China – Heading into a more inflationary 2013-1401-14 15:42 Caijing
• We introduce our new CPI leading indicator, which predicts CPI out to six months
• Pork, residential and service-sector prices will be key drivers of the 2013-14 inflation cycle
• We forecast average CPI inflation of 4% in 2013 and 5% in 2014, higher than market consensus of 3.1% and 3.5%
• We expect PBoC to hike interest rates once in Q4-2013, by 25bps, followed by four more hikes in 2014
Inflation was not a problem in 2012. We outlined this view in late 2011, when growth was slowing and China's three-year inflation cycle meant that CPI inflation, after averaging 5.4% in 2011, was due for a quiet patch (On the Ground, 15 December, 2011, ‘China – A year of not worrying about inflation’). CPI inflation was at 2.5% y/y in December 2012 and averaged 2.6% for the year, proving our inflation call for 2012 largely correct.
We believe the inflation story for 2013-14 will be very different (see On the Ground, 17 September, 2012, ‘China – A sneak peek at 2013’). We estimate that CPI inflation will average 4% in 2013 and 5% in 2014, above the Bloomberg consensus estimates of 3.1% and 3.5%. Inflation, we believe, is likely to become a serious problem from Q4-2013 through H1-2014. As a result, we expect the People's Bank of China (PBoC) to hike the benchmark interest rate once in Q4-2013, by 25bps, followed by four more hikes in 2014. This would take the 1Y benchmark lending rate to 7.25% by the year-end from 6% today. This is our base case; if further interest rate reform happens (for instance, if the ceiling on the deposit rate is raised), this will have an impact on our forecasts; more on this below.
In today's report, we introduce our leading indicator for CPI, the CPILI (Figure 1). Our CPILI has a good six-month lead on official inflation and had a correlation of 83% with headline CPI inflation from January 2000 to December 2012. Our indicator significantly outperforms official M2 in leading inflation (the correlation between M2 growth and CPI inflation was only 25% from 2000-12). The CPILI suggests that CPI inflation has already bottomed and will break above 5% by Q4-2013. We believe this will trigger an explicit shift in monetary policy.
In this report, we also explore the outlook for key components of the CPI basket, including pork and residential and service-sector prices. In weighted growth terms, these three components account for about 50% of overall CPI inflation. They have already begun to rise, and this should only intensify as the economy continues to recover.
The ingredients of our CPILI
The CPILI is composed of five variables:
• Adjusted CNY loan growth
• Money demand
• Electricity growth
• Industrial production (IP) growth
• The Shanghai stock exchange A-share index
In calculating the CPILI, we use the three-month moving average (3mma) of the y/y growth rates for these five variables (except for money demand, for which we use a 9mma). We have selected the electricity, IP and A-share index data because they track activity as well as forward-looking sentiment towards the economic cycle, and thus the demand-inflation cycle. The other two components require a little explanation.
The adjusted Chinese yuan (CNY) loan growth in our equation is different from the official number released by the PBoC. At end-2002, CNY loans made up 92% of total social financing (TSF, which measures all forms of credit extension, including CNY/FX loans, entrusted/trust loans, bank acceptance bills and bond/equity issuances). This share had fallen to about 28% by end-2012. Significantly increased usage of other financing channels has had a sizeable impact on CNY loan growth. To remove this effect, we have assumed that each month after 2003, a constant share of 92% of TSF was made up of CNY loans, to calculate a new series of outstanding CNY loans and their y/y growth.
There has been a significant divergence recently between growth in our adjusted CNY loan measure and the official number. As Figure 2 shows, our adjusted CNY loan growth has picked up markedly since July 2012, while the official loan growth rate has weakened modestly. Anaemic official loan growth seems inconsistent with both the re-acceleration in monetary supply growth (official M2) and the fact that non-food inflation has been edging up since September. In addition, we prefer to use our adjusted loan growth figure to calculate our CPI leading indicator, as its 6-month lead series has a much higher correlation with non-food CPI inflation – 36% between 2002 and 2012, compared with a correlation of only 7% between official loan growth (6-month lead) and non-food inflation over the same period.
We define money demand as the ratio between the 7-day repo rate and the product of the average of demand and 1Y deposit rates and the required reserve ratio (RRR). The 7-day repo rate is a market price for money – the higher it is, the stronger the demand for money in relation to money supply. In order to filter out the effects of money supply on the 7-day rate, we use the ratio between the 7-day rate and other rates. (Changes in deposit rates alter the cost of money supply, and changes in the RRR affect the amount of money available in the market.) This ratio provides a way to track changes in demand for money, and hence demand for goods and services. Figure 3 shows the relationship between money demand strength (6-month lead) and CPI inflation. The recent uptick in money demand suggests that underlying inflationary pressures are on the rise.
What the CPILI says about the 2013-14 inflation outlook
The CPILI comprises two sub-indices: one leading indicator for food inflation (the FLI) and another for non-food inflation (the NFLI). The two sub-indices have some differences in construction, designed to allow them to better track the development of these two forms of inflation.
Figure 4 shows our FLI and headline CPI food inflation. The dashed lines represent our forecasts for the variables' trajectories in 2013-14. Our FLI leads headline CPI food inflation by six months, so we know that the headline number will pick up strongly throughout H1-2013. The correlation between our FLI and official food inflation was 88% between 2002 and 2012, significantly outperforming official M2 growth, which had only a 19% correlation with official CPI food inflation over the same period.
For 2013-14, we expect official CPI food inflation to rebound from 4.2% y/y in December 2012 to a peak of about 12% y/y in H1-2014, before coming off to 4% y/y by end-2014 as a result of policy tightening starting from Q4-2013. We estimate that CPI food inflation will average about 7.5% in 2013 and 9.1% in 2014. On a weighted basis, this means that food will be responsible for about 2.4ppt of the 4% CPI inflation we forecast for 2013, and 2.9ppt of the 5% we forecast for 2014.
For non-food inflation, as indicated by our NFLI (see Figure 5), we expect average rates of 2.3% in 2013 and 3.1% in 2014, up from 1.6% in 2012. On a weighted basis, this means that non-food items will contribute about 1.6ppt of the 4% CPI inflation we expect in 2013, and 2.1ppt of the 5% expected in 2014. (The correlation between the NFLI and official non-food CPI inflation from 2002 to 2012 was 55%, compared with a 7% correlation between official loan growth and non-food CPI inflation over the same period.)
Our CPILI is calculated as the weighted average of the FLI and NFLI, which have respective weightings of 70% and 30%, in line with their relative contributions to overall CPI inflation in weighted growth terms. (As Figure 6 shows, food contributes about 70% of total CPI inflation, and non-food items account for 30%, of which the residential sub-component accounts for about half.) The correlation between our composite CPILI and headline CPI inflation was 83% from 2000 to 2012, as shown in Figure 1. Over the same period, the correlation between M2 and CPI inflation was only 25%.
As Figure 1 shows, we expect CPI inflation to be on an upward trend in 2013 and H1-2014, with the headline figure breaking above 5% in Q4-2013. Inflation pressures should gradually ease in H2-2014 as a result of policy tightening starting from late 2013. We estimate 4% average CPI inflation in 2013 and 5% in 2014. As a result, the authorities are likely to shift from a prudent monetary policy stance to outright tightening from Q4-2013. We forecast one interest rate hike of 25bps by the PBoC in Q4-2013, followed by another four in 2014.
We now look at some of the key components of CPI. We focus here on pork, residential and service prices.
The pig problem will return
Pork inflation looks set to become a problem again in 2013-14. The two key indicators here are the pork-to-corn price ratio and the total pig stock. The former measures the profitability of pig farming, while the latter is a good tracker of supply over the following year. If the pork-to-corn ratio is low, hogs are sold or slaughtered as farmers exit the business; this leads to a big upswing in pork prices after about a year due to a supply shortage.
Figure 7 shows the y/y change in the pork-to-corn ratio and pork inflation (the dashed lines are our forecasts for 2013-14). We calculated the pork-to-corn ratio using the current pork price and the 6mma of the corn price. The current rate of growth in the pork-to-corn ratio, at about -3% y/y, is too low from a historical standpoint. As the ratio normalises in 2013-14, pork inflation should rise.
Slowing pork supply growth will also add to price pressures. As we show in Figure 8, pork supply growth is set to slow in 2013, as indicated by the pig stock in 2012 (official pork supply data for 2012 has yet be released, so the end date for pork supply in the chart is 2011, not 2012). The negative relationship between growth in pork supply and in pork prices seen since 2002 suggests strong reasons to expect a reflationary pork cycle in 2013-14.
As a result, we estimate that pork prices will rise 16% in 2013 and 30% in 2014, after falling by 3.5% in 2012. Given that pork has a 2.9% weighting in the CPI basket, it is likely to add 0.45ppt to headline inflation in 2013 and 0.87ppt in 2014, after subtracting 0.1ppt in 2012.
Pay no less attention to other foods
Pork will not be the only source of food-price pressure in 2013-14. It has accounted for less than 30% of food inflation over the past decade. Other likely candidates for price increases include:
• Beef, whose price has been picking up strongly, as Figure 9 shows.
• Edible oils and fish as urban middle-class consumption growth picks up again
• Rice, whose price will continue to rise as part of official government policy
Overall, on a weighted basis, we expect food inflation (excluding pork) to be about 2% for 2013-14.
Residential and service prices underpin our inflation call
Residential and service prices are already picking up. Residential prices within the CPI include building and decoration materials, public and private housing rents, water, electricity and fuel. (We believe that rental prices here have two significant problems – they are under-weighted in the residential basket, and they are based on mortgage costs rather than actual rents. We therefore suspect that the official figures underestimate rent and residential price inflation.) As Figure 12 shows, residential prices are already bottoming out, and they should soon rise.
Slower growth in labour supply, combined with continued strong growth in demand for services (as reflected in the consistently above-55 reading in the non-manufacturing PMI) and moderate productivity growth, will continue to push up prices. Figure 13 shows a few examples of rising service prices, including nursery fees, education, car maintenance costs and personal service charges. We expect overall non-food CPI inflation to average 2.3% in 2013 and 3.1% in 2014, up from 1.6% in 2012.
Inflation is returning to China in 2013-14. Although most of the pressure will come from food, service prices will also lend a hand. By mid-2013, this issue will be moving up policy maker's list of things to worry about. We forecast that CPI inflation will average 4% in 2013 and 5% in 2014, with the headline figure breaking above 5% in Q4-2013; it may stay there for most of H1-2014. Given the likelihood that this bout of inflation will not be as serious as the 2007-08 and 2011 cycles, we believe the authorities should be able to move gradually to raise overall borrowing costs. We also hope that the central bank will be given more leeway to set interest rates, and that the government reform agenda will not be sidetracked by inflation-fighting.
We think the PBoC will hike benchmark interest rates once in Q4-2013, by 25bps, followed by four more hikes in 2014. This should take the 1Y benchmark lending rate from 6% at end-2012 to 7.25% by end-2014. A further loosening of interest rate controls may mean that rates can go higher without benchmark hikes. But at the end of the day, corporates will still need to prepare for higher lending costs.
Editors’ Picks »
- 1China Central Bank Withdraws Liquidity for Second Time in a Week
- 2Highlights: Premier Li's Government Work Report
- 3Chinese Regulators Vow Support for Internet Finance Products
- 4Last 3 Suspects Caught in Kunming Manhunt
- 5Jingdong Seeks Cornerstone Investors Ahead of IPO
- 6Japan’s Fast Retailing Surges in HK Debut
- 7HSBC, China Official PMI Drop for Third Consecutive Month in Feb.
- 8Yuan Among Most-Used Currencies
- 9CNOOC Licensed to Seek Arctic Oil
- 10Geely Buys British E-Car Startup