NPD: Behind the Numbers, May 2009

[Gamasutra's industry-leading analysis of May 2009's U.S. console hardware and game sales gives some perspective on the startling industry slump, also discussing Wii hardware slowdown and PSP Go pricing.]

It's official: the videogame industry is feeling the pain of the global recession, just like almost every other business in the world. Overall U.S. market was down 23% in the month of May compared to the previous year, according to the retail tracker, the NPD Group.

Hardware unit sales are down on every system except the Xbox 360 and the Nintendo DS/DSi, and coupled with the Xbox 360 price cut from September 2008, hardware revenue is down 6% so far this year.

Software unit sales and corresponding prices are down, and as a result software revenue is off by 8% from the same time in 2009. Even accessories are down 5% year-on-year.

Below we'll cover the damage in more detail but also point out some solid reasons to temper the gloom that currently hangs over the industry.

Industry Figures at a Glance

The latest NPD Group data shows that the U.S. videogame industry was down in every major category in May 2009. The top-line figures are shown in the following table:

Stronger sales in January and February have provided some buffer for the year-to-date (YTD) figures.

Many analysts, such as Michael Pachter of Wedbush Morgan Securities and Jesse Divnich of EEDAR, still expect middle single-digit annual growth by the end of the year, but such growth will require strong sales in the second half of the year, particularly through the holiday months.

Three Facts for Perspective

In a moment we'll show how the industry peaked between January and February of this year, but for the moment, let's get some perspective on the industry's slump. Here are three reasons that could have contributed to lower year-on-year sales in May 2009, irrespective of the current economic conditions.

This isn't to dismiss the economy as a factor in slowing videogame industry revenue. Quite the contrary, the economy may indeed be a key reason that sales are down.

However, it is worth noting that even with the deck stacked against the industry as listed above, industry revenue should still show modest growth by the end of the year.

The Industry Peaked, Briefly

We can measure the health of the videogame industry by looking at a 12-month simple moving average of weekly revenue.

For example, the $415.9 million per week given February 2009 in the figure below is computed by adding up the revenue for March 2008 through February 2009 and then dividing by the number of weeks over which that revenue was accrued.

Viewing the data in this way smooths out seasonal adjustments, since each 12-month period contains exactly one November and December, the months that so strongly skew single-month comparisons.

With one single month exception (September 2008), this simple moving average increased each month for over three years, from $199.3 million per week in February 2006 to $415.9 million per week in February 2009.

At that peak, the industry was at his greatest point in history: a 12-month period during which revenue totalled over $21.6 billion. For comparison, the official total for calendar 2008 was $21.3 billion.

(Our reliable industry revenue data does not extend into 2004, preventing us from following this trend back further.)

Then in March 2009, the rate dropped to $410.3 million per week. It dropped again in April and then again in May, ending at $401.3 million per week.

This kind of drop is unprecedented in the data we have available, and it points to a sudden and dramatic shift in the retail-based game industry's fortunes. All previous drops were single-month drops, after which the market revenue had begun climbing again.

In order to match the industry's total revenue for 2008, this moving average will have to creep back up to the $410 million per week level by December 2009. That will require several months of very modest year-on-year growth later in the year or a handful of months with extraordinary growth.

In another couple of months, we can return to this measure of the retail industry's health and see whether it has slowed its descent or begun to rise again.

(It's worth mentioning that there's significant and likely increasing digital-only revenue, including online game subscriptions and microtransactions, not encapsulated in NPD's monthly reports. This continues to be difficult and treacherous to estimate.)

Wii Pace Slows, Microsoft Grows

While many analysts expected Wii hardware sales to slow, it was still novel to see Wii hardware sales drop under 75,000 units per week. The Wii has sold slower in only two prior months.

The first was March 2007 at a rate of about 52,000 units per week, right before an increase in hardware production finally reached retailers.

The second was January 2008, when sales dropped to 68,500 systems per week after Holiday 2007 sales drained every available unit from retail and left relatively nothing for the first month of the year.

Is the drop in May 2009 significant? Yes, but not necessarily because there is anything wrong with Nintendo's business. The previous sales record-holder, the PlayStation 2, managed just under 71,000 per week during its third May on the market in 2003.

If indeed supply has finally surpassed demand, then the figure in May 2009 may well be the new normal sales level for the Nintendo Wii. It is lower than historical Wii sales levels, but it is also more than 25,000 units per week higher than its nearest competitor, the Xbox 360.

Suffice to say that both Sony and Microsoft would probably be thrilled to claim that their lead platform had sales as strong as the Wii did in May.

To put the sales in perspective, we have noted the three exceptionally slow months for the Wii on the figure below.

After the release of the NPD Group data on Thursday of last week, Microsoft commented that its Xbox 360 is the only console to experience growth in 2009. This statement is true, but requires some context.

For example, year-to-date (YTD) Wii sales are down at this point in 2009 and Xbox 360 sales are up, but this misses the crucial relative sizes of those two figures.

Microsoft has sold about 1.4 million Xbox 360 systems so far in 2009 while Nintendo has sold approximately 2.7 million Wii consoles -- quite an important fact that is overlooked when merely comparing relative growth or decline.

Furthermore, Xbox 360 had some of its weaker sales at the beginning of 2008. With the September 2008 price drop, sales increased substantially (as happened to the PlayStation 3 with its November 2007 price drop). Without further price drops, it is possible that the Xbox 360 will end the year with little or no growth over 2008.

Permanent Generational Rankings?

For precisely a year, the Nintendo Wii has had the largest installed hardware base of all the current generation systems. The Xbox 360 has been in second while the PlayStation 3 has been running at a distant third. As of May 2009, the installed bases for those systems stand at 20.2 million, 15.3 million and 7.75 million units, respectively.

If, indeed, the current systems are destined for a 10-year hardware cycle, these may very well be the rankings that will persist to the end.

Let's assume that the Xbox 360 is considered a living system through to November 2015, a decade after its launch, and the PlayStation 3 catches up with it by that time. Over that period, the PlayStation 3 will have to outsell the Xbox 360 by over 19,000 systems per week on average.

Outside of June 2008, when the Metal Gear Solid 4 launch helped the PS3 surpass the Xbox 360 by 37,000 per week for that lone month, the PS3 has only bested Microsoft's system by at most 9,800 systems per week in any given month.

Perhaps more strikingly, the PS3 has sold better than the Xbox 360 in only five of the 31 months that both systems have been on the market. And for the year so far in 2009, the Xbox 360 has outsold its HD rival by over 20,000 systems per week.

For Sony to reverse the roles entirely – behind by 20,000 per week to ahead by 19,000 systems per week – would be an extraordinary change of momentum, one that seems unlikely even with the most generous (yet realistic) price cut to the PlayStation 3 hardware.

Consumers Buying Inexpensive Software

According to data provided by Cowen and Company, console software revenue has been harder hit by the market downturn than has handheld software.

Software for consoles typically sells at a much higher average sale price (ASP) than does handheld software, and therefore any drop in console sales is felt more acutely than is a drop in handheld sales.

Ideally we would be able to compare three relevant figures for console and handheld software: unit sales, revenue, and ASP. However, we only have access to the revenue figures, and they are plenty revealing.

For convenience, we have graphed the relevant figures below.

When comparing the YTD 2008 figures with the YTD 2009 figures, the figure above shows that console software lost over $230 million. By comparison handheld software lost a more modest $33 million dollars.

In relative terms, that means that console software revenue dropped 8.5% while handheld software only lost 3.5%.

While we don't have unit sales information for each of these segments, Mr. Pachter of Wedbush Morgan Securities, did provide some insight into unit sales for software overall.

All the software purchased in the first five months of 2008 had an ASP for $39.89, while the software in the same period of 2009 had an ASP of only $38.01, a decline of $1.88 (or 4.7%). Mr. Pachter attributes this shift in part to consumers buying more catalog (older) titles and fewer new releases.

Consider also that unit sales for the year are down very little, as shown in this figure.

With a loss of only 3 million software units from 2008 to 2009, it is harder to argue that consumers are buying significantly fewer software titles overall.

In fact, a single big release on the scale of Super Smash Bros. Brawl or Grand Theft Auto IV might well have tipped the software market over from a loss to a gain for the year, at least in terms of unit sales if not revenue.

(On Friday of last week we provided an analysis of the top 20 software chart.)

Hardware Pricing and the PSP Go

First, a bit of keeping score. Last month we predicted that Sony was likely to announce a price cut for the PlayStation 3 hardware during E3 2009 in early June. We were wrong, and Sony has reportedly said that it felt that a consumer products show was the wrong setting in which to announce a price cut.

Some analysts currently expect the PS3 to receive a $100 price drop in time for the launch of Madden NFL 2010 in August 2009. Elsewhere Mr. Pachter of Wedbush Morgan expects a smaller cut, probably $50, and a bundle announcement, possibly at the Tokyo Game Show in late September.

We still believe that Sony will have difficulty reaching its stated PS3 sales goals for its current fiscal year without a $100 price drop.

Talk of price drops and bundling for other systems is also increasing. For example, the Wii with a different pack-in is a possibility for the holiday season, although the price might not necessarily need to drop for that strategy to be effective.

Moreover, Microsoft could be compelled to shift its pricing or bundling in response to a Sony price cut, especially since the effectiveness of its last price drop appears to be waning.

Finally, we conclude with a discussion of the PSP Go, Sony's latest update to its PlayStation Portable platform, due in October of this year.

The new system adopts slide-out controls for smaller size, eschews the UMD mechanism and replaces it with 16 gigabytes of flash memory, and retails for $250. We'll focus on possible retailer reaction and the pricing.

Sony's John Koller, director of hardware marketing, told Ben Kuchera of Ars Technica that they have “changed the model from a margin perspective from the 3000”, suggesting that retailers were possibly getting a bigger cut of the $250 retail price. However, we spoke with Mr. Pachter of Wedbush Morgan in email and he does not believe that “the higher pricing has anything to do with adding extra margin for retailers.”

We suggested, and Mr. Pachter agreed, that retailers will heavily bundle the PSP Go at the outset – adding cases, media, and other accessories – in order to increase margins on their own.

With PSP software revenue comprising only about 4% of all software revenue (that would be approximately $18 million in May 2009), most big-box retailers will accept the loss in UMD sales and stock the PSP Go prominently as an upper-end consumer electronics product.

(Anecdotally, we have observed that PSP shelf space has decreased dramatically at several retailers like Target, Wal-Mart, and Best Buy. A digital-distribution PSP Go would fit easily within those existing spaces.)

Mr. Pachter explained that the PSP Go would likely sell well to existing (hardcore, we presume) PSP owners who wish to trade up. As a result, we believe Sony has priced the system at launch to maximize its revenue from that population and that system bundles from retailers will likely drive the system price even higher as they attempt to do the same.

[As always, many thanks to the NPD Group for its monthly release of the videogame industry data. Additional credit is due to Michael Pachter, analyst for Wedbush Morgan Securities, and Doug Creutz of Cowen & Company for their industry analysis. Finally, many thanks to colleagues at Gamasutra and commenters on NeoGAF for many helpful discussions.]