Over the course of the last six months or so, our analysis of what lies behind the recent global slowdown has changed somewhat. A year ago, we were anticipating a sharp slowdown in the US, which would ripple round the rest of the world through three main channels: trade, financial markets and confidence. Economies that were exposed to the US via trade, or which had close links with the US through their financial markets, or where confidence was anyway fragile, would be hit hardest. But the US would be hit hardest of all, being the primary source of the shock. Monetary and fiscal policy would loosen to offset the shock, but the impact of that loosening would be delayed, so some of the shock at least would be unavoidable. Since then, the US has slowed down sharply – in fact, rather more sharply than we expected - with the events of 11 September causing the slowdown to get worse. But the rest of the world has been hit at least equally hard. As outlined in the assessment article in the last World Economic Prospects, it is no longer appropriate to characterise the global slowdown merely as a US shock rippling around the world: rather it is a sharp, synchronised global slowdown, with all regions affected by similar factors. This article assesses the contribution of the boom-bust cycle in the ICT sectors globally to the recent slowdown – and to the pace of growth that preceded it. First of all, some definitions. For the purposes of this article, the information and communication technology sector includes both manufacturing and services. On the manufacturing side, it includes the production of PCs, other IT hardware (servers, semi-conductors, disks etc), and the production of mobile and land-line telephones, faxes, modems etc (but not radios, televisions etc). On the services side, it includes IT services such as hardware and software consultancy and software supply and data processing; and telecoms services such as consultancy plus the transmission of sound, images and data. Of course, the production of ICT equipment and services does not exhaust the wider contribution that the sector can make to overall activity, for four reasons. First, spending on ICT need not equal gross output: net trade in ICT can account for any discrepancy (though at a global level this discrepancy must sum to zero). Second, ICT firms undertake spending on non-ICT products: ICT manufacture involves intermediate inputs from a range of different non-ICT sources (both goods and services), and ICT firms also invest in plant and machinery and equipment of other kinds. Third, ICT equities represent a significant proportion of the total value of equities, so a collapse in ICT equity prices will have a direct impact on aggregate wealth, reducing consumption. Fourth, ICT firms are significant employers, so a collapse in their profits will involve widespread lay-offs, reducing average incomes directly and indirectly, by increasing the pool of unemployed and putting downward pressure on earnings. And fifth, the investment in ICT may have beneficial spillover effects on productivity economy-wide. However, in this article we restrict ourselves to estimating the direct impact on growth from ICT production in the US, Japan, the Eurozone and the UK over the period 2000 to 2002