Disclaimer that is, it aims to signal if we are heading towards a turning point in the cycle rather than if we have reached a turning point. The HLI-M leads the ISM manufacturing index by 6 months on average. Interpreting the indicators Given that the aim of the HLI-C is to identify turning points in the economic cycle, one should focus on the direction rather than the level of the indicator. The HLI-C is also subject to data revisions, and hence the level of the indicator can change from month to month. Despite the revisions in the level, the HLI-C does a very good job of reliably identifying turning points in real time. The HLI-M is a diffusion indicator that should be read in a similar way to a PMI with the 50% line separating periods of expansion and contraction. The HLI-M indicator is not subject to revisions. We publish the first estimate of the HLI-M for each month rather than the latest estimate. That is, for each monthly reading going back in time, we calculate the HLI-M only using data that was available when then indicator was published that month; we dont “improve” the estimate using subsequent information. This means that the HLI-M series is a real-time indicator over its full history, rather than just for the latest reading. We look at this in a bit more detail in the “Data revisions” section on page 13). Overview The HLI-Cycle leads the US ISM manufacturing index by 3 months on average The HLI-Momentum leads the US ISM manufacturing index by 6 months on average 3 MULTI ASSET GLOBAL 20 February 2018 What is in this publication? In the remainder of this document we provide more information on the indicators and how they can be used as part of an investment process. In particular: We provide sector and regional sub-indicators for the HLI-C; We compare the HLIs to the widely watched ISM manufacturing index; We look at how the HLIs can be used to inform asset allocation decisions; We look at the impact of data revisions; We list the indicator constituents and look in detail at the construction methodology. For more on the HLIs, please see The new HSBC Leading Indicators: Ahead of the pack, November 2013. MULTI ASSET GLOBAL 20 February 2018 4 The constituents of the HLI-C are selected because they have consistently led the ISM manufacturing index. Most of the indices therefore behave in a similar manner. Occasionally, however, there are divergences at a regional and a sector level that provide some useful additional insights into the business cycle. HLI-C Breakdown Region and sector breakdowns for the HLI-C Regional breakdown for the HLI-C 9294969810010210492949698100102104199519971999200120032005200720092011201320152017EuropeAsia-PacificAmericasEmerging MarketsDeveloped MarketsSource: HSBC, Thomson Reuters Datastream Sector breakdown for the HLI-C 949698100102104949698100102104199519971999200120032005200720092011201320152017 ConsumerIndustrialsTradeMoney that is, any indicator could potentially make it into our global aggregate, it just needs to have shown a consistent lead historically. This also means that the list of potential indicators that we look at is vast; in fact, we start with a database of nearly 6 million potential series. This in turn means that our two indicators contain a lot more components than most leading indicators, which we believe allows us to get a more reliable estimate of the cycle. We also focus only on economic data and do not include financial indicators, such as the stock market. Other leading indicators include financial series, but since our objective is ultimately to lead markets, we do not do this because the reasoning becomes circular: using financial data to lead financial data. Why do you publish both a cycle (HLI-C) and diffusion indicator (HLI-M)? The HLI-C and HLI-M provide slightly different information. The HLI-C is designed to track the cycle directly so it aims to identify turning points as early as possible. The HLI-C also has a broader range of constituents. In contrast, the HLI-M only includes the timeliest indicators. The HLI-M is designed to signal changes in the momentum in the cycle. For example, it aims to show if the economy might be slowing during a phase of expansion and we are thus heading towards a peak. Why do you focus on the global cycle? Does such a thing even exist? Although there are, of course, regional variations in economic performance, we would argue that there is also a global business cycle that affects most countries. To get an estimate of this cycle, we look at 37 economies for which we have fairly complete GDP data over a long period; combined, these economies account for over 83% of GDP. A single common cycle accounts for over 50% of the variation in these series. We consider this to be the global cycle, and its this were trying to lead. By how much do the indicators lead the business cycle? The HLI-C leads the ISM manufacturing index (which is itself m