to overhaul healthcare.gov after it failed to work on its initial launch cost $1.7 billion, compared in an initial budget of $93.7  million.Footnote ⁵⁸

States with their own exchanges must fund a large part of their ongoing operations. California estimated it would spend $534 million, excluding federal grants, by the end of FY2017 on administration of Covered California with ongoing annual costs of over $350 million, funded out of plan assessments.Footnote ⁵⁹ Even a smaller state like Vermont will spend about $50 million annually to run its state exchange.Footnote ⁶⁰

Costs also include opportunity costs. The efforts needed to bolster and refine the exchanges have consumed health insurance regulators – at both the state and the federal level. They have commanded oversized technocratic analysis of exchanges and their successes and shortcomings, with some of the most talented researchers and think tanks consumed by this task.Footnote ⁶¹

For example, the Department of Health and Human Services proposed, revised, and issued hundreds of pages of federal regulations to implement the exchanges.Footnote ⁶² From the passage of the ACA through the end of the Obama Administration, the Centers for Medicare and Medicaid Services (CMS) promulgated twenty-four new rules and generated sixty-four guidance documents with respect to the exchanges alone.Footnote ⁶³ An entirely new office, the Center for Consumer Information and Insurance Oversight, was established within CMS in part to implement them.Footnote ⁶⁴ In parallel, state regulators have been doing the same.

Even more, sociologically, this labored creation and preservation of a market-based structure reinforces the idea that choice of health plan is sacred and should be a primary goal, even if the choice of plan it enables offers little meaning or value. By becoming part of this market bureaucracy, the public option would reinforce the value that is at its very core: choice defined in a narrow, microeconomic sense. Ironically, as compared to public options in other domains like swimming pools or shipping services that can improve access or quality, a public health insurance option would perpetuate prioritizing choice as a value over high-quality, universal, and affordable access. In some domains, as Sitaraman and Alstott assert, a competitive public option might simultaneously advance values of equity and freedom or autonomy, but when it comes to health insurance and when freedom is defined as market choice, these two values are at odds.

6.4.1 Baseline Coverage

One way public health insurance and private health insurance can coexist is with public health insurance as a baseline plan to cover basic health needs. This approach is undoubtedly the surest way to achieve Sitaraman and Alstott’s vision of what public options can uniquely deliver: guaranteed access at a controlled price. A baseline plan is available to all. The government sets payment rates for goods and services, working to keep spending as low as possible while maintaining sufficient supply of providers.

Health-care systems all around the world follow this model, in countries such as Canada, France or England. The details can take many forms. Benefits can be more or less extensive. This, in turn, defines the nature of its relationship with private supplemental coverage that fills in what is not covered. Medicare for All is one idea in this mold, and the details would determine how the public and private coverage coexists. A more comprehensive version, like the model advanced by Senator Bernie Sanders, is not a baseline model because it would leave little role for private insurance, but a more politically realistic version of Medicare for All would likely be similar to what other countries do.

Private insurance would thus serve the kinds of roles it does elsewhere – supplemental coverage to fill in gaps if the public coverage does not pay for the full costs of care, or complementary coverage to cover goods and services not publicly financed at all. For example, in the existing Medicare program, original public Medicare is a baseline. It pays on average about 60 percent of the total health-care costs of the enrolled population, and 90 percent of Medicare enrollees have a secondary plan to fill in the gaps.Footnote ⁶⁵ Current enrollees use private coverage in two ways. Some people buy a private supplemental plan, which they layer on top of original Medicare. Others choose a private Medicare Advantage plan as a replacement for traditional, fully public Medicare, and this privately administered plan covers everything that original Medicare would and fills in the gaps.Footnote ⁶⁶ Either way, a lion’s share of health care costs is publicly financed, directly or in the form of payments to private Medicare Advantage insurers.

A less transformative version of a baseline plan would be to offer a baseline to everyone, but to allow certain people to opt out, on a temporary or permanent basis. A more permanent opt out might allow people with incomes above a certain level to opt for private coverage as their primary coverage, as in Germany.Footnote ⁶⁷ But as Germany illustrates, this approach can be thorny because it invites in inequity in access, where people who opt out might buy into the better doctors, quicker access, and higher quality care.

Although a universal baseline approach would best serve the goal of health equity, it is the hardest to achieve politically. Efforts at universal, public health-care financing have been proposed and defeated repeatedly in the United States over the past century.Footnote ⁶⁸ It may prove impossible to create a hybrid system with public coverage as a baseline in a country where there is already a strongly embedded private system (although starting from scratch can be done, as Taiwan demonstrated). Hybrid systems may work best in domains where the public option exists first and the private ones fill in later, as in primary education or libraries. Yet, even then, there are significant challenges to maintain a balance between the two when the boundaries of play overlap or are not well demarcated.

6.4.2 Gap-Filler

Another way to use public health insurance productively is as a gap-filler for everyone who does not have an alternative. For example, Medicare or Medicaid could be made available to anyone who does not currently meet the criteria for these programs and who does not have quality access to ESI or a subsidized private health insurance plan on an exchange.

A more ambitious version of this idea would be to subsume the ACA’s exchange population as well. In the end, half of the population would have private coverage through ESI and the other half would have public coverage of some flavor – Medicare, Medicaid, CHIP, VHA, and IHS. The public coverage would still vary by program, so it would lack any uniform policy characteristic, but it would at least ensure that people did not fall through the cracks and the government could attempt to control spending for at least half of the population. In this version, public and private would coexist population-wide, with people sorted roughly into one or the other. It would offer the flexibility Hacker discusses in Chapter 5 and the potential, even if small, for private insurance innovation. The greatest challenge would be to ensure that the two paths remain equitable.

6.4.3 The Creeping Public Option

A compromise between a universal baseline and a gap-filler is to start with gap-filling and to expand over time. This was the initial vision that the architects of Medicare had in the 1960s – that it would first cover older Americans and would over time expand to cover the whole population.Footnote ⁶⁹ This vision is also what led opponents to universal, public coverage to advocate for simultaneous passage of Medicaid.Footnote ⁷⁰ They bet (correctly) that once the elderly, pregnant women, and children were covered, it would reduce motivation for any monumental expansion of public insurance.

A creeping expansion, however, could still happen now. Candidates in the Democratic presidential primaries in 2019, including then Senators Kamala Harris and Elizabeth Warren, proposed creeping public options. Legislation introduced by Representatives Rosa DeLauro and Jan Schakowsky, and informed by Jacob Hacker (Chapter 5), proposes to use Medicare first to fill in existing gaps.Footnote ⁷¹ It also gives employers the option to enroll their employees in public coverage instead of offering ESI. Over time, it envisions Medicare would grow in its reach by enrolling all newborns, which over a generation would eventually create a universal baseline program and could take any of the forms of private and public coexistence described earlier.

Even if the creep stops short of universal, a public option that covers many Americans could improve health equity and the health-care system significantly. In fact, the larger and more heterogenous of a population a public option reaches, the more durable it will be politically and the more people will be invested in its success, as illustrated by high popular and political support for Medicare.Footnote ⁷²

These kinds of proposals for creeping public options recognize the challenge of transitioning from the current system to one where a public, baseline option serves a broader social function and benefit. They also will face deep resistance from the same parties whose livelihood will be threatened or transformed if public insurance expands – namely, private insurers and providers and suppliers whose reimbursement rates will be squeezed. They may, in fact, be no more politically feasible than a direct step to Medicare for All and much more complex, growing the market bureaucracy even more.