The Next Great Recession

“There will always be a business cycle, and white-collar workers will get hit in the next recession like they always do in recessions.”

Robert Reich – American economist, Secretary of Labor 1993–1997

 

“The global financial crisis – missed by most analysts – shows that most forecasters are poor at pricing in economic/financial risks, let alone geopolitical ones.”

Nouriel Roubini – American economist

 

The Great Recession, lasting from December 2007 through June 2009 in the US, led to a global recession in 2009, adversely affecting global pharma and healthcare markets (1, 2, 3, 4, 5, 6, 7). In the US, in particular, the Great Recession had a transformative effect on pharma – shaking the very structure of the industry. It is now considered more recession-sensitive than ever before, with cost-shifting from payers to the patient and the growth of expensive personalized medicines, often involving orphan drugs to treat rare diseases, adding greater recession sensitivity to drug demand (8, 9, 10, 11).

There are multiple signs that, despite 2018 ending on some good macroeconomic news, structural problems persist (12). Fears about the next recession in the US – which would have a knock-on effect in global markets – started in the summer of 2018 and really began to manifest in significant stock market volatility in December 2018, with continued risks being called out to this current day (13, 14, 15). In the US, the Federal Reserve (“the Fed”) reduced the federal funds rate, a benchmark used for rates on credit cards and mortgages, by a quarter point at the end of July 2019 – seen by economists and market analysts as an expansionary move – in response to a growing fear about domestic and global economic and geopolitical conditions (16, 17). This action mirrors sentiments recently expressed by the Federal Reserve Federal Open Market Committee (FOMC) meeting in June 2019, which noted that risks to the future economic outlook were balanced, meaning there are both positive and negative signs in continued economic growth (18). Participants also expressed gloom about the global economic future at the annual Federal Reserve Economic Policy Symposium in August 2019 (19).

Economists surveyed earlier this year by The Wall Street Journal (WSJ) assessed a 25 percent and 57 percent chance for a recession in 2019 and 2020, respectively (20). Markets nervously reacted to the August 2019 US jobs report, noting that the numbers reflected a global slowdown occurring outside the US, stunting growth – likely caused by trade uncertainty created by the Trump administration (21, 22). The Federal Reserve finally announced a small quarter-point reduction in the discount rate to just above 2.0 percent on September 18, 2019, reflecting slowing economic growth in Europe and China, and the effects of global trade uncertainties that add to a drag on future growth prospects (23).

So, while the prospects of when a recession will start are uncertain, two key questions to ask are: 

i) “How does a recession affect pharmaceutical drug demand?”
ii) “Are pharma companies prepared for this event?” 

Or, put another way: Have any lessons from the Great Recession experience been incorporated into future pharma business planning when the next recession occurs?

Shifting drug demands

There is an overall dearth of empirical evidence connecting the sensitivity of biopharma industry demand for specific drugs to a recession. Publicly available research literature tends to look at the effects of a recession on drug spending in the aggregate, which masks the different types of recessionary effects on drug demand relative to factors that can be affected by company actions, such as sales, marketing, payer contracting, and pricing. This comparison of effects is important since macroeconomic trends, such as a recession, are taken as a given to an individual company, while sales and marketing are under management control. Thus, building models that measure how the advent and severity of a recession can be mitigated by management control variables is important insight for executives. We know the Great Recession had different effects by geography, especially by specific region or metropolitan area. An analysis at a highly disaggregated geographic level and by specific brand is needed to understand true recession effects – and, importantly, how mechanisms available to executives can dampen those effects.
Using economic theory and practical experience on the determinants of pharmaceutical demand functions, a severe recession would have effects on specific drug demand via four empirically measurable mechanisms.

1) The disposable income effect
A reduction in real (inflation-adjusted) disposable income caused by falling or stagnant wages reduces affordability of (and access to) drugs. A decrease in disposable income relative to any out-of-pocket cost to access drugs will decrease drug demand. One should also see substitution effects in the proportion of biologics and branded drug demand relative to biosimilars and generics demand due to disposable income effects.
 

2) The unemployment/labor force participation/loss of insurance effect
Unemployment often results in a loss of patient health insurance. Since the Great Recession, the labor force participation rate (the number of people who are employed and unemployed but actively looking for a job divided by the total number of eligible workers between the ages of 16–64) in the US has stabilized at around 63 percent, a number not seen since the mid-late 1970s (32). The passing of the Patient Protection and Affordable Care Act (ACA) and the availability of health insurance through exchanges not connected to employment and subsidized premiums based on income may mitigate the effects from a poor labor market and declining access to employer-provided health insurance. The degree to which access to drug insurance through the ACA mitigates recessionary effects depends on the cost of premiums relative to people’s limited income and the choice of plan.
 

3) The wealth effect
A severe recession may reduce the value of financial and physical assets; for example, changes in the value of equities and bonds versus changes in assets like the price of housing. People who are in or close to retirement may use these assets for future spending, thereby reducing drug affordability – similar in response to a disposable income effect.
 

4) The Government effect
A recession depresses tax revenues, while increasing entitlement-program spending for the poor, which means public deficits will grow. In the US, this effect is more severe for states since governments at this level generally must run balanced budgets, causing pressures to reduce spending. This means government-provided health insurance programs may become more restrictive in their drug benefits.

Given that a severe recession produces varying consequences on different segments of the population based on their relationship to the economy (for example, a rising unemployment rate has little effect on drug demand for therapy classes dominated by patients who are elderly or retired individuals), the above four effects would impact the pharma industry in the following ways:

• Lower utilization of patented biologics and branded drugs, with greater sensitivity seen for more expensive specialty medicines (which should exhibit greater price and income elasticities).
• Greater utilization of biosimilars and generics as less expensive substitutes, meaning that a severe recession will trigger faster and greater adoption of biosimilars and generics.
• Lower utilization of drugs in therapy classes where patients must absorb a proportionally greater out-of-pocket expense.
• Lower drug compliance (the filling of a prescription received from a physician) and adherence (how patients take their medications). For example, patients spreading daily medication usage over two days.
• Greater utilization of mail order relative to retail pharmacy as a channel to receive medications at a lower cost.
• Greater physician demand for samples, especially in geographic areas or population segments that are more sensitive to changes in economic conditions.
• Greater demand for enrollment in company patient-assistance programs to offset the effects of losing drug coverage and affordability/access issues due to lower income from unemployment.
• Patients with multiple conditions in difficult economic straits will more likely choose continuing drug therapy for symptomatic conditions over asymptomatic ones. Thus, for example, older individuals who have osteoarthritis, diabetes, and hypertension will more likely choose continuing their osteoarthritis medication over the latter conditions, even though controlling their diabetes and hypertension is likely more medically important. In short, patients may make suboptimal healthcare choices.
• Greater movement by patients into catastrophic higher deductible health plans, as they are less expensive but will also translate into receiving fewer medications given the higher out-of-pocket expense. Less access to healthcare and drugs will have adverse consequences on health outcomes and overall medical care spending.
• Greater geographic variations in drug demand utilization seen around the world as the recession could generate different local and regional effects. For example, during the Great Recession, specific cities and regions in the US that relied more on heavy manufacturing and auto production for their economic base (like Detroit and “rust-belt” states) were severely impacted. Greater drug demand effects would be seen in local areas and regional economies that were less economically diversified and more susceptible to any one change in a recession index. 

Datasets exist that capture all the above relationships which in turn can be empirically measured using various econometric models to support sales/marketing/payer tactics.

Preparing for the worst

“What we know about the global financial crisis is that we don’t know very much.” Paul A. Samuelson, Nobel Prize-winning economist.

Is the US pharma industry prepared for another recession? The best educated guess from this author and economist: “No!” Pharma commercial organizations, by their nature, presume management control variables, such as sales, marketing, and contracting efforts, directly affect prescription volume to the exclusion of external factors. They are not looking at macroeconomic effects on prescription sales, nor do they have the modeling expertise and experience in analyzing such effects. In addition, as macroeconomic conditions are beyond the impact of individual companies, the view may be that their effect on prescription volume is not something that needs be considered in the business planning process. Such a view is incorrect. Though macroeconomic conditions are exogenous factors to be treated as a given by companies, it is important to know the effects of such trends on prescription volume and other key outcomes – and what can be done to mitigate those effects.
Given the typical risks and uncertainties of making future forecasts, there are a number of precautions that pharma companies can take to prepare for the effects of another recession.

• Predict when a recession will start. Admittedly, such predictions are challenging! However, an array of forecasts exists from governmental institutions, private sources, and academic organizations. Waiting for an official proclamation that a recession exists is a far worse scenario. The measurement that a recession exists essentially means that at least two consecutive quarters of an economic downturn have occurred. Waiting for the official announcement means at least a half-year has already passed, meaning the effects from a recession have had time to affect patients and other healthcare system stakeholders (which may not be reversible), and less time for business plans to be enacted to help mitigate the effect.
• Estimate the severity. Assessing severity will help determine the amount and type (sales, marketing, or managed markets) of resources that need to be deployed to mitigate recession effects.
•  Determine how long it will last. From a business planning standpoint, you need to know how long resources must be devoted to mitigating recession effects at the brand level.
• Assess the differential effects. The Great Recession had differential effects by geography, industrial sector, and socio-demographics. And the next one will, too.
• Prepare contingency plans that can be put in place at the first sign of a recession. Pharma companies have a number of processes where the risks and effects of a recession to the business can be incorporated into future plans. First, brands go through quarterly and annual business reviews that assess both strategic and tactical plans. Second, prescription and financial forecasts are developed based on market and environmental trends. These forecasts can easily incorporate recession risks at the national level, but more importantly at the metropolitan area, especially for the top local markets key to brand success. Third, companies generally create a “risk register” or a list of future potential events that could occur (along with their likelihood of occurrence, severity of business impact, and rating of business importance or priority) that could affect business operations. The occurrence of a recession could be included on this list of potential future events that place the company at risk.

Implications for US pharma

One key difference between the Great Recession and any future recession relating to pharmaceutical demand is the availability of health insurance that is now detached from employment through market exchanges via the ACA, which was not previously in existence; in theory, people do not have to lose their coverage when they lose their job. However, if the insurance is too expensive, people facing economic hardships will likely drop coverage, significantly raising the out-of-pocket drug costs, thereby decreasing patented drug demand and patient adherence. Moreover, much of the health insurance coverage expansion was for Medicaid, which for pharma companies is low-margin business and where plans generally have strong biosimilar and generic preferences. Also, the repeal of the ACA mandate and financial penalty has likely contributed to a rollback of healthcare coverage, making people and, thus, drug demand more susceptible to an economic recession. Lastly, people may choose to change the quality of their health plan, from more comprehensive and lower deductible/co-pay coverage to essentially a catastrophic plan but with poor coverage and higher out-of-pocket expenses when it comes to general health and drug maintenance and coverage to reduce premium costs.

We can therefore draw the following conclusions about the next recession and its effect on the US pharma industry:

• Signs are building for a recession in the near future – potentially 2020 – according to recently surveyed economists and uncertainties concerning global economic/geopolitical conditions.
• Many signs point to a potentially deeper recession than the Great Recession of 2007–2009 given structural economic problems, slower growth outside the US, international trade issues between the world’s largest trading partners, and geopolitical risks/uncertainties.
• The pharma industry’s shift to expensive specialty medicines, coupled with payer costs shifting to patients, means drug demand is increasing in recession sensitivity.
• The ACA will likely do little to help patients pay for medicines and will not mitigate the negative effects on drug demand when a deep recession occurs.
• Issues with government debt (also seen in many countries worldwide) will mean added economic constraints by the public sector against spending increases to maintain healthcare and drug services. Such constraints will likely result in increasing controls in the US and beyond against patented biologics and branded drug utilization and spending, and favoring substitution to biosimilars and generics.
• Pharma companies are likely unprepared. They may not currently have the analytics or processes in place to predict recessionary effects on drug demand or know how to implement changes in sales, marketing, and payer tactics.

The important relationships that hold the key to understanding the full effects of a recession on drug demand (and the resulting impacts) can be predicted, estimated, and assessed prior to the actual event. Models and analytics can help design and implement plans to mitigate any effects from a recession on physician, patient, and payer behavior.

Alleviating adverse effects

What role should company commercial analytics have in mitigating recession-induced drug demand effects? There is a wealth of historical economic data that already exist, while economic forecasting companies project forward trends on the types of measures that would trigger recession-induced drug demand effects. Econometric inference models – developed at the local or regional level – can help determine the extent of drug demand effects from changes in specific economic factors relative to company management control and market-oriented variables (sales, marketing, and market access). Empirical results would reveal variations by geography in these relative effects given the wide diversity of economic conditions. Results from inference models could then be used to estimate future drug demand effects based on projections of economic activity and assumptions on company management control variables going forward.

There are numerous potential business insights gained from the previously stated research path of applying commercial analytics in combination with econometric inferential and prediction models to economic and non-economic data on drug demand. For example:

• The results may reveal surprising insights that economic trends play a much more significant role in affecting drug demand relative to management control variables than first thought. Given the growing trend toward launching expensive specialty medicines, the structure of economic variables is likely to play an even greater impact on drug demand relative to traditional sales and marketing channels. Marginal, elasticity, and relative importance (standardized coefficients) estimates could be derived from a wide range of drug demand models.
• A company can position commercial resources differently by local area according to economic dynamics. Local/regional differences in economic effects may suggest variations in managed care contracting, demonstration of greater drug value through promoting disease management programs, differentials in drug messaging through personal/non-personal/consumer promotion channels, and so on.
• National, regional, and key local area company financial forecast accuracy could be improved by introducing the effects from economic variables, which could determine the extent of any drag on financial forecasts from recessionary effects.
• Models could be used to determine not only specific drug demand by prescription type, but also by payer channel and brand to generic substitution ratios. One would expect that, as a recession becomes more severe and lasts longer, forecasts could be developed to see how many prescriptions move from third-party commercial to Medicaid, or from retail to mail order.
• More advanced modeling can be developed to measure the negative effects of a recession on drug utilization and adherence on patient health and economic outcomes. The reason for this connection is due to an increasing prevalence of performance-based contracts between pharma companies and payers/pharmacy benefit managers (PBMs). Recent published evidence looking at Detroit revealed that the Great Recession reduced overall population health (33). Making such connections in real time will require the use of patient-level claims data and electronic health records, along with applications of artificial intelligence and machine learning. Pharma companies can also understand more fully the effects of co-pay offset programs to mitigate recessionary effects on health and economic outcomes.
• Pharma companies can use artificial intelligence and machine learning to predict which patient segments per drug will have more affordability problems paying for medicines during a deep recession. Such algorithms can also be used to predict enrollment applications in patient assistance programs. Algorithms can initiate the next best action towards locality based on the predicted effects of changes in economic conditions.
• Analysis could estimate differences in the price elasticity of demand for different drugs by geographic regions that see variations in economic distress. Such analysis will also likely pick up differences in drug price elasticities for certain life-threatening and rare conditions versus other conditions; similarly, we should see differences between symptomatic versus asymptomatic health conditions.

In short, another recession is highly probable. Companies need to be aware of the full extent of the potential impact. Management control should be exercised now to mitigate adverse effects.

This article was co-published with Axtria, a big data and analytics company: https://insights.axtria.com/whitepaper-how-will-the-next-great-recession-affect-the-us-pharma-industry