Annual Report on the Japanese

Economy and Public Finance


- No Gains Without Reforms IV -

July 2004

Cabinet Office

Government of Japan

 [Toc]  [Prev]  [Next]  [Annual List

Section 1 Regional Disparities in Economic Recovery

1. Disparities in Production Recovery

State of regional economy
    Looking at this economic recovery phase, in step with the national economy having hit a trough and moved toward recovery in January 2002, most regions have bottomed out or rallied by mid-2002. Since then, some regions have come to a slight halt in recovery between the end of 2002 and the first half of 2003, but there was again a stronger trend toward improvement beginning in the second half of 2003. All regions except Hokkaido have achieved economic recovery or are firming up as of May 2004. According to the economic trends shown in the Economy Watchers Survey, business confidence has improved in all regions compared to the level in January 2002 at the beginning of the economic recovery. As of May 2004, the DI for current economic conditions is over 50 in the majority of regions. Since 50 is considered a benchmark for determining the direction of the economy, this indicates that the number of people who think the economy is getting better exceeds the number of people who think the economy is getting worse in most regions (Figure 2-1-1).(1) Moreover, according to the Economy Watchers Survey, the regional disparities in business confidence have somewhat diminished compared to previously.
Figure 2-1-1 Business Confidence Improves in All Regions
    Meanwhile, an examination of industrial production trends since January 2002 demonstrates that there are significant disparities from region to region (Figure 2-1-2). In regions with strong production growth, export- and digital electronics-related production in particular is giving a boost to overall regional production. Looking at production trends by industry on a regional basis, electronic parts and devices account for a relatively high proportion of production in Tohoku and Kyushu and production is increasing, reflecting the recent growth in domestic sales and exports of digital electronics (Figure 2-1-3). In Chubu, the production of electronic parts and devices is rising and the production of transport vehicles, which has seen solid exports, is also contributing to production growth in the region. The production of transport vehicles, geared primarily toward exports, is rising in the Chugoku region. In the Kinki and Kanto regions, the production of electronic parts, among other things, is growing, and the impact of export growth in response to the increased demand from China and other factors is also evident in the materials industry such as chemicals. Meanwhile, in Hokkaido where production recovery is delayed, machinery and other areas remain robust. However, food production, which is the mainstay of production, remains roughly flat. Ceramics, soil and stone are also declining, influenced by the contraction of public works, among other things, and production growth in the entire region is generally modest.
Figure 2-1-2 Regional Industrial Production
Figure 2-1-3 Dependence of Production on Exports and IT

Production disparities and export dependence
    Are the regional disparities in production especially salient in this recovery phase? The coefficient of variation (standard deviation divided by the mean) was calculated in order to determine the change in regional disparity in production of the year-on-year growth rate of industrial production in the past decade. According to this calculation, the regional disparities in production are slightly larger in this recovery phase compared to the past two recovery phases (Figure 2-1-4). Noting the region with the highest growth in production and the region with the lowest growth in production in this recovery phase, the growth in production from the trough of the latter is roughly the same as in the previous recovery phase, but the degree of recovery in production from the trough of the former is larger than in the previous recovery phase. Hence the production disparities in this recovery phase were brought about by the remarkable increase in production growth in some regions due to factors such as the steady sales of digital electronics and expanded exports, rather than the continued stagnation in production in some regions.
Figure 2-1-4 Production Disparities in This Economic Recovery Phase Widen
    There are two major underlying factors in these regional disparities in production recovery. The first factor is the differences in the situation of sites of the industries and export dependence in each region. In this recovery phase, robust domestic demand in electric machinery, transport machinery and other exports as well as production recovery in industries stimulated other industries. As a result, there is a trend in which regions with many sound industries recovered ahead of other regions. In particular, the degree to which production responds to export growth differs in each region, a point that can be confirmed in the Inter-regional Input-Output Table. Looking at how production is fueled by exports in each region, the Chubu and Chugoku regions, where many export industries are concentrated, are structured such that 16% and 13% of gross production is stimulated by exports, respectively. Meanwhile, this percentage remains low for Hokkaido and Okinawa, at 2% and 5%, respectively (Appended Table 2-1). Consequently, if economic recovery is export-led, then production will rise relatively significantly in regions such as the former and result in regional growth in regional production.(2)

Regional differences in the ripple effect
    The second factor causing regional disparities in production is presumably the significant differences among regions in their sensitivity to demand changes in other regions. What is meant in such case is that while the economy in one region recovers, it does not spread to some other regions and therefore causes regional disparities in business confidence. Looking at this trend in the Inter-regional Input-Output Table, it is evident that the sensitivity to demand is different for each region, depending on the extent to which it trades products with other regions (Figure 2-1-5). The characteristics of the regional economies based on their current situation are as follows.
Figure 2-1-5 Regional Characteristics: Production Stimulation Rate and Dependence on Demand
    First, regions where production is relatively strong at present such as Tohoku, Chubu and Chugoku have "trading-style" economies with a high proportion of products traded with other regions. These regions have a significant effect on stimulating production in other regions and are also greatly affected by demand in other regions. Currently, production is increasing in these regions, stimulated by the demand growth outside the region such as overseas. As a result, it appears that a favorable environment has been created whereby demand outside the region as well as investment has been stimulated, in turn spurring production in other regions.
    Second, Hokkaido, which is experiencing low production growth, is based on factors such as its geographic condition an "inward transfer-style" economy highly dependent on imports from other regions while the size of its exports to other regions is small. Therefore, Hokkaido has an economic structure which is insensitive to economic recovery in other regions and demand growth within Hokkaido easily leaks out in the form of increased imports from other regions. Nonetheless, in Okinawa, which has a similar "inward transfer-style" economic structure, business confidence has improved remarkably due to strong tourism. It is therefore necessary to keep in mind that the economic structure concerning production is not necessarily the only factor influencing actual economic trends.(3)
    Third, Kanto and Kyushu fall midway between the two abovementioned models in terms of the strength of production. They are "autonomous (self-contained style)" economy with a low proportion of products coming in from or going out to other regions. In other words, these regional economies have a considerable effect on stimulating production in their own regions and are highly dependent on demand in their own regions. However, since the production scale in the Kanto region is large, the ripple effect of production to other regions is significant even though it has a low proportion of products going out of the region.
Impact of reductions in public works
    Public works are being reduced in all regions with a view to realizing fiscal soundness at both the national and local levels. Looking at regional trends in public works based on changes in contract amounts for public works, most regions saw double-digit declines in 2003. An examination of the correlation between dependence on public works and business confidence of enterprises (Diffusion Index (DI) for judging the business condition of all industries, Short-term Economic Survey of Enterprises in Japan (Tankan)) shows that business confidence is weak in regions that are greatly dependent on public works (Figure 2-1-6).
Figure 2-1-6 Regional Impact of Reductions in Public Works

2. Employment Situation Still Severe in Some Aspects

Unemployment rates for some regions remain at high levels
    There are also regional disparities in employment, reflecting the economic structure of each region. The unemployment rate is relatively high in Hokkaido, Kinki, Kyushu and Okinawa, and relatively low in regions such as Tokai, Chugoku, Kita-Kanto and Koshin (Figure 2-1-7). Looking at changes in the unemployment rate of each region after January 2002, when the economy hit a trough, although the unemployment rate declined in all regions, regional disparities in the unemployment rate persist. Moreover, looking at long-term changes in the unemployment rate, the unemployment rate in each region has historically remained at high levels and the employment situation is still severe except in a few regions. As in the case of production, the coefficient of variation was calculated in order to see how the degree of the regional disparities in unemployment changed in a time series. According to the calculation, as a general trend, regional disparities are diminishing, but recently the variation coefficient has been rising after bottoming out in 1999 (Figure 2-1-8). The recent widening disparity has manifested itself as a trend in which the unemployment rate in regions with high unemployment rates increases more than in other regions.
Figure 2-1-7 Regional Unemployment Rate
Figure 2-1-8 Unemployment Rate Disparities Slightly Widen in Recent Years

Structural changes in regional employment
    An underlying factor in such regional changes in the labor market is that the regional employment structures are changing dramatically. Taking a somewhat long-term view of the fluctuations in the number of employers by industry in each region, first of all, employment increased in most industries, except agriculture and forestry in each region between 1987 and 1992, which inclusive of the bubble years (Figure 2-1-9). One of the characteristics of this period is that growth in employment in the tertiary industries and construction was comparatively high in the large metropolitan areas, reflecting the bubble economy.
Figure 2-1-9 Change in Number of Persons Engaged in Work by Industry
    Second, an examination of the changes from 1992 to 1997 after the collapse of the bubble economy reveals that even though employment in the tertiary industries and construction continued to grow in most regions, employment in manufacturing declined in most regions, reflecting expanded overseas production and other factors. Furthermore, influenced by the significant boost in public works, employment in construction in the provinces rose.
    Third, observing the changes between 1997 and 2002 shows that the number of employees fell in most regions. As for the type of industry, manufacturing continued to decline sharply in all regions and employment in construction also decreased in all regions due to factors such as the reduction in the size of public works. Thus growth in employment can be attributed to the tertiary industries in both the large metropolitan areas and the provinces.
    A close look at the breakdown of employment in the tertiary industries shows that between 1996 and 2001, employment increased in the service industries including medical care, welfare and business services and the information and communications technology (ICT) sector in most regions, and the wholesale and retail sector decreased in most regions (Figure 2-1-10). With respect to regional characteristics, the service industries such as medical care and welfare grew in the provinces as well as in the large metropolitan areas, while the ICT sector is experiencing outstanding growth in the large metropolitan areas.
Figure 2-1-10 Change in Number of Workers Engaged in Tertiary Industries
    Furthermore, considering the effects the public works reduction has on the regional employment situation, there is no strong correlation between the dependence on public works as of 2001 (public investment as a share of gross prefectural spending) and the changes in the number of construction workers, but there is a positive correlation between the fluctuations in public works in the past five years and the number of construction workers. In regions with a substantial decline in public works, the number of construction workers also fell sharply (Figure 2-1-11).
Figure 2-1-11 Relationship Between Changes in the Number of Construction Workers and Dependence on Public Works As Well As Changes in Public Works

Importance of labor migration between industries
    Structural changes in the employment situation as illustrated above have increased the need for labor migration between industries in the regions, although if it does not occur swiftly, it will result in higher unemployment, among other things. As such, looking at the industries that reemploy people leaving jobs in construction and manufacturing, which is experiencing a sharp downturn in employment,(4) more people in construction are reemployed in the same industry than manufacturing. This is presumably because of the nature of the construction industry is such that people often switch companies every time there is a new construction project, resulting in a trend wherein a high proportion of people are reemployed in the same industry. Aside from such intra-industry migration, many people transfer to the wholesale, retail, restaurant and pub industries, and service industries (Figure 2-1-12).
    Amid advancements in changes in the industrial structure, labor demand in each industry will change and presumably labor migration between industries will become essential. In the past, such changes in labor demand between industries were met by employing new graduates in growth industries sustained by an abundant young labor force. However, given that the size of the young labor force will dwindle in the future, it is expected that greater adjustments through labor migration between industries including of middle-aged and older workers will be required. To this end, a policy that promotes the smooth transition of labor between industries is necessary.
    This section has hitherto explained the background of the long-term upward trend in the regional unemployment rates. The next section will describe the regional disparities in the unemployment rate in more detail.
Figure 2-1-12 Labor Transition Rate by Industry Among Former Construction and Manufacturing Workers

 [Toc]  [Prev]  [Next]  [Annual List