“Our favourite holding period is forever” – Warren Buffet
The next bubble?
Shares in the technology sector have soared in the first half of the year, up some 24%. Comparisons with the 2000 tech bubble are multiplying amongst nervous commentators. The technology sector has been one of our preferred equity sectors for the majority of this economic cycle. Below, we explore whether the time has come to turn more cautious.
First, let us put the recent market expansion into perspective. The sector now comprises roughly 16% of the total MSCI World market capitalisation. While the market cap weight of technology stocks has increased in recent years, the move has been gradual compared to the sharp spike seen during the dotcom bubble. More importantly, the increase in the sector’s market cap weight has corresponded with its rising share of total earnings. This is a far cry from the late 1990s, when valuations rallied and the sector’s market cap weight significantly exceeded its earnings share. It took nearly eight years after the dotcom crash for the chasm between earnings and market cap share of the technology sector to close, underscoring the enormity of the distortions caused by the internet bubble.
Such distortions make finding a reasonable frame of reference for relative valuations complicated. The structural de-rating of the sector following the bursting of the dot-com bubble took relative valuations of technology stocks to a pessimistic extreme in 2013/2014. Consequently, today’s relative valuations look elevated versus 3 or 4 years ago, yet multiples have merely rebounded from very depressed levels. The broad technology sector currently trades at a modest 5% premium to the broad equity market on a forward P/E ratio basis. Such a premium does not seem particularly rich for a sector with a strong balance sheet, superior return on equity and above market earnings growth prospects.
Viable business models…
The lack of viable business models was a key feature of the dotcom bubble. It took significant time until advertising revenues could be harvested as a source of funding, leaving many businesses high and dry while waiting. Figure 3 plots return on equity as a measure of profitability over time. The graph confirms that, except for the few large companies that are able to cross-subsidise their R&D capital expenditure, profitability in the IT sector was well below the overall market average during the dot- com boom. Similar to the gap between market cap weight and share of total earnings, the gap only closed years later. The good news is that, following the financial crisis, the situation has reversed and broad IT profitability is well above the market average.
Should we nonetheless be worried that, in addition to an expected strong correlation with growth and momentum recently, the sector also showed a co-movement with the ‘bond-proxy’ consumer staples and utilities sectors, and minimum volatility and Treasuries? However, history tells us that this type of co-movement is nothing new. Ultimately, the biggest threat to the relative performance of the technology sector stems not from earnings or valuations, but from a more pronounced rotation into deep cyclical plays. Such a move would likely require both a sustained rise in bond yields and commodity prices, so bolstering investors’ confidence in the prospects of other cyclical sectors such as energy, materials and finance. For our part, we do believe bond yields will work their way higher in the second half of the year. However, commodity prices face headwinds due to buoyant global supply and a slowdown in China’s heavy industries – the latter being driven by a cooling of the country’s property market. A backdrop of gradually rising bond yields and flat-to-down commodity prices is likely to remain supportive of the relative performance of technology stocks.
Information technology is cyclical enough to benefit from the current economic climate in the context of other deep-cyclical sectors like energy and materials remaining unattractive. We see further upside potential for the sector in the medium term. While some bumps of course can’t be ruled out, we continue to recommend an overweight stance in our US regional model.