Japan bearing orders: autos great, machinery weak

by SR on February 9, 2012

On 9 February the ESRI released Japan machinery orders for December 2011, a closely watched indicator of purchasing sentiment. As usual I’m going to ignore the headline number, which covers the whole machinery sector, and focus on the orders, sales and book-to-bill ratio of the bearings sub-sector.

Why should you care? Because bearings are used in all machinery and thus despite being a small sector bearings are a broad indicator; because the orders have a 3- to 6-month lead time and so are forward-looking; because Japan’s bearing suppliers are dominant in a number of areas and so reflect global demand. (Related companies: JTEKT; NSK; NTN; Minebea; Nachi-Fujikoshi; SKF; Timken; Komatsu; Hitachi Construction Machinery; Sany Heavy.)

  • Big differences. The overall figures looked decent, but there was significant variation. Auto-related orders were abnormally strong, while machinery orders were well below the seasonal trend. No big surprise there: Toyota continues to rebuild inventory and this drives demand for auto bearings.
  • Prefer tyres. In terms of sector allocation, this discrepancy in the figures strengthens my view that auto parts (see here for Denso’s impressive Q3 figures) and especially tyres in the shape of Bridgestone are a better bet than machinery-related names for the next couple of quarters.
  • Don’t bottom-fish. I would be reluctant to start picking up these stocks in anticipation of a spring 2012 recovery in machinery orders,  given that the signals are so weak (remember that forklift exports collapsed last month). It’s tempting to buy bombed-out cyclicals but that could be a short-lived and painful trade. The worst news is not yet out.
  • Next up: tyres and diggers. The next data point I would be watching for (after Japan tyre sales, which should be out in the next few days) is China excavator sales. That data is usually released round the middle of the month.

Those figures in brief

Total orders came to Y58.4bn, rising 3.3% year-on-year and up 0.5% month-on-month. Sales were flat month-on-month at Y59.7bn and rose 3% YoY. For the calendar year of 2011, orders increased by 4.1% YoY to Y682.9bn, while sales posted a larger gain of +7% YoY to Y689.3bn.

Japan_monthly_bearing_orders_pct_change_year-on-year_to_

In a previous post I talked about the book-to-bill ratio for the Japan bearings and how I keep an eye on this as an indicator of the health of the global industry market. The “why” is that pretty much all machinery that is sophisticated enough to add value uses bearings and the Japanese bearing suppliers have high market shares in the auto and machinery markets. (Of course this largely reflects the competitive strength of the Japanese companies operating in these sectors that are the customers of the bearing suppliers.)

Japan_bearing_book-to-bill_ratio_(x)_to_

In that post I suggested that we would see a book-to-bill ratio of 0.99x for Japan bearings December 2011 after the 0.97x of November 2011. The actual figure was 0.98x so a tiny positive incremental change. The reason for the BB ratio coming a little lower than I had expected was not weaker orders, but stronger sales. As the figure below shows sales typically fall month-on-month in December, but in the December just passed, sales didn’t fall as much as usual. I’m guessing that was due to strength in products with a short lead-time, probably auto-related items.

Month-on-month_pct_change_in_Japan_bearing_sales_for_each_December_since_1988

For the orders data, however, the devil however is in the details. Autos had a great month, with orders increasing by 5% YoY and falling only 0.3% month-on-month. Look at the graphic below, showing the month-on-month change in December relative to November for every year from the late 1980s onward. As you can see, there is a pronounced seasonal trend for December to fall sharply. Since 1987 auto orders in December have never risen relative to the previous month of November; they always fall. December 2011 also fell MoM, but it posted the smallest sequential decline since 1987.

Month-on-month_pct_change_in_Japan_bearing_auto_orders_for_each_December_since_1988

Now look at orders over the same period for bearings used in general machinery. The seasonal trend for these products is far less distinct and orders show a greater range of movement. For about a third of the years since 1987, orders for machinery bearings have risen in December compared to the previous month. Nevertheless, December 2011 posted an emphatic sequential decline and was the fourth-worst month since 1987 in that sense.

Month-on-month_pct_change_in_Japan_bearing_machinery_orders_for_each_December_since_1988

The backlog fell again to Y159.3bn, -3.9% YoY and down 0.8% MoM. As the figure below shows, we’ve come a long way from the glory days of 2008 when the backlog briefly peeked over the Y220bn level in August and September before collapsing in 2009 to a low of Y152bn. We are still some way off that Y152bn trough point, but if machinery orders remain weak then we will start chewing through that backlog at a fair rate and we could hit it within a few months.

In theory things are not as bad for the bearings suppliers this time round, because the auto industry (by far the biggest contribution, accounting for 35% of total orders at least) is in better shape than it was during the global financial crisis. However, at that point China’s economy was still strong and despite a blip or two managed to throw off the weakness in the rest of the world. That was important for machinery (especially construction machinery) demand and the robust and I feel that the early recovery of China machinery demand was a major factor in cushioning the blow for the Japanese bearing suppliers. What makes me uneasy today is that the industrial (non-auto) bearings have far the higher margins, so compression in revenues of these products would have a disproportionately large impact on profitability.

Japan_bearing_order_backlog_(billion_yen)_to_December_2011

Incidentally, one of my readers commented drily that my talk of canaries suddenly “stopping singing” suggests a taste for the theatrical. She asks for a reality-check: what did the companies themselves say at the time those February 2011 orders were released, and what has happened since?

As previously mentioned, I wrote a short and gloomy note on those February 2011 orders when they came out in early April 2011. I also spent some time getting the views of the bearing companies. One of those companies was distinctly unhappy that I was making a bearish call on data that (they felt) arguably does not reflect the wider global economy. “We’re fine” they said, “we’re very busy”. The other two bearing suppliers I spoke to also played down the figures.

All three of those companies have since revised down their earnings guidance for FY3/12. They are, in no particular order, NSK 6471, NTN 6472 and JTEKT 6473, with NTN having the most exposure to the bearing market and JTEKT, for which steering systems are more important, having the least exposure. NSK is also a power-steering supplier. NTN has revised down its FY3/12 guidance twice and is now forecasting Y20bn in operating profit (-18.6% YoY), down from the forecast of Y35bn it put out in May 2011. Previously management had forecast 42.5% YoY growth in operating profit for FY3/12, to Y35bn yen.

Again, I want to make it clear that the bearing figures have to be looked at in the context of the other macro factors. There’s nothing magical about the data, it just happens to be a fairly forward-looking indicator for a very widely used product.

Previous post:

Next post: