WAG/RAD: Pressure on generic dispensing margins likely to be much more permanent than guidance implies

ginna
Print Friendly
Share on LinkedIn0Tweet about this on Twitter0Share on Facebook0

Richard Evans / Scott Hinds / Ryan Baum

SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES

203.901.1631 /.1632 / .1627

https://twitter.com/images/resources/twitter-bird-blue-on-white.png richard@ / hinds@ / baum@ssrllc.com

@SSRHealth

October 14, 2014

WAG/RAD: Pressure on generic dispensing margins likely to be much more permanent than guidance implies

  • WAG & RAD both recently lowered guidance; both blame falling generic dispensing margins. In each case the companies point to pressures on both sides of the margin equation – rising generic acquisition costs, paired with lower-than-expected reimbursement
  • Unravelling the relative importance of these two effects is strategically crucial; rising generic acquisition costs are relatively transient; rising reimbursement pressures are more likely to be permanent
  • Evidence indicates generic acquisition costs played little if any role in reducing companies’ generic margin expectations; this implies falling reimbursement pressures played the major role, and that lower generic dispensing margins are here to stay
  • The companies make 3 arguments regarding the generic acquisition cost trend: 1) acquisition costs grew at a surprising rate on ‘existing’ products; 2) acquisition costs on newly launched generics fell more slowly than normal; and 3) newly launched generics are being made by fewer manufacturers than expected
  • None of these arguments are supported by evidence. On a sales-weighted basis, ‘existing’ product generic acquisition costs indeed experienced major hikes through February 2014, but they remained stable thereafter through the months when the companies issued/confirmed higher guidance (and subsequently lowered it). Also on a sales-weighted basis, acquisition costs of newly launched products fell faster more recently (Nov. 2013 – July 2014) than they did last year. Finally, the number of manufacturers for new generics has been no different than should have been expected given the dollar size of the parent brands, the presence or absence of an OTC version of the parent molecule, and the complexities of manufacturing any given new generic
  • By simple process of elimination this argues that reimbursement pressures are likely to be the major cause of falling generic dispensing margins at retail. The companies acknowledge the trend to narrow networks has played some role in reduced reimbursement; we would argue that narrowing of networks appears to have played a major role, will continue for some time, and will result in steady additional pressure on generic margins. As context, consider that REAL retail pharmacy dispensing margins have grown faster than real pharmaceutical prices since at least 2001, despite the fact that there are no fewer outlets now than in 2001
  • Unmentioned as a driver of lowered guidance is the shift toward reimbursing pharmacies on the basis of NADAC (which is likely to result in lower generic margins than reimbursements based on AWP). NADAC became available to all US retail pharmacies in April of this year, meaning 2014 has been the first negotiating season (for 2015 benefits) in which payors could reasonably be expected to ask retailers for NADAC-based contracts

Both WAG and RAD have significantly lowered guidance, and both have blamed pressure on generic margins. And, the companies have laid blame on both sides of the generic margin equation – the ‘buy-side’ cost of acquiring generics, as well as the ‘sell-side’ level of reimbursement received from third party payors

The question of which effect is larger – rising generic costs or falling reimbursement – is of crucial strategic importance. Rising generic costs presumably are transient; in contrast, pressures on reimbursement are more likely to be permanent

‘Buy-side’ trends are unremarkable

Price inflation on ‘existing’ generic products

Both companies point to ‘same basket’ price inflation, i.e. accelerating list price growth on existing products, as a source of buy-side pressure on generic margins. To put this in scale, we calculated a sales-weighted index of monthly change in generic acquisition costs from 2Q13 to present, shown as Exhibit 1. Sales-weighted generic acquisition costs indeed spiked significantly multiple times through February 2014 as well as in the last 30 days; however their lull from mid-February to early September coincides with the period when WAG and RAD issued and later ratcheted down guidance. In WAG’s case, taking the WSJ’s timing of WAG events at face value, the internal April 10 forecast of $8.5B in fiscal ’16 pharmacy profits was made not long after the largest spike in acquisition costs in our time series, and there were no further significant moves in generic acquisition costs between that internal April 10 forecast and 1) the retraction of fiscal ’16 sales goals on June 24th; 2) the internal revision of the fiscal ’16 pharmacy profit forecast on July 9th; and 3) the issuance of lower fiscal ’16 guidance on August 6th. Just as important, it seems implausible that an unexpected inflationary spike would be of enduring relevance when new guidance was issued more than a year later. Similarly, RAD didn’t issue fiscal ’15 guidance until April 10th; their second downward guidance revision did occur immediately following the most recent cost spike in September, however this spike was well within the proportions of many earlier spikes that did not result in lowered guidance. The timing of spikes in acquisition costs appears unrelated to either internal changes in estimates or external changes in guidance. The only reasonable conclusion we can draw is that spikes in generic acquisition costs are not a major driver of the companies’ lowered expectations

Rate of price deflation on new generic products

Regardless of the price trend on ‘existing’ products, another stated explanation has been that prices for new generics didn’t fall as quickly as expected, particularly since new generics have a disproportionately large impact on overall margin. We compared the sales-weighted trajectory of new generic prices as a function of weeks since launch for the cohorts of new generic products launched from November of 2013 to July of 2014, and from February of 2013 to October of 2013. We find that sales-weighted prices for the more recent new generics (solid green line, Exhibit 2) fell faster than sales-weighted prices for new generics entering the market in the preceding period (solid black line, Exhibit 2). Thus if anything, the companies’ experience with price trends on new generic launches was better than it was in the period that fully preceded the more aggressive guidance that ultimately had to be lowered. Here too, the only reasonable conclusion is that changes to the new generic price erosion trend are not a major driver of the companies’ lowered expectations

Number of manufacturers of new generics

The final argument regarding generic acquisition costs was advanced by RAD on their September 18th call. Specifically, the company argued that “We also had some newer generics that lost exclusivity, that didn’t come to market as profitably as we thought, honestly, in part because they’re just weren’t as many manufacturers as we expected”. Our experience is that the number of generic manufacturers launching a given molecule is in rough proportion to the dollar sales of the parent brand, the complexity of producing the generic, and whether an OTC version of the parent molecule also is being launched. All evidence indicates that this normal relationship held in the period preceding the companies’ lowering of guidance. Exhibit 3 plots the number of manufacturers (y-axis) offering a given generic molecule as a function of the parent brand’s sales (log x-axis). Black dots represent new generic drug launches in the February 2013 to October 2013 timeframe, green dots the more recent November 2013 to July 2014 timeframe. Excluded from Exhibit 3 are ‘special cases’ that are more complex to manufacture, namely topicals, sublinguals, narcotics; and also drugs with OTC equivalents. Exhibit 4 is identical to Exhibit 3, but includes the ‘special’ cases (earlier period special cases are open black circles; recent period special cases are open green circles). We find no evidence to indicate that either company should have been surprised by the number of generic entrants in the period immediately preceding the lowering of guidance

Delays in key generic launches (e.g. Nexium)

Retail margins on generic drugs are largest in the period immediately following the entrance of cheap generic alternatives to a brand, and are correlated with the amount of sales at stake. The continued delay in the launch of Ranbaxy’s generic version of blockbuster Nexium was mentioned by WAG and RAD as a contributing factor to margin pressure in the then-current/next couple quarters. While this specific case is surely negative in the present, it is nonetheless difficult to understand how the companies’ previous guidance discounted regulatory risks surrounding a single manufacturer with a checkered past of infractions. In any event –as the companies clearly remark – such special cases clearly have no bearing on guidance for future fiscal years

By simple process of elimination, ‘sell-side’ reimbursement pressures appear to be the main force behind companies’ reduced expectations for generic gross margin

Both companies acknowledge reimbursement pressure as an important driver of falling generic margins, though the specific mechanics of tightening reimbursement are left unclear

Transient MAC/AWP pressures

We recognize that generic reimbursement ‘naturally’ tightens when generic prices are rising. Payors often are slow to update maximum allowable cost (MAC) lists when retailers’ acquisition costs are increasing; because ‘MAC’ defines the amount retailers are paid on many products, generic margins compress until the MAC values are updated to reflect retailers’ higher generic acquisition costs. And, generic manufacturers can be slow to update their list (aka AWP or ‘average wholesale price’) prices. Because retailers’ reimbursement is often set using AWP as a basis – especially on newer products – rising generic acquisition costs pressure margins until and unless AWP ‘list’ prices are increased proportionally

It is likely that certain WAG/RAD reimbursement contracts were designed with unfavorable terms (i.e. provisions allowing ‘sticky’/fixed rates) that consequently have maintained margin pressure for protracted periods following generic cost spikes. Notwithstanding, the delayed timing of the companies’ guidance revisions relative to the cost spikes (see Exhibit 1) and, in WAG’s case, the distant period of guidance (fiscal ‘16) again discredits temporary payment rate lag as a major driver of tightening generic margins

The durable and accelerating trend toward narrow networks

The companies also acknowledge falling average reimbursement in general, and in particular from payors that are moving toward narrower pharmacy networks. The story around WAG seems to center more on narrowing of Part D networks than on narrowing of non-Medicare commercial networks. The story around RAD seems to draw less of a distinction between Part D and non-Medicare commercial payment pressures, which syncs with what we know about tightening of commercial networks – ESRX being the most prominent example

Looking at the trend in retail pharmacy margins, and its relation to the number of US pharmacies, it’s easy to see the potential for payors to reduce pharmacy margins by moving to narrower networks. REAL retail pharmacy dispensing margins[1] (solid green line, Exhibit 5) have grown steadily since 2001 – faster even than average prescription drug prices[2] (dashed green line, Exhibit 5) – even though the number of retail outlets per 100,000 US persons has remained roughly constant (grey area, right axis, Exhibit 5). This trend is remarkable in multiple respects: real gains in retail margins have grown faster than the (considerably rapid) real growth in the price of what’s being sold, and in the absence of significant change in the supply of retail outlets. Equally remarkable is that real pharmacy dispensing margins have grown even as the mix of retail outlets has shifted toward very large retailers who have every reason to use their pharmacies as loss leaders to drive purchases of non-pharmaceutical merchandise (e.g. WMT, TGT, KR, others)

The elephant in the room: NADAC

Missing from the general discussion is the impact of the new national average drug acquisition cost (NADAC) data, which has been available in all or very nearly all US retail pharmacy outlets since April of 2014 – making the 2014 negotiating season for 2015 benefits the first in which payors could reasonably ask retailers to accept reimbursement based on NADAC, as opposed to AWP or MAC

Unlike AWP, which as a ‘list’ price bears no predictable relationship to retailers’ true acquisition costs for generics, NADAC is an index that much more closely reflects retailers’ generic acquisition costs – in fact the index is compiled using actual generic acquisition costs from a rolling nationwide sample of retail pharmacies

Under AWP-based reimbursement, retailers are paid AWP less some percentage for generic prescriptions; because AWP bears no predictable relationship to actual acquisition costs, and because retailers may know more about their true acquisition costs than whomever is paying them based on AWP, the retailer is on the winning side of the resulting information asymmetry. However with NADAC, the information asymmetry is sharply narrowed. As a result, NADAC-based ‘cost-plus’ reimbursements are likely to be lower (particularly for new generics, because NADAC tracks falling new generic costs much more quickly and accurately than AWP) than legacy AWP-based reimbursements

Though it’s unclear what role NADAC may have played in the recent lowering of guidance by WAG and RAD, it is clear that NADAC is here to stay: NADAC-based reimbursement will certainly become more pervasive going forward, and will almost certainly result in lower generic margins

  1. Bureau of Labor Statistics (BLS) Producer Price Index (PPI) series 4461101, deflated by all-urban CPI
  2. Bureau of Labor Statistics (BLS) Producer Price Index (PPI) series WPUSI07003, deflated by all-urban CPI
Print Friendly