The current console generation started nearly seven years ago -- exactly how has U.S. video game retail been dealing with such an extended cycle? Gamasutra analyst Matt Matthews takes a close look. In many ways, the decline of video game retail in the U.S. resembles the pause that appears to happen right before a new generation of video games is launched. Not only are overall revenues down according to the latest NPD Group market estimates, but the drivers of revenue -- software and hardware -- are selling at lower volumes despite lower prices. This last happened in 2005, when Microsoft planned to launch the Xbox 360 in November and the Nintendo Wii and Sony PlayStation 3 were not due for another 15 months. Of course, this year we have the Nintendo Wii U launching by November, and the new systems from Microsoft and Sony are not expected until next year. It's not as simple as just the launching of a new set of consoles, however. This generation has had a higher peak than the last, relative to its starting point, and it has gone on for a couple of years longer. Along with the higher peak in sales, there has been a corresponding fall. The figure below shows the last generation's retail revenue growth curve. It's useful to compare the situation now to that in 2005. Seven years ago July was the second worst month in terms of revenue, at a mere $133 million per week. This year the revenue in July came in around $137 million per week, and it's the third worst in 2012. Back in 2005 the Game Boy Advance, Nintendo's best-selling handheld at the time, had hardware sales that were down nearly 40 percent from the year prior, despite a price cut from $100 to $80 less than a year earlier. Nintendo's other handheld, the Nintendo DS, was struggling through its first year with sales of fewer than 100,000 units for all of July. Sales of the Nintendo GameCube were even lower. The situation is not so dire for Nintendo this year, since Wii sales are hitting historical lows right before the launch of a successor. Still, Nintendo sold about 370,000 hardware units across three platforms seven years ago and about 330,000 this July. The original Xbox was on a terminal trajectory in 2005, but still managed 130,000 units per month on its way out the door. Sales of the PlayStation 2, destined to be the leading console for another year, were basically twice the Xbox figure. At that time, current generation consoles accounted for around 440,000 units and today their modern counterparts totaled around 410,000 in July 2012.

Accelerated contraction

With those kinds of figures being so close to the figures from 2005, it is tempting to say that the industry didn't grow at all in the past seven years -- but of course we know that's not true. Today's video game industry isn't simply a retail market. Seven years ago, the Xbox Live Arcade service was less than a year old and the Xbox Live Marketplace hadn't yet launched. Valve Software's Steam service was less than two years old. Apple's iOS App Store wouldn't see a release until July 2008. What I will say is that the traditional video game business model, centered around retail, has been fundamentally altered. Fewer systems are being sold, but at significantly higher prices. Every system on the market today has local storage, multimedia functionality, and the facility to sell products or connect with paid services, in part justifying the higher price of entry. As a result, the economic power of these devices cannot be measured by retail figures alone. Regrettably, the most public and detailed data we have is about the retail market, so that is where I'll focus today. Later this week I have some more information to share about the digital market, so stay tuned. Even though the industry is surely experiencing some pressure from a longer than usual generation of consoles, there are still some shorter-term issues that are worth discussing. In particular, we are now over a year into a period of exceptionally poor retail sales. I know that my columns can come across as at lot of doom, but the problem here isn't just contraction but an acceleration of the contraction. As I've said many times before, the U.S. retail market for video games began to shrink around March 2009. However, if you look closely at the last 19 months of sales results, in particular the year-over-year (YOY) changes, you'll see an interesting sequence of results. The graph below shows these YOY rates of change. The last really good month, at least by this measure, was April 2011 when Mortal Kombat (2011) and Portal 2 both launched. And from September through November, the industry's sweet spot, sales were roughly flat ($5.15 billion in 2010 versus $5.12 billion in 2011). But the accelerated contraction I was talking about is pretty evident: Outside of those three flat months, every other month in the past 13 has had a YOY decline of 20 percent or more. I don't know precisely what, but something changed last summer, and it wasn't just in one segment of the market. As of July 2011, the Xbox 360 stopped having consistent year-over-year hardware sales increases and the Nintendo DS had its last non-holiday month with sales of 300,000 or more units. Hardware sales for the PS3 also stumbled, but this was mitigated somewhat by the $50 cut to its price in August of last year. Of course, a price cut also means lower revenue. Software unit sales across all platforms took a dive last summer and have experienced regular downward pressure on average prices. The result: 20-40 percent contraction in 8 of the last 12 months of software sales. Even big months like May 2012, when Max Payne 3 and Diablo III both launched, look terrible in that figure above. I had thought, prior to the July results, that we would see a turnaround in the market, that somehow U.S. consumers had collectively reset their spending habits on video games to a lower level, and once a full year had elapsed we would see the market stabilize. Unfortunately, that didn't happen and instead sales were down another 20 percent this July. As noted in previous months, part of the problem here is systemic: publishers are releasing fewer titles per month this year than they have in previous years. According to the NPD Group's Liam Callahan, there were 39 new releases (SKUs) in July 2011 and only 22 this year, across consoles, portables, and PC. If the August retail results also show a 20 percent or greater decline, then we will have had three months this year with less than $600 million in sales. That hasn't happened since 2004, eight years ago. And it isn't just my own pessimism here. Just read what the NPD Group's own analyst, Anita Frazier, had to say: "Based on year to date sales, and taking into account the release slate for the back five months of the year as well as the anticipated launch of the Wii U, annual sales for the new physical channel should come in around $14.5 billion for the year." That figure, the $14.5 billion for the year, represents a 15 percent year-over-year decline for the retail industry during calendar 2012 and a 5 percent decline over the last five months of the year. Even the NPD Group itself is expecting a tough end to the year after taking into account the software releases that have been announced, like Halo 4 and Call of Duty: Black Ops II, and the launch of a new platform. Prior to the July results, Michael Pachter of Wedbush Securities was expecting a similar result for software through the end of the year, saying that "the introduction of compelling console software later this year will help to mitigate monthly sales declines, with the result that a modest rebound in sales (to somewhere between down single digits and flat) will likely occur in September through December." If this year comes in at around $14.5 billion for the total take at retail, then the generational curve -- similar to the one I showed above for the last generation -- will look like the figure below.

August turnaround?

Last generation, the total size of the retail industry grew almost 60 percent from the launch of the PS2 to the launch of the Xbox 360. This generation, if that $14.5 billion holds, then the industry will have grown nearly 40 percent from the launch of the Xbox 360 to the launch of the Wii U. It would be ideal to include all revenues -- retail and digital -- in these curves, but that kind of data simply doesn't exist, not in public and not in reliable form. Looking ahead, I'm particularly keen to see the August results, which we will have in about a month. The introduction of the Nintendo 3DS XL and the launch of New Super Mario Bros. 2 should bring in some extra revenue, along with Darksiders II and Sleeping Dogs on the HD consoles. Madden NFL 13 won't arrive until 28 August, and therefore will be counted as part of September's results. We won't know for months whether the retail game industry can really break its streak of 20 percent declines, but if August is indeed a turnaround and the packed release slate helps buoy sales during the last four months of the year, it's at least a sign that consumer do still have some appetite for retail products -- just not outside of the holiday season. Later this week, I'll have an update on July's software sales chart and some interesting information on digital sales as compared to retail.