September 13, 2021
According to a draft document being circulated, the Democrats are looking to save $600 billion over a decade by negotiating lower prescription drug prices. The basic story here is that the United States pays more than twice as much, on average, for prescription drugs than Germany, France, Canada, and other wealthy countries. If we paid something closer to what these countries pay for drugs, there would be very large potential savings.
The Democrats are hoping to use these savings to pay for expanding Medicare (possibly also lowering the age of eligibility; 64 would be a good start), free community college, extending the child tax credit, and all sorts of other good things. While there is some skepticism as to whether the government can actually save $600 billion over a decade on prescription drugs, this is actually a very low target. Arguably, the full cost of the $3.5 trillion (1.2 percent of GDP) reconciliation package could be covered by savings on prescription drugs alone.
Projected Spending on Prescription Drugs
There is very little awareness of how much money the country spends on prescription drugs. This is partly due to the fact that most spending is by third parties, either the government or private insurers, rather than directly out of the pockets of patients. There also is little realization of how high drug prices are relative to the cost of manufacturing and distribution.
The figure below shows projected spending on prescription drugs over the next decade. Spending for 2021 is projected to be just over $500 billion, or 2.2 percent of GDP. The growth rates projected by the Centers for Medicare and Medicaid Services (CMS) imply that spending will rise to $894 billion by 2031, or 2.6 percent of GDP. The base for this spending is taken from the National Income and Product Accounts, Table 2.4.5U, Line 121. It is considerably higher than the spending data from CMS because it includes spending on drugs in hospitals, nursing homes, and other institutions. The CMS data only measures spending on drugs by patients directly.
Source: CMS, BEA, and author’s calculations.
As a target for potential spending, the figure assumes that without patent and related protections, we would be spending 20 percent as much as is currently projected. This is a plausible, and possibly even conservative target.
According to data from the Association for Accessible Medicines (AAM), the trade group for the generic drug industry, brand drugs accounted for 74 percent of spending even though they were only 11 percent of the prescriptions sold. By contrast, generic drugs accounted for just 26 percent of spending even though they were 89 percent of prescriptions. This implies that the average generic prescription cost just 3.6 percent of the price of the average brand prescription, or $29.70 per prescription. This figure would mean that we could save 96.4 percent of the money spent on brand drugs if we immediately got rid of protections and allowed them to be sold as generics. (As a practical matter, we would not see generic competition push down prices right away, since it would take several years for a substantial number of competitors to enter the market.)
But even this figure understates the true potential savings from eliminating patents and related protections. The AAM’s data is based on averages, but these numbers are skewed by the fact that many generics sell for high prices in the period immediately after they are introduced. This is due to the fact that they have limited competition, in part by design.
The first generic to enter a market is granted six months of exclusivity, a period in which they enjoy an effective duopoly with the brand drug. For example, a recent analysis found that price declines averaged just over 50 percent in the first year after a generic is introduced. This means that the average cost of generic drugs in the AAM analysis is inflated by the generic drugs that are sold at high prices in the first years after they come onto the market. Also, the price of generics is inflated due to the fact that the generic industry must incur legal costs to contest patent battles with the brand industry. In addition, generic drugs may also be paying royalties on secondary patents that are still in force even after the key patent for a drug expired.
In a world with no patents or related protections, it is likely that generics would cost close to half as much as we now spend since the market would be more fully competitive and the industry would face far lower legal expenses. The table assumes an 80 percent reduction in spending, which is likely a conservative figure. If the average price of a generic drug in a patent-free world was 50 percent of the current price and all drugs were sold as generics, the savings would be more than 85 percent of current spending.
The 80 percent reduction in projected spending would imply savings of $5.6 trillion over the next decade. Of course, the government does not pay for all prescription drugs in the country, so only around half of these savings, or $2.8 trillion, would accrue to the government. Although, since most spending on health insurance is tax-deductible, the government should recoup some of the private sector savings in the form of higher tax revenue. Also, from the standpoint of reducing demand in the economy and limiting the risk of inflation (the purpose of taxation at the federal level), the reduction in private sector spending on drugs is likely to have almost as much impact as the savings to the government.
Research and Development Costs
This picture leaves out the research and development costs incurred by the industry. In 2020, the industry spent $105.7 billion on research and development. If we assume that spending on research increases at the same pace as spending on drugs, then the industry would be spending $1.5 trillion on research and development over the course of the decade, on its current path. (This is in addition to the $45 billion a year that it currently spends through NIH and other agencies.) If the government had to replace this spending, and we assume that a dollar of government spending has the same value as a dollar of private spending, then we would need to subtract $1.5 trillion from our savings on drug prices to cover these research costs. That would still leave $4.1 trillion in savings.
Whether or not a dollar of government spending has the same value as a dollar of private spending under the patent monopoly system is a debatable point. It could be argued that private companies, with their own dollars on the line, will be more efficient, but there are reasons the story could go the other way. (Moderna, which developed its coronavirus vaccine almost entirely on the public dime, is a great example of public funding being very effective.)
First, there are better and worse ways of doing public funding. In my view, the best route would be to pay out money for research to drug companies on long-term contracts, similar to the way the Defense Department contracts out for major weapons systems. (I describe this more fully in chapter 5 of Rigged [it’s free].) The contractors would have a strong incentive to ensure their money was well spent since presumably, they will want their contracts renewed when they expire.
A big advantage this system would have over military contracting is that everything would be fully open. While there are good reasons for secrecy in designing weapons systems, there is no reason to want to keep research findings on a cancer drug or treatment for heart disease secret. In fact, putting all results on the web in a timely manner should be a condition of getting public funding. This rule would apply not only to contractors but also to all subcontractors. (It is likely that any major drug company winning a contract would contract out for much of the research, just as they do now. Innovative startups are likely to be more efficient with research dollars than large well-established companies.)
If all research is quickly made available to the entire scientific community, we are likely to make progress more quickly than in a world where drug companies are looking to lock up their work behind patent monopolies. Researchers can quickly learn from each other’s research, building on successes and avoiding failures.
In addition, research that is primarily rent-seeking in nature would be largely avoided. In many cases, drug companies look to develop drugs that are not expected to produce major medical benefits over existing drugs, but rather will simply give them a portion of the patent rents earned by a major breakthrough drug. The Food and Drug Administration puts more than 90 percent of its drug approvals in the “standard” category, meaning that they are not a qualitative improvement over existing drugs.
To be clear, duplicative drugs are not of zero value. People may react poorly to a certain drug, whereas they may be fine with another drug in the same class. Also, some drugs do not mix well with other drugs. However, as a general rule, we would probably like to see research dollars spent finding drugs for conditions where effective drugs do not currently exist, rather than developing the seventh, eighth, or ninth drug for a condition.
There also will be some amount of research spending that will be needed to ensure that production standards are safe and efficient. This spending would be necessary even if everything was sold as a generic, since it will be needed for any manufacturer. The government would not need to replace this spending, which is likely in the neighborhood of 10 percent of the total.
The Opioid Scandal and Patent Monopoly Financings
There is one other very important reason for believing that publicly-funded research could be more effective on a per-dollar basis than patent monopoly financing. Patent monopolies give drug companies an incentive to lie about the safety and effectiveness of their drugs.
While it is common for drug companies to be less than completely honest about the safety and effectiveness of their drugs, we saw a very dramatic example of this dishonesty with the opioid crisis. Purdue Pharma and other major opioid manufacturers paid out tens of billions of dollars in settlements in response to lawsuits alleging that they deliberately concealed evidence of the addictiveness of their drugs in their efforts to promote them to doctors.
To be clear, the allegation was not that these companies made a mistake. The claim was that they knew their drugs were highly addictive, but claimed the opposite when they tried to convince doctors to prescribe them to patients.
There will always be mistakes in research, some with very large consequences. But the issue here was deliberate deception. Had it not been for the opportunity to sell their opioids at patent monopoly prices, the drug companies would have had far less incentive to misrepresent their addictiveness.
If we eliminated patent monopolies for prescription drugs, we also eliminate this perverse incentive. This is likely to lead to considerably better care for patients, since doctors will have honest research guiding their decisions in prescribing drugs.
Drug Savings and Covering the Cost of the Reconciliation Package
If we think clearly about how much we are spending on drugs, compared to the cost of manufacturing and distributing them, it should be clear that the $600 billion savings figure being tossed around by Democrats is very doable. It is plausible to envision savings many times larger, if we are prepared to jettison the patent monopoly system of financing drug research and replace it with something more modern.
Of course, we should be looking to other routes for offsetting the spending in the reconciliation package, most obviously by increasing taxes on the rich and corporations. My favorite in the latter category involves switching the basis for the corporate income tax from profits to returns to shareholders. The latter are simply dividends and capital gains. This can be done on a simple spreadsheet. (That would be a single spreadsheet for all publicly traded companies.) We could put the tax shelter industry out of business since there is no way to hide what is being doled out as dividends or the rise in share price. (To be clear, this is a corporate income tax, not an individual tax – the corporations will be taxed based on the money they made for shareholders.)
The Democrats in Congress are in a position to make changes that will have a huge impact on the lives of tens of millions of people. Covering the cost should not be a problem. We can get much or all of the way there from savings on prescription drugs alone, but we have many other good options to add to the mix.
 The projections in the chart use calendar year 2019 spending on prescription drugs as a base, since the 2020 numbers were distorted by the pandemic. This is taken from the National Income and Product Accounts, Table 2.4.5U, Line 121. The numbers for later years use the projected growth rates from the Center for Medicare and Medicaid Services, National Health Expenditure Projections, 2019-2028, Table 2. The spending growth rate for 2027 to 2028 (6.0 percent), is assumed to continue for the next three years.
 National Income and Product Accounts, Table 5.6.5, Line 9.