Economy in Brief

This week, we assess Asia through a risk lens, given the recent flare-ups across the region. Japan–China tensions, following Japanese Prime Minister Takaichi’s recent remarks on Taiwan, have prompted some retaliatory measures from China, including travel advisories against visiting Japan. This raises the risk of a dent to Japan’s tourism income this year, given the significance of aggregate Chinese tourism expenditure (chart 1), although any such impact would come at a time when Japan is already grappling with the effects of overtourism. Beyond Japan–China relations, investors are also bracing for the Bank of Japan’s (BoJ) final scheduled policy decision of the year, where a rate hike is expected despite Takaichi’s pro-growth fiscal stance. This expectation reflects ongoing yen weakness (chart 2), alongside encouraging results from Japan’s Q3 Tankan survey, among other factors. While the anticipated hike is not, in itself, a risk, the associated increase in financing costs—coupled with the likelihood of higher government bond issuance to support growth objectives—adds to investor concerns.

Risks are also evident in China, particularly after the November data docket showed a continued deterioration in growth across multiple sectors (chart 3). While China’s 5% growth target for the year remains within reach, achieving it appears to hinge on a narrow set of growth drivers, notably trade. On this front, China has defied earlier investor concerns to record a substantial widening of its trade surplus this year, despite a worsening bilateral trade balance with the US (chart 4). However, while supportive of growth, this development risks drawing increased pushback from non-US trading partners that have been absorbing China’s excess capacity.

Elevated risks are also present elsewhere in Asia, particularly in Southeast Asia, where renewed military clashes between Thailand and Cambodia have once again pushed bilateral tensions to the fore (chart 5). Thailand additionally faces political uncertainty ahead of early elections, following Prime Minister Anutin’s dissolution of parliament last week. Lastly, we turn to semiconductor- and AI-related risks, which are especially pertinent to Asia given the dominance—and reliance—of Taiwan and South Korea in the sector. A key risk lies in the current extent of market optimism surrounding future AI-related earnings (chart 6), suggesting greater scope for disappointment than for further upside surprises.

Japan-China tensions Geopolitics will likely remain front and centre in Asia heading into the new year, with several flashpoints carrying the risk of broader regional spillovers. One such case is the recent flare-up in tensions between China and Japan, following new Japanese Prime Minister Takaichi’s parliamentary remarks on Taiwan. Since then, Chinese authorities have rolled out a series of measures — urging Chinese citizens to avoid travelling to or studying in Japan, reinstating an import ban on Japanese seafood, among others. While none of these steps are particularly substantial on their own, they underscore a rapidly deteriorating bilateral dynamic. The economic implications are not trivial. As shown in chart 1, spending by Chinese tourists has historically made up a significant share of Japan’s total inbound tourism receipts — roughly a quarter in recent quarters — with especially strong concentrations in certain regions such as Osaka. Although the absence of Mainland Chinese tourists has been keenly felt since tensions escalated, early indications suggest that increased arrivals from other countries have helped offset some of the shortfall. More recently, a series of close military encounters between China and Japan has further heightened concerns. The risk of miscalculation has grown, raising the possibility of an inadvertent escalation at a time when geopolitical sensitivities in the region are already elevated.

Initial claims for unemployment insurance rose 44,000 to 236,000 during the week of December 6, after declining to 192,000 during the week of November 29, revised from 191,000. The four-week moving average of initial claims was 216,750, an increase of 2,000 from the previous week unrevised average of 214,750, which was the lowest level since January 2025.

Federal government employees initially filed 643 (NSA) jobless insurance claims in the week ended November 29, after 1,126 filed in the week ended November 22, revised from 1,125.

Continuing claims for unemployment insurance posted a sharp decline of 99,000 to 1,838,000 in the week ending November 29, from 1,937,000 the prior week, revised from 1.939 million. The four-week moving average decreased by 27,000 to 1,918,000 in the week ended November 29, from 1945,250 the prior week, revised from 1.945 million. The insured unemployment rate stood at 1.2% in the week of November 29, a decrease from 1.3% in the prior week.

Federal government employees continuing benefits fell to 12,732 (NSA) in the week ended November 22, from 17,862 in the week ended November 15, revised from 17,851.

The insured unemployment rate varied greatly across individual states and territories. In the week ended November 22, the highest unemployment rates were in New Jersey (2.18%), Washington (2.23%), Massachusetts (1.95%), Connecticut (1.83%) and Rhode Island (1.82%). The lowest rates were in Florida (0.27%), South Dakota (0.34%), North Carolina (0.39%), Louisiana (0.42%), Alabama (0.44%), Virginia and Tennessee (both 0.46%), and Arkansas and Nebraska (both 0.48%). Rates in other notable states include California (1.74%), New York (1.63%), Pennsylvania (1.46%), Illinois (1.45%), and Texas (1.07%). These state data are not seasonally adjusted.

Data on weekly unemployment claims are from the Department of Labor itself, not the Bureau of Labor Statistics. They begin in 1967 and are contained in Haver’s WEEKLY database and summarized monthly in USECON. Data for individual states are in REGIONW back to December 1986.

Sweden’s inflation rate has developed into a very interesting oddity as headline inflation as measured by the harmonized index of consumer prices (HICP), that is used in Europe, has come to diverge sharply from the Swedish national rate of inflation measured by its national CPI. The HICP inflation rate is 2.2% over the last 12 months; Sweden’s national rate is up by only 0.2%. If we rank each of these two inflation rates on data back to 2001, the rank of the HICP is in its 72.2 percentile while the rank of the national CPI is in only its 19.4 percentile, the lower one-fifth of its historic range of national inflation rates. These two rankings occur over the same periods and so there's nothing about the periodicity that's responsible for this; there's simply very different performance over the recent span.

About measuring inflation The HICP was adopted in Europe once the monetary union was formed and the countries of Europe were not able to get together to decide how to treat housing costs, which are important element in the CPI. They had entered into CPIs in different countries in different ways at that point. In the United States, inflation statistics are often criticized because of the way the housing component fits into them. Housing has such a large weight in the U.S. inflation measure. U.S. housing costs consider the cost of running the house in terms of taxes and utility costs; they look at the rental equivalent values, and they also have an owner imputed rent for people who live in their own homes. Of course, owner equivalent rent is a made-up number; it's completely imputed. Because a lot of European mortgages are short-term and they reprice when the central bank changes its rate, it’s very important to be careful how interest rates are treated in the housing cost indexes in Europe. The central bank did not want to be in a position where it would be raising interest rates to fight inflation and causing inflation to be higher because they were raising the mortgage interest rates on homes and causing the cost of housing to go up. For this reason, the HICP measures do not include housing costs. As we can see in the Swedish details for this month, housing and fuel costs are down 3.1% year-over-year. If we look at the ranking of housing costs for year-over-year growth, we see that they rank in the lower 6.4 percentile and are one of the most important factors in driving the national CPI lower and, of course, that metric is absent from the HICP.

Inflation is always rife with controversy: All of this is simply just another way to point out that inflation, while we measure it and talk about it, it's a controversial subject. How we measure it does matter. However, for monetary policy purposes, if we construct a measure that is reasonable and if we follow it consistently over the years, it should give us good signals and a good way to measure inflation and engage monetary policy. Yet, very clearly, if there are two competing gauges, it could become a real problem to jump from one to the other because they track different circumstances.

The current state of Swedish inflation: Swedish inflation in November fell by 0.3% month-to-month after rising by 0.5% in October and falling by 0.1% in September. The diffusion, or breadth, of inflation acceleration was only 33.3% in November. It jumped up to 55.6% in October, but it had been lower in September at 33%. Looking at Swedish inflation over 12-month, 6-month, and 3-month periods, the HICP goes from 2.2% over 12 months to an annualized pace of 5.7% over 6 months and then back down to 0.5% over 3 months. The national CPI has the same sort of topography but different growth rates with the CPI up by only 0.2% over 12 months, accelerating to 1.5% over 6 months and then declining at a -0.9% annual rate over 3 months. Inflation diffusion over 3 months in the national index is only 11.1%, over 6 months it jumps up to 66.7%, and over 12 months it's at 44.4% which is below 50% marking inflation as having more deceleration than acceleration.

Controlled inflation: On balance, Swedish inflation appears to be under control using either method. The ranking for inflation by category shows high inflation for food & nonalcoholic beverages, education, other goods & services, and medical care. However, inflation is below its median pace and bears low rankings for clothing & footwear, housing, fuel & lighting, recreation & culture, and transportation. Inflation is always somewhat mixed across sectors. That’s another thing that makes it hard to pin down exactly what inflation is, because people who consume different baskets of goods are going to experience different inflation rates. However, Sweden appears to have its inflation situation under control whichever metric you look at. The chart makes it clear that after its spike during COVID, inflation has come back down to a reasonable long-term pace that appears to be close to something that the central bank would find acceptable. The recent rise in HICP inflation seems to be unwinding and year-over-pace is down to a moderate 2.2%.