40
USC Dornsife Institute for New Economic Thinking Working Paper No. 15-13 The Regional Spillover Effects of the Tohoku Earthquake Robert Dekle, Eunpyo Hong & Wei Xie February 20, 2015
USC Dornsife Institute for New Economic Thinking
Working Paper No. 15-13
The Regional Spillover Effects of the Tohoku
Earthquake
Robert Dekle, Eunpyo Hong & Wei Xie
February 20, 2015
The Regional Spillover E↵ects of the Tohoku
Earthquake
∗
Robert Dekle†, Eunpyo Hong‡, and Wei Xie§
Department of Economics, Kaprielian Hall, USC
February 20, 2015
Abstract
In this paper, we trace out how a decline in industrial productionin one region can be propagated throughout Japan. We model Kantoas the dominant region, which includes the Tokyo area, and then use themodel to analyze how a shock to industrial production in Tohoku–owing tothe earthquake–can be propagated throughout Japan. In our econometricmodel, regions and industries within regions are linked by an input-outputstructure and this input-output structure disciplines how the shocks arespatially propagated.
Keyword:Tohoku earthquake, regional spillovers, industrial production,dominant region, propagation of shocksJEL Classification: R11 R15
∗We thank the helpful comments from Professors Etsuro Shioji and other participants fromthe March 2013 Gakushuin Workshop on the Economic Recovery from the Great East JapanEarthquake. We deeply thank the financial support from the Japan Foundation Center forGlobal Partnership grant, without which this project would not have been possible.
†[email protected]‡[email protected]§[email protected]
1
1 Introduction
March 11, 2011, a devastating earthquake and tsunami hit the Tohoku andNorthern Kanto regions of Japan. The damage was mostly concentrated in theIwate, Miyagi, and Fukushima prefectures. In particular, all three prefectureswere swept by the tsunami, with much of the immediate damage caused by thetsunami. In the areas impacted by the tsunami, industrial production declinedby over 95 percent between March and July of 2011.
Nearly 23000 people were killed (or missing) in these prefectures; and inthe days after the earthquake, about 125,000 people (or 2 percent of the threeprefectures’ populations) evacuated. Destruction to the capital stock was esti-mated to be about $180 billion, or 10 percent of the total capital stock in thethree prefectures.
The overall weight of Iwate, Miyagi, and Fukushima in Japan is small, com-prising about 4 percent of both the Japanese population and GDP in 2010.Still the immediate impact of the earthquake and tsunami on Japanese aggre-gate production was huge, with the negative e↵ect on aggregate GDP lingeringon for a year or more. This is because these three prefectures were major pro-ducers of electronics and other intermediate parts used for production in otherJapanese regions (and even the world), and the stoppage in production of theseintermediate parts meant that production of the final goods in the electronics,automotive, and other industries were stalled all over Japan. For example, To-hoku accounted for 42 percent of the micro-semiconductors and 40 percent ofthe flat screen filters used in the Japanese production of automobiles and cellphones.
The importance of this collapse in Tohoku intermediate input production canbe seen in how Japan’s aggregate GDP declined in the immediate aftermath ofthe earthquake. Compared to the previous quarter (before the earthquake),Japanese aggregate GDP declined by 1.9 percent in the first quarter of 2011.1
The declines in aggregate consumption and inventories contributed 0.9 and 0.6percent to the overall GDP decline, respectively.2 Inventories dropped sharply,as firms nationwide dug into their inventories to supply the intermediate parts–disrupted by the earthquake–necessary for production.
In subsequent quarters, while consumption recovered, inventories continuedtheir depletion. Between the last quarter of 2010 and the third quarter of 2012,aggregate GDP grew by 0.5 percent. Aggregate consumption contributed 1.2
1It is important, however, to keep the magnitude of the impact of the Tohoku earthquakein perspective. In fact, the negative impact of the global financial crisis in late 2008 onoverall Japanese GDP was far larger than the negative impact of the Tohoku earthquake.Moreover, how the 2008 global financial crisis caused the Japanese recession at that time isvastly di↵erent from how the Tohoku earthquake caused the latest Japanese recession.
While the recession after the 2008 financial crisis was caused by a decline in Japanese invest-ment and an exogenous fall in exports, owing to a collapse in foreign demand, the recessionpost-earthquake was related to the inability of Japan to produce inputs to production, suchas intermediate products and energy, which led to a drawdown in inventories, a decline in theability to supply exports, and the increased imports of raw materials.
2Let GDP=C+I. Then in an accounting sense, the contribution of variable C to the growthin GDP is approximately (C/GDP)*�C/C.
2
percent to this growth, but the depletion of inventories and the decline in netexports contributed to dragging down GDP by 0.6 percent and 1.8 percentbetween the last quarter of 2010 and the third quarter of 2012.3
In this paper, we trace out how a decline in industrial production in oneregion can be propagated throughout Japan. We take Kanto (Tokyo) as thedominant region and explain how a shock to industrial production in Tohoku–owing to the earthquake–can be propagated throughout Japan. Kanto (Tokyo)was chosen as the ”dominant” region, since we estimate our model using datafrom 1998, and during most of this period, more and large shocks have emanatedfrom Kanto than from other regions.4Our maintained a priori assumption is thatduring the period including the Tohoku earthquake, Kanto can be treated as adominant region. This allows us to consistently estimate separate conditionalerror correction models for di↵erent regions of Japan, which we then combinewith a model for Tohoku to solve for a full set of spatio-temporal impulse re-sponse functions. Conditional impulse response analysis traces out the e↵ects ofshocks over time. However, with a spatial dimension, dependence is both spa-tial and temporal. In our impulse responses using our econometrically estimatedmodel, we trace out the e↵ects from a shock to Tohoku.
In our econometric model, regions and industries within regions are linked byan input-output structure and this input-output structure disciplines how theshocks are spatially propagated. Our emphasis on the input-output structurein the propagation of shocks after the Tohoku earthquake is motivated by thefact that much of the immediate impact of the Tohoku earthquake on otherregions was driven by the decline in intermediate inputs produced in Tohoku.The shocks to Tohoku are propagated spatially to other regions. The otherregions in turn impact other regions with a delay. We also allow these laggede↵ects to echo back to Tohoku.
This is not the first paper to trace out the e↵ects of the earthquakes andother natural disasters on Japanese output and industrial production. Davisand Weinstein (2002), Okazaki, Ito, and Imaizumi (2002), and Tokui, Kawasaki,and Miyagawa (2013) examine how the distribution of economic activity withinJapan are impacted by natural disasters. Uchida, Miyakawa, Hosono, Ono, andUesugi (2013) examine how shocks arising from earthquakes, when interactedwith fianncing constraints, can lower firm-level industrial production.
3During this longer period, net exports declined because of the fall in total exports andthe increase in total imports. The decline in total exports contributed to dragging down GDPgrowth by 0.6 percent and the rise in total imports contributed to dragging down GDP growthby 1.2 percent. Much of the increase in imports was driven by the increase in natural gasand other fossil fuel imports. Energy imports increased, since Japan was faced with an energyshortage. The energy shortage was caused by a shutdown of almost all of the country’s nuclearpower plants, which normally provides 30 percent of Japan’s total energy.
4The assumption of choosing Tokyo as a dominant region seems more justifiable when weare estimating our model using data from 1998. In an earlier version of our paper, we choseTohoku as the ”dominant” region in our estimates, but our impulse responses were una↵ected.
3
2 The Impact of the 2011 Tohoku Earthquake
on Aggregate and Regional Industrial Produc-
tion.
GDP includes a sizable component of non-manufacturing production, includingthe production of services. To better isolate the impact of the disruption ofthe production of parts in Tohoku on Japanese manufacturing production, forthe remainder of the paper, we focus on the measure of industrial production,which mainly captures manufacturing production. Figure 1 depicts the patternin industrial production from the third quarter of 2008 to the third quarter of2012. We can observe that disruptions owing from the Lehman crisis sharplylowered Japanese aggregate industrial production in the first quarter of 2009.Compared to the decline in production from the Lehman crisis, the decline inproduction from the earthquake was far milder.
This aggregate pattern, however, masks the wide regional disparities in theimpact of the earthquake. Not surprisingly, the decline in production in Tohokuwas far larger during the earthquake than during the financial crisis. The impactof the earthquake was much more regionally concentrated than the impact ofthe financial crisis.
In Figure 2, we show a map when the 47 prefectures are aggregated into 8regions. We aggregate the prefectures up to this level, since the input-outputtables that we use extensively below are only available at this regional break-down. With this aggregation, Tohoku now includes Aomori, Akita, and Yam-agata, in addition to the three heavily impacted prefectures of Iwate, Miyagi,and Fukushima. The Kanto region includes Japan’s largest cities of Tokyo andYokohama (Kanagawa); and the Chubu region includes the important heavymanufacturing prefectures of Aichi and Shizuoka. In this aggregation, sinceChubu also includes the Hokuriku region, Chubu also turns out to be adjacentto Tohoku.
Figures 3(a) and 3(b) plot the monthly regional industrial production indices(seasonally adjusted) from 1998 to 2012 for the eight regions. Compared toFebruary 2011, industrial production in Tohoku fell by 35 percent in March2011. This decline in industrial production was much steeper than the post-financial crisis decline of 28.6 percent (between December 2008 to February2009) in Tohoku.
While the decline was not as steep as during the financial crisis, productiondeclined sharply post-earthquake in Kanto and Chubu. In March 2011, indus-trial production fell by 20 percent in Kanto and 25 percent in Chubu. TheKanto prefectures of Chiba, Saitama, Ibaragi, Tochigi, and Tokyo were directlyimpacted by the earthquake, but not the tsunami, so the direct damage to theircapital stock was minimal. However, the Kanto region has many factories usinginputs produced in the Tohoku region, so production was halted in many of thefactories. Likewise, the Chubu region is Japan’s industrial heartland, and manyof the factories located there such as the automobile factories used inputs madein Tohoku.
4
Despite its geographic proximity to Tohoku, Hokkaido was spared of muchof the impact of the earthquake. Kyushu, Shikoku, Kinki, and Chugoku are alllocated far from Tohoku. While Chugoku and Shikoku’s industrial productiondeclined after the earthquake, Kyushu’s industrial production, while decliningslightly after the earthquake has bounced back strongly. It is said that Kyushuproduces many products that are substitutes to Tohoku’s, so that Kyushu was infact a beneficiary of the damage to Tohoku’s production facilities. Surprisingly,Kinki, while including the industrial cities of Osaka and Kobe, was spared ofthe direct e↵ects from the supply disruption of the intermediate parts producedin Tohoku.
3 Indices of Interactions Among Japanese Re-
gions.
As discussed above, the earthquake to Tohoku a↵ected di↵erent regions in dif-ferent ways. Some regions like Kanto and Chubu experienced a sharp fall inindustrial production, while industrial production in Kinki, Chugoku, and otherSouthern regions barely budged. We have argued that the di↵erent propagationmechanisms in industrial production may be related to how di↵erent regionsused the inputs produced in Tohoku or were substitutes to the inputs producedin Tohoku.
In this Section, using input-output matrices that include 17 industries in our8 regions, we show how the di↵erent regions in Japan are ”interrelated.” Weconsider three measures of ”interrelatedness.” The 17 industries and 8 regionsare depicted in Table 1. The measures of ”interrelatedness” are: 1) how tworegions are ”similar” (Conley and Dupor [2003]); 2) how much two regions buyfrom each other; and 3) the geographicaladjacency of two regions.
3.1 ”Interrelatedness” Measures
We use the Japanese regional input-output matrices for 2005 compiled by RI-ETI, in which there are N = 8 regions. The raw input-output matrices includesrows (suppliers of commodities) and columns (purchasers of commodities) thatdo not correspond to any industries. On the column side, besides intermedi-ate users of commodities such as manufacturing, mining, and construction, theinput-output table contains columns for other components of gross domesticproduct: consumption, investment, change in business inventories, and govern-ment purchases. On the row side, the input-output table contains rows forcompensation to nonindustries such as wages and taxes. We address these com-ponents of the regional input-output table by: (a) removing all the final-usecolumns of the input-output table; and (b) dropping all additional rows of thetable. Finally, the original matrix has 29 industries, but we drop ”public ad-
5
ministration”, ”medical services”, ”business services”, ”personal services”, and”others”, to arrive at M = 17 industries, which are primarily in manufacturing.
3.1.1 Notation
� is the input-output matrix of dimension N ⇥M by N ⇥M . A typical (s, b)-thelement of � is �(s, b), which is the total value of transactions between s’s supplyand b’s purchase. In other words, the s-th row of � corresponds to the value ofsales of s, and the b-th column of � corresponds to the value of purchases of b.
For i, j = 1, · · · , N and m,n = 1, · · · ,M , denote ��i(m), j(n)
�as the total
value of sales from region-i’s industry-m to region-j’s industry-n.
3.1.2 ”Similarity” Regional Matrix
This economic distance measure holds that two regions are close if they buygoods from similar industries. We use the argument that regions with similarinput requirements are likely to have similar technology; so that the same shockto a given region is likely to a↵ect the output of another ”similar” region.
Steps to compute the ”similarity” matrix.
• calculate Bm
Bm(i, j) =��i(m), j(m)
�P
k ��k(m), j(m)
�
• calculate B
B(i, j) =X
m
Bm(i, j)
• calculate D
b
D
b(i, j) =
(X
k
[B(k, i)�B(k, j)]2)1/2
for i, j = 1, ....N .
This matrix is depicted in Table 2(a). According to this matrix, prefectures mostrelated to Tohoku (in order) are: Kanto, Hokkaido, Kinki, Shikoku, Chubu,Chugoku, and Kyushu.
6
3.1.3 Mutual Buying Regional Matrix
Our second measure of ”interrelatedness” measures how much two regions arebuying from each other, relative to their purchases from other regions. The morethe two regions are buying from each other, the more dependent or ”interrelated”are the two regions.
X with (i, j)-the element
X (i, j) =
Pm,n �
�i(m), j(n)
�P
k,m,n ��k(m), j(n)
� +
Pm,n �
�i(m), j(n)
�P
l,m,n ��i(m), l(n)
�
The first term is the weight of sales from region i to region j among all theregions’ sales to region j. The second term is the weight of purchases by regionj from region i among all the regions’ purchases from region i. This matrix isdepicted in Table 2(b). According to this matrix, prefectures most related toTohoku (in order) are: Kanto, Chubu, Kinki, Hokkaido, Kyushu, Chugoku, andShikoku.
3.1.4 Contiguity Matrix
The last matrix of ”interrelatedness” simply assigns a value of one if theregion shares a border with another region, deeming that if they share a border,they are ”similar.” This matrix is depicted in Table 2(c). According to thismatrix, prefectures most related to Tohoku (in order) are: Hokkaido, Kanto,Chubu, Kinki, Chugoku, Shikoku, and Kyushu.
4 Regional Spillover E↵ects
4.1 Model of Regional Spillover E↵ects
Holly, Pesaran, and Yamagata [2011] designed a method for analyzing the spatialand temporal di↵usion of shocks to a dominant region, which was applied toevaluate the e↵ects on UK housing prices due to shocks on the housing priceto London. The method treats the house price of London as a common factorand then models the contemporaneous as well as lagged dependencies amongregions conditional on London house prices.
We employ the di↵usion model of Holly, Pesaran, and Yamagata [2011] toassess the shock of Tohoku earthquake on the other regions in Japan. We makeuse of monthly data of industrial production for the 8 Japan regions defined inthe previous section. The data ranges from January 1998 to October 2012, sothat T = 178.
Denote pit as the industrial production data of region i at time t, for i =1, · · · , N and t = 1, · · · , T . The di↵usion model treats the dominant region (i =1) and the rest of the regions (i = 2, · · · , N) di↵erently by allowing for the shockon the dominant region to a↵ect the other regions not only contemporaneouslybut also through lagged impacts, while allowing for no contemporaneous e↵ectsfrom the rest of the regions on the dominant region.
7
For regions i = 2, · · · , N ,
�pit = �is
�pi,t�1 � p̄
si,t�1
�+ �i1 (pi,t�1 � p1,t�1) + ai
+kiaX
l=1
ail�pi,t�l +kibX
l=1
bil�p̄
si,t�l +
kicX
l=1
cil�p1,t�l + ci0�p1t + "it (1)
For region i = 1, �11 and c10 are set to be 0 in the above equation (1), where
p̄
sit =
NX
j=1
Sijpjt, withNX
j=1
Sij = 1
Sij � 0 is the (i, j)-th element of weighted spatial matrix S, which measuresthe spatial connection between region i and region j. Notice that S is rowstandardized in that each row of S sums up to 1. In practice, S is estimatedby row standardizing the ”interelatedness” measures defined in the previoussection, namely, the ”Similarity Matrix”, the ”Mutual Buying Matrix”, and the”Contiguity Matrix”.
As pointed out by Holly, Pesaran, and Yamagata [2011], the error correct-ing specification of equation (1) is a parsimonious representation of pair-wisecointegration of the data across regions. In addition, weak exogeneity of �p1t
in equation (1) can be tested by the procedure of Wu [1973].In the estimation of the model above, we take Kanto (Tokyo) as the dominant
region. Tokyo’s industrial production is assumed to be only a↵ected by its ownlagged industrial production and the lagged e↵ects of its neighbor’s industrialproduction. The industrial production of other regions is assumed to be a↵ectedby not only the lagged e↵ects of Tokyo and the remaining regions, but also thecontemporary e↵ects of the shocks to Tokyo.
4.2 Spatio-temporal Impulse Response Functions
We can use the estimates from the model above to examine impulse responsesboth over time and space.The persistence profile of shocks to the system overtime and across regions can be evaluated using generalized impulse responsefunction (GIRF), initially advanced by Pesaran and Shin [1998].
For horizons h = 0, 1, · · · , the impulse response of a unit (i.e. a standarddeviation) shock on the dominant region is computed as
g1(h) = E(pt+h|"1t =p�11,Ft�1)� E(pt+h|Ft�1)
=p�11 hRe1 (2)
where pt = (p1t, · · · , pNt)0 is the vector of industrial production data attime t, Ft is the filtration of information up to time t, �11 = var("1t), ande1 = (1, 0, · · · , 0)0. By stacking the N regressions in (1), Holly, Pesaran, and
8
Yamagata [2011] derived that5
�pt = a+Hpt�1 +kX
l=1
(Al +Gl)�pt�l +kX
l=0
Cl�pt�l + "t
where a, H, Al, Gl, and Cl are matrices of model parameters. It can besolved from the above expression to get
�pt = µ+⇧pt�1 +kX
l=1
�l�pt�l +R"t
where k = maxi{kia, kib, kic}, µ = Ra with R = (IN � C0)�1, ⇧ = RH,�l = R(Al +Gl +Cl).
In a VAR form, this implies that
pt = µ+k+1X
l=1
�lpt�l +R"t
where �1 = IN +⇧+ �1, �l = �l � �l�1 for l = 2, · · · , k and �k+1 = ��k.Then for h = 0, 1, · · · , h in equation (2) is defined as
h =k+1X
l=1
�l h�l
5 Empirical Results
5.0.1 Regions and their Connection
Kanto (Tokyo) is set as the dominant region in model (1) to account for bothof its contemporaneous and intertemporal impacts. We follow Holly, Pesaran,and Yamagata [2011] to estimate model (1) equation by equation using OLS..
We construct the weighted spatial matrix basing on three measures of re-gional ”interrelatedness”.
5.0.2 Total Industrial Production
Table 4, Figures 6-8 , and Figure 9, contain the results when evaluating theTohoku di↵usion e↵ects using total industry production data.
Table 4(a) reports the results based on the row standardized ”Similarity”matrix, Table 4(b) reports the results based on the row standardized ”Mutual
5See Holly, Pesaran, and Yamagata [2011] for detailed derivations of the generalized impulseresponse function in the spatial temporal model.
9
Buying” matrix, and Table 4(c) reports the results based on the row standard-ized ”Contiguity” matrix. We can see that results from Table 4(a), (b), and (c)are similar in the following ways.
”Own lag” is the estimatedPkia
l=1 ail. A positive ”own lag” e↵ect impliesthat the series continues to drift in the same direction as last period, exhibitingeither an upward trend or a downward trend. A negative ”own lag” e↵ectimplies that the series adjusts to last period’s increase by a decrease in thecurrent period, exhibiting a property like mean reverting. Estimation basedon the ”Similarity” matrix identifies the own lag e↵ects of Tohoku, Hokkaido,Chubu, Kinki, Chugokku, and Shikoku to be significant. Estimations based onthe ”Mutual Buying” matrix and the ”Contiguity” matrix identify the same setof significant own lag e↵ects, ie. own lag e↵ects are only found to be insignificantfor Kanto and Kyushu.
”Neighbour lag” estimates the dynamic spillover e↵ectsPkib
l=1 bil. A positive”neighbour lag” e↵ect implies that the series moves in the same direction as theweighted average of its neighbour in the last period. A negative ”neighbour lag”e↵ect implies the series moves in the opposite direction. Both the estimationbased on the ”Similarity” matrix and the estimation based on the ”MutualBuying” matrix identify the same set of significant neighbour lag e↵ects inHokkaido, Kinki, Chugoku, Shikoku, and Kyushu. Estimation based on the”Contiguity” matrix identifies significant neighbour lag e↵ects in Chubu, Kinki,Chugoku, Shikoku, and Kyushu. Finally, based on all three ”interrelatedness”measures, the estimated neighbour lag e↵ects on all the regions are positive,except for the neighbour lag e↵ect on Tohoku and the neighbour lag e↵ect ofKanto when the Contiguity matrix is used as the ”interrelatedness” measures.
”Kanto lag” is the estimated lagged e↵ect of Kanto. A positive ”Kanto lag”e↵ect implies that the series moved in the same direction as Kanto did in the lastperiod. Based on all the connectedness measures, the estimated ”Kanto lag”e↵ects are found to be significantly positive for Tohoku. Significantly positive”Kanto lag” e↵ects are also observed for Chubu when using the ”Similarity ma-trix” and ”Mutual buying matrix” and for Hokkaido when using the ”Contiguitymatrix”.
”Kanto current” is the estimated contemporaneous e↵ect of Kanto, ci0. Apositive ”Kanto current” e↵ect implies that the series simultaneously moves inthe same direction as Kanto. Based on all the connectedness measures, theestimated ”Kanto current” e↵ects are similar, and all of them are significantlypositive.
EC1 is estimated �i1, which is referred to as the error correction term of(pi,t�1 � p1,t�1), the deviations of region i from Kanto. The estimated EC1are similar based on the three connectedness measures, which give a signifi-cantly negative EC1 for Chugoku; the Similarity matrix additionally identifiesthat Tohoku also has a significantly negative EC1. EC2 is the estimated �is,the error correction term of (pi,t�1 � p̄
si,t�1), the deviation of region i from its
neighbours. The estimated EC2 based on the three ”interrelatedness measures”identify Chugoku and Shikoku to have significantly negative EC2; the ”MutualBuying matrix” and the ”Contiguity Matrix” both identify Tohoku to have a
10
significantly negative EC2.WH-stat is the Wu-Hausman test statistics (Wu [1973]) testing the null
hypothesis that production changes in the dominant region Kanto are exogenousto production changes in the other regions. The results show that most of theregressions passed the Wu-Hausman test, except for the regression of Hokkaidobased on the ”Contiguity” matrix.
kia, kib, and kic are all selected by the Schwarz Bayesian criterion (SBC).Based on all three ”interrelatedness” measures, SBC selected kia to be equal to 1and kib to be equal to 1 or 2. SBC selected the lag orders kic = 0, producing theestimated ”Kanto lag” e↵ects,
Pkic
l=1 cil, to be 0 for Kinki, Chugoku, Shikoku,and Kyushu.
Figure 6, Figure 7, and Figure 8 plot the estimated generalized impulseresponse functions (GIRF) caused by a 1 unit (i.e. 1 standard deviation) positiveshock to Tohoku’s industrial production. Figure 6, Figure 7, and Figure 8 allowus to observe a similar pattern across regions.
The persistence profile of Tohoku shows that it takes about 2 years forTohoku to absorb 1/2 of a positive unit of shock to its monthly IP level. Thepersistence profiles of all the regions other than Tohoku show that a positiveunit of shock to Tohoku can be quickly and well adjusted within about a year’stime. Figure 6, Figure 7, and Figure 8 also reveal similar persistence profiles interms of the shape and the degree of the persistence basing on the three di↵erent”interrelatedness” measures.
For selected time periods h = 0, 3, 5, 10, 20, 50, Figure 9 depicts the GIRFacross regions and over time. The results are estimated for the ”Similarity,””Mutual Buying,” and ”Contiguity” matrices. The regions are ordered on thehorizontal axis from left to right according to their ”closeness” (according toeach of the three ”interrelatedness” matrices) to Tohoku. For example, in Figure9(a), according to the ”similarity” matrix, the ordered horizontal axis shows thatthe ”closest” region to Tohoku is Kanto, followed by Hokkaido, Kinki, Shikoku,Chubu, Chugoku, and Kyushu.
If ”proximity”–according to the various definitions– results in higher spillovers,then we should see a declining pattern in the graphs. As the regions becomefurther from Tohoku, the impact of the Tohoku shock should dissipate. In gen-eral we see no such pattern in the graphs, except for Kanto. A positive Tohokushock always tends to raise Kanto and Shikoku industrial production. WhileKanto is ”close” to Tohoku by all three measures, Shikoku is rather ”far,” butis strongly a↵ected by Tohoku shocks. Chubu is relatively ”close” to Tohoku,but its total industrial production is scarcely a↵ected by Tohoku.
5.0.3 Key Industries
There certainly have been other shocks hitting Japan in the few months fol-lowing the 2011 earthquake. There was a nationwide adverse demand shock inthe Spring and Summer of 2011, owing to negative consumer sentiment. The
11
regulation and forced conservation of electrical power usage throughout Japanlowered production in other sectors. In principle, the impulse response method-ology above controls for these other factors, since the impulse response shockis supposed to identify only the partial e↵ect on other regions from the adverseearthquake shock to Tohoku.
To focus more on the role of Tohoku’s intermediate products production onthe output in the rest of Japan, below we trace out the shocks emanating fromthe decline in the production of electrical and automotive parts in Tohoku.
Electric Machinery Industry Tables 7(a) and (b), Figure 4, and Figures12-13 contain the results when evaluating the e↵ects of the Tohoku earthquakeusing production data of only the Electric Machinery Industry. Tables 5(a) and5(b) report the results based on the ”Mutual Buying” and ”Contiguity” matrices(the ”Similarity” matrix is undefined when we have only one industry.) All ofthe estimates in the Tables appear reasonable.
Figure 12 depicts the e↵ects of a one standard error positive shock to electri-cal industry production in Tohoku. The impacts are relative large and persistentin Hokkaido, Chubu, Shikoku, and Kyushu. The e↵ects are relatively small inKinki and Hokkaido. In Figure 14, we order the regions according to theirproximity to Tohoku in the electrical machinery industry. As expected from theimpulse responses, the e↵ects to Kanto and Kinki are small, while they are largeand persistent for Hokkaido, Chubu, Shikoku and Kyushu. It appears that theshocks to the electrical machinery industry had relatively large and importante↵ects on the electric machinery industries throughout Japan, except for thevery large (in terms of GDP and size) regions of Kanto and Kinki, where thee↵ects are minimal.
Transportation Equipment Industry Tables 8(a) and (b), Figure 5, andFigures 15-17 contain the results when evaluating the e↵ects of the Tohokuearthquake using production data of only the Transportation Industry. Tables6(a) and 6(b) show that the e↵ects are relatively large for Chubu, Kinki, andKyushu, although the e↵ects dissipate quickly for Kinki. In Figure 17, weorder the regions according to their proximity to Tohoku in the transportationequipment industry. With regards to the e↵ect on Chubu, a major automotiveproducer, the immediate e↵ects are large, but they dissipate after one period,then the e↵ects grow again. The e↵ects are relatively large for Kyushu andHokkaido through all periods.
6 Conclusion
In this paper, we traced out how a decline in industrial production in one re-gion can be propagated throughout Japan. We take Kanto as the dominant
12
region and explain how a shock to industrial production in Tohoku–owing to theearthquake–can be propagated throughout Japan. In our econometric model,regions and industries within regions are linked by an input-output structureand this input-output structure disciplines how the shocks are spatially propa-gated.
Our emphasis on the input-output structure in the propagation of shocksafter the Tohoku earthquake is motivated by the fact that much of the im-pact of the Tohoku earthquake on other regions was driven by the decline inintermediate inputs produced in Tohoku.
In general, while we definately find e↵ects on industrial production from theTohoku earthquake, the regional e↵ects do not seem to depend on our defini-tions of proximity, although we observe significant heterogeneity in how di↵erentprefectures were a↵ected by the spillovers from the Tohoku earthquake.
13
Table 1: Regions and Industries(a) Regions
01 Hokkaido02 Tohoku03 Kanto04 Chubu05 Kinki06 Chugoku07 Shikoku08 Kyushu + Okinawa
(b) Industries020 Mining030 Beverages and Foods040 Textile products050 Timber, wooden products and furniture060 Pulp, paper, paperboard, building paper070 Chemical products080 Petroleum and coal products090 Plastic products100 Ceramic, stone and clay products110 Iron or steel products120 Non-ferrous metal products130 Metal products140 General machinery150 Electrical machinery160 Transportation equipment170 Precision instruments180 Miscellaneous manufacturing products
14
Table 2: Distance Measures(a) Similarity Matrix
Tohoku Hokkaido Kanto Chubu Kinki Chugoku Shikoku KyushuKanto 0 11.366 12.227 13.034 12.703 14.064 12.999 13.905Tohoku 11.366 0 11.843 12.612 11.996 13.105 11.998 13.295Hokkaido 12.227 11.843 0 12.352 12.127 13.169 12.046 13.476Chubu 13.034 12.612 12.352 0 11.497 12.993 12.050 13.356Kinki 12.703 11.996 12.127 11.497 0 11.648 9.848 12.287
Chugoku 14.064 13.105 13.169 12.993 11.648 0 10.812 12.414Shikoku 12.999 11.998 12.046 12.050 9.848 10.812 0 11.796Kyushu 13.905 13.295 13.476 13.356 12.287 12.414 11.796 0
(b) Mutual Buying MatrixTohoku Hokkaido Kanto Chubu Kinki Chugoku Shikoku Kyushu
Kanto 0 0.269 0.160 0.261 0.182 0.106 0.109 0.166Tohoku 0.361 0 0.038 0.080 0.062 0.032 0.016 0.037Hokkaido 0.239 0.045 0 0.107 0.059 0.019 0.011 0.021Chubu 0.256 0.117 0.063 0 0.206 0.097 0.076 0.140Kinki 0.224 0.083 0.043 0.205 0 0.130 0.147 0.106
Chugoku 0.147 0.039 0.025 0.120 0.165 0 0.109 0.119Shikoku 0.187 0.025 0.014 0.096 0.137 0.070 0 0.076Kyushu 0.147 0.029 0.011 0.083 0.107 0.096 0.056 0
(c) Contiguity MatrixTohoku Hokkaido Kanto Chubu Kinki Chugoku Shikoku Kyushu
Tohoku 0 1 1 1 0 0 0 0Hokkaido 1 0 0 0 0 0 0 0Kanto 1 0 0 1 0 0 0 0Chubu 1 0 1 0 1 0 0 0Kinki 0 0 0 1 0 1 1 0
Chugoku 0 0 0 0 1 0 1 1Shikoku 0 0 0 0 1 1 0 1Kyushu 0 0 0 0 0 1 1 0
15
References
Timothy G. Conley and Bill Dupor. A spatial analysis of sectoral complemen-tarity. Journal of Political Economy, 111(2):311–352, 2003.
Sean Holly, M. Hashem Pesaran, and Takashi Yamagata. The spatial and tem-poral di↵usion of house prices in the uk. Journal of Urban Economics, 69(1):2 – 23, 2011.
H. Hashem Pesaran and Yongcheol Shin. Generalized impulse response analysisin linear multivariate models. Economics Letters, 58(1):17–29, 1998.
De-Min Wu. Alternative tests of independence between stochastic regressorsand disturbances. Econometrica, 41(4):733–50, 1973.
16
Tab
le3:
Estim
ationresultsof
region
specificdi↵usion
equationforTotal
Industrial
Production
(a)Sim
ilarityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.3
(-2.62
4)0.88
(6.149
)-
--
--
12
-Toh
oku
-0.262
(-3.25
3)-0.026
(-0.16
8)0.44
2(3.220
)0.81
5(11)
--0.22(-3.65
3)-0.815
11
1Hok
kaid
-0.478
(-6.51
8)0.39
4(4.042
)-
0.30
5(3.666
)-
-1.13
41
10
Chubu
-0.217
(-3.03
1)0.47
1(3.818
)-
1.01
2(14.49
7)-
-0.26
31
10
Kinki
-0.541
(-4.78
5)0.50
6(5.437
)-
0.51
3(7.752
)-
-0.108
(-2.34
3)0.09
32
10
Chugo
ku
-0.119
(-1.39
4)0.30
7(2.86)
-0.62
9(7.424
)-
-0.37(-4.56
1)-0.632
11
0Shikok
u-0.59(-4.77
9)0.51
9(3.319
)-
0.52
6(4.192
)-
-2.09
92
10
Kyusyu
-0.059
(-0.65
8)0.1(0.622
)0.36
6(2.896
)0.78
5(10.88
9)-
-0.07(-2.13
8)1.03
91
11
(b)Mutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.309
(-2.56
6)0.86
8(6.022
)-
--
--
12
-Toh
oku
-0.254
(-3.02
4)0.44
5(4.132
)-
0.81
0(11.39
1)-
-0.254
(-3.60
9)0.38
71
10
Hok
kaid
-0.482
(-6.70
3)0.40
1(4.528
)-
0.32
8(4.100
)-
-0.52
71
10
Chubu
-0.262
(-3.43
2)0.54
4(4.153
)-
1.00
1(14.38
3)-
-0.14
21
10
Kinki
-0.590
(-5.11
1)0.52
9(6.113
)-
0.52
2(97.76
0)-
--0.906
21
0Chugo
ku
-0.174
(-2.04
8)0.33
0(3.102
)-
0.62
5(7.150
)-
-0.251
(-3.56
9)-0.912
11
0Shikok
u-0.617
(-4.99
3)0.56
3(3.698
)-
0.51
9(4.187
)-
-1.64
02
10
Kyusyu
-0.252
(-2.56
4)0.60
7(5.283
)-
0.78
0(10.99
6)-
-0.87
22
10
(c)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.159
(-1.30
6)0.57
6(4.82)
--
--
-1
2-
Toh
oku
-0.216
(-2.62
8)-0.152
(-1.03
8)0.5(3.444
)0.83
1(11.89
8)-
-0.286
(-4.09
8)0.09
11
1Hok
kaid
-0.454
(-6.34
5)-0.089
(-1.06
4)0.44
4(2.537
)0.37
3(4.696
)-
--0.699
11
1Chubu
-0.193
(-2.60
3)0.36
9(3.2)
-1.04
2(15.05
2)-
--0.458
11
0Kinki
-0.570
(-5.05
6)0.11
4(1.245
)0.36
3(3.663
)0.57
4(8.578
)-
-0.104
(-2.60
5)-1.812
21
1Chugo
ku
-0.185
(-2.30
3)0.27
6(2.916
)-
0.63
1(6.888
)-0.148
(-2.88
2)-
-1.649
11
0Shikok
u-0.292
(-3.81
0)-0.371
(-1.55
4)0.53
4(2.855
)0.67
7(4.925
)-0.097
(-2.35
9)-
-0.543
12
1Kyusyu
-0.252
(-2.68
1)0.14
6(2.142
)0.43
1(2.838
)0.83
1(11.59
7)-
--1.027
21
1
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
17
Tab
le4:
Estim
ationresultsof
region
specificdi↵usion
equationforElectricMachinery
(a)E
lectricMutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to0.19
(1.546
)0.17
2(2.009
)-
--
--
21
-Toh
oku
-0.278
(-3.26
7)0.34
6(2.663
)-
0.59
5(5.485
)-
-0.224
(-3.33
8)0.18
91
10
Hok
kaid
-0.166
(-2.03
3)0.51
2(2.201
)-
0.60
4(3.08)
--0.091
(-2.12
1)0.45
41
10
Chubu
-1.011
(-9.30
4)1.27
4(5.688
)-
0.20
3(1.392
)-
--0.153
22
0Kinki
-0.344
(-4.38
)0.12
5(0.932
)0.31
4(1.045
)0.47
4(5.175
)-
-0.07
11
11
Chugo
ku
-0.149
(-1.73
1)0.50
2(3.013
)-
0.54
1(4.916
)-0.093
(-2.27
8)-
-0.525
12
0Shikok
u-0.191
(-2.32
9)0.33
7(1.993
)-
0.60
8(4.276
)-
-0.06
61
10
Kyusyu
-0.186
(-2.32
5)0.72
8(3.14)
-0.4(2.822
)-
-0.103
(-2.47
1)1.02
11
20
(b)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to0.26
4(2.332
)0.07
4(1.359
)-
--
--
21
-Toh
oku
-0.314
(-3.92
3)0.02
4(0.223
)0.34
3(2.661
)0.59
2(5.441
)-
-0.13(-2.94
5)-0.293
11
1Hok
kaid
-0.223
(-2.84
9)0.48
5(2.278
)-
0.49
3(2.445
)-
-1.29
51
20
Chubu
-1.026
(-9.26
2)1.49
6(5.459
)-
0.10
5(0.686
)-
-0.44
92
30
Kinki
-0.337
(-4.29
8)0.04
1(0.538
)0.37
5(1.3)
0.48
(5.247
)-
-0.89
81
11
Chugo
ku
-0.132
(-1.58
7)0.36
7(2.928
)-
0.58
5(5.339
)-0.089
(-2.17
9)-
-0.175
12
0Shikok
u-0.141
(-1.76
4)0.10
3(0.693
)-
0.61
2(4.119
)-
--0.114
11
0Kyusyu
-0.268
(-3.48
9)0.72
1(4.349
)-
0.36
4(2.656
)-
-0.094
(-2.49
7)0.23
21
20
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
18
Tab
le5:
Estim
ationresultsof
region
specificdi↵usion
equationforTransportation
(a)Transp
ortation
Mutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to0.26
1(2.171
)-0.070
(-0.88
3)-
--
--
21
-Toh
oku
-0.142
(-1.74
2)0.31
3(2.457
)-
1.04
9(12.75
4)-
-0.053
(-1.81
5)-0.489
11
0Hok
kaid
-0.052
(-0.76
7)-0.091
(-1.79
6)-
0.93
5(7.839
)-
-0.056
(-2.40
4)-0.786
11
0Chubu
-0.017
(-0.21
3)-1.67(-4.24
9)-
-0.012
(-0.04
8)-
-0.326
(-5.43
4)0.29
91
30
Kinki
-0.68(-5.65
8)-0.044
(-0.62
3)0.25
9(2.655
)0.55
2(8.481
)-0.118
-3.218
--1.493
21
1Chugo
ku
-0.073
(-1.02
7)0.15
4(1.462
)-
0.77
5(9.441
)-
-0.114
(-3.11
6)-0.445
11
0Shikok
u-0.259
(-3.35
3)0.08
9(0.752
)-
0.25
7(2.462
)-
-0.93
61
10
Kyusyu
-0.205
(-2.56
1)-0.294
(-2.36
1)0.51
9(3.303
)0.90
1(9.944
)-
-0.028
(-1.40
1)0.91
71
22
(b)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to0.24
(2.119
)-0.056
(-1.21
7)-
--
21
--
-Toh
oku
-0.141
(-1.71
1)-0.006
(-0.08
7)0.31
4(2.672
)1.03
2(12.67
8)-
-0.087
(-2.58
6)-0.648
11
1Hok
kaid
-0.160
(-1.96
8)0.16
2(1.639
)-
0.99
0(8.358
)-
--0.818
11
0Chubu
-0.039
(-0.48
2)-1.69(-3.84
)-
-0.055
(-0.21
6)-
-0.282
(-5.16
)-0.677
13
0Kinki
-0.682
(-5.69
3)-0.031
(-0.57
5)0.24
8(2.628
)0.55
2(8.472
)-0.118
(-3.22
5)-
-1.367
21
1Chugo
ku
-0.026
(-0.35
9)0.03
9(0.368
)-
0.76
7(9.393
)-
-0.162
(-3.78
)-0.189
11
0Shikok
u-0.260
(-3.34
2)0.06
9(0.628
)-
0.25
2(2.412
)-
-1.03
71
10
Kyusyu
-0.162
(-2.03
1)-0.041
(-0.41
2)0.32
9(2.561
)0.93
2(10.83
)-
-0.085
(-2.55
9)-1.360
11
1
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
19
Tab
le6:
Estim
ationresultsof
region
specificdi↵usion
equationforTotal
Industrial
Production(1998-2008)
(a)Sim
ilarityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.113
(-0.856
)-0.223
(-1.216
)-
--
--
11
-Toh
oku
-0.631
(-4.758
)0.31
5(1.32
9)
0.76
3(3.752
)0.79
3(15.44
9)
--0.219
(-3.37
1)-0.750
22
3Hok
kaid
-0.787
(-7.00
1)
-0.276
(-1.60
9)
0.51
9(2.821
)0.57
9(9.495
)-
-1.06
82
11
Chubu
-0.231
(-3.39
6)0.92
4(5.478
)-
0.63
8(11.48
5)
-0.02(-1.46
8)-
-0.029
13
0Kinki
-0.777
(-8.05
0)
0.39
8(5.025
)-
0.59
2(12.63
8)
--
1.06
92
10
Chugo
ku
-0.161
(-2.10
1)
0.46
9(3.335
)-
0.66
1(11.56
4)
--0.446
(-5.46
7)
0.44
11
20
Shikok
u-0.256
(-3.53
3)
0.15
3(1.066
)-
0.48
5(4.836
)-0.219
(-4.12
3)-
1.10
81
10
Kyushu
0.12
5(1.144
)0.02
2(-0.17
6)
0.21
7(2.301
)0.65
9(15.40
4)
--0.087
(-2.57
6)
0.79
33
11
(b)Mutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.287
(-2.29
6)1.54
1(5.66
)-
--
--
14
-Toh
oku
-0.605
(-3.63
8)1.38
1(5.225
)-
-0.73(8.55
1)-
-0.221
(-2.89
6)-0.412
23
-Hok
kaid
-0.483
(-6.14
6)0.35
1(2.99
5)-
0.33
6(3.59
7)-
-0.00
81
10
Chubu
-0.26(-3.18
)0.39
7(2.64
3)-
0.95
2(12.34
1)-0.016
(-1.13
7)-
-0.418
11
0Kinki
-0.251
(-2.94
1)0.35
1(3.33
9)-
0.50
4(6.85
3)-
-0.144
(-2.78
1)-1.105
11
0Chugo
ku
-0.182
(-1.96
5)0.35
4(2.60
7)-
0.65
5(6.904
)-
-0.296
(-3.48
8)-1.878
11
0Shikok
u-0.275
(-3.38
5)-1.357
(-3.06
5)-
0.62
4(4.209
)-0.26(-4.58
1)-
1.05
51
40
Kyushu
-0.117
(-1.29
2)0.50
4(4.733
)-
0.64
9(9.7)
--
-0.82
11
0
(c)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.24(-2.01
9)0.78
8(3.83
5)-
--
--
14
-Toh
oku
-0.509
(-3.47
3)-0.225
(-1.48
4)1.04
7(4.38
8)0.64
2(7.53
4)-
-0.298
(-4.10
7)-0.944
21
3Hok
kaid
-0.457
(-5.86
7)-0.164
(-1.74
9)0.37
6(1.87
1)0.30
8(3.03
)-
--0.189
11
1Chubu
-0.255
(-3.20
5)0.21
4(1.72
7)-
0.89
5(10
.824
)-
-0.33
91
10
Kinki
-0.67(-5.38
2)0.14
7(1.193
)0.27
2(1.73
2)0.5(6.36
2)-
--2.8
21
1Chugo
ku
-0.274
(-3.00
2)0.14
7(1.55
6)-
0.50
2(4.98
3)-
-0.194
(-2.68
1)-1.58
21
1Shikok
u-0.244
(-2.90
4)0.09
5(0.43
7)-
0.36
4(2.385
)-0.203
(-3.58
1)-
3.97
41
10
Kyushu
-0.204
(-2.28
6)0.06
3(1.05
3)0.32
5(2.309
)0.60
1(8.59
8)-
--0.877
11
1
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
20
Tab
le7:
Estim
ationresultsof
region
specificdi↵usion
equationforElectricMachinery(1998-2008)
(a)E
lectricMutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to0.33
1(3.088
)0.18
3(2.10
7)-
--
--
21
1Toh
oku
-0.284
(-3.58
3)-0.093
(-0.52
)0.36
8(2.46
2)0.63
9(6.91
7)-
-0.17(-2.97
9)0.13
61
11
Hok
kaid
-0.185
(-2.41
5)0.31
0(1.62
7)-
0.50
4(3.05
1)-
-0.172
(-3.18
3)1.43
81
10
Chubu
-0.587
(-5.01
2)1.30
2(4.82
9)-
0.08
2(0.50
3)-0.03
1(-1.66
1)-
-0.814
22
0Kinki
-0.276
(-3.50
5)0.12
1(0.87
7)0.28
6(2.43
7)0.58
4(6.98
8)-
-0.132
(-2.62
6)-1.024
11
1Chugo
ku
-0.091
(-1.13
6)0.16
(1.41
)-
0.56
8(5.00
2)-0.106
(-2.12
2)-
-0.779
11
0Shikok
u-0.341
(-4.76
6)0.2(1.56
)-
0.47
0(4.33
1)-
-0.066
(-2.23
2)1.67
11
10
Kyushu
-0.126
(-1.62
9)0.72
(2.86
9)-
0.34
9(2.35
8)-
-0.198
(-3.86
3)0.49
01
20
(b)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to0.28
6(2.26
9)0.07
5(1.12
9)-
--
--
21
-Toh
oku
-0.302
(-3.38
6)-0.049
(-0.43
6)0.41
4(3.13
7)0.56
4(5.026
)-
-0.156
(-2.81
7)0.00
11
11
Hok
kaid
-0.127
(-1.36
5)-0.147
(-0.94
3)-
0.58
6(2.817
)-
-0.232
(-3.12
7)1.25
71
10
Chubu
-1.149
(-10
.199
)0.91
9(4.049
)-
0.09
4(0.67
8)-
-1.06
62
20
Kinki
-0.3
(-3.48
2)-0.053
(-0.58
3)0.43
(4.07
8)0.49
5(5.11
2)-
-0.088
(-2.25
6)0.23
91
11
Chugo
ku
-0.13(-1.40
5)0.45
(2.483
)-
0.60
3(4.93
)-0.096
(-1.79
5)-
-0.611
12
0Shikok
u-0.26(-2.96
8)0.05
4(0.34
7)-
0.55
8(3.91
4)-
-0.40
31
10
Kyushu
-0.252
(-2.92
1)0.29
8(1.55
6)-
0.35
4(2.59
2)-
--0.018
12
0
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
21
Tab
le8:
Estim
ationresultsof
region
specificdi↵usion
equationforTransportation
(1998-2008)
(a)Transp
ortation
Mutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.294
(-2.62
8)-0.014
(-0.13
9)-
--
--
11
-Toh
oku
-0.199
(-2.23
6)0.38
7(2.69
2)-
1.04
3(10.12
2)-
--0.489
11
0Hok
kaid
-0.24(-2.94
5)0.06
6(1.147
)-
0.78
8(6.59
5)-
-0.86
11
0Chubu
-0.089
(-1.
)-0.096
(-0.39
5)-
0.34
8(1.60
7)-0.367
(-4.62
)-
-0.284
11
0Kinki
-068
2(-4.89
1)0.10
3(1.231
)-
0.54
8(7.01
3)-0.133
(-2.89
6)-
1.70
82
10
Chugo
ku
-0.243
(-1.89
5)0.21
(1.733
)-
0.84
1(8.81
3)-
-0.122
(-2.61
4)0.13
82
10
Shikok
u-0.119
(-1.40
3)0.23
7(1.81
7)-
0.45
(3.92
1)-
-0.26
91
10
Kyushu
-0.13(-1.72
8)0.10
9(1.258
)-
0.83
(8.627
)-
-2.18
21
1
(b)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Kan
to-0.891
(-5.13
1)-0.017
(-0.33
2)-
--
--
31
-Toh
oku
-0.049
(-0.66
5)0.12
9(1.46
)-
0.72
9(7.898
)-0.072
(-2.73
2)-
2.92
41
10
Hok
kaid
-0.049
(-0.61
9)0.46
2(2.89
2)-
0.26
3(2.11
5)-
-0.154
(-3.36
3)1.97
61
20
Chubu
-0.041
(-0.53
8)0.01
8(0.09
8)-
0.35
9(2.28
9)-
-0.509
(-6.38
8)-0.157
12
0Kinki
-0.341
(-5.22
7)0.07
3(1.24
9)-
0.40
1(6.13
3)-0.184
(-4.30
7)-
1.96
11
10
Chugo
ku
0.06
4(0.86
4)0.10
4(1.153
)-
0.59
(7.11
2)-0.101
(-3.45
3)-
2.64
31
10
Shikok
u-0.138
(-1.72
3)0.81
(3.75
4)-
-0.017
(-0.13
7)-
-0.122
(-2.92
5)2.01
12
0Kyushu
-0.193
(-2.50
1)0.35
5(2.83
5)0.83
4(2.074
)0.48
2(4.63
4)-
-1.71
51
22
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
22
Tab
le9:
Estim
ationresultsof
region
specificdi↵usion
equationforTotal
Industrial
Production(C
hubu)
(a)Sim
ilarityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
-0.304
(-2.58
)1.18
(4.945
)-
-0.057
(-1.98
)-
--
12
-Toh
oku
-0.683
(-6.33
2)-0.041
(-0.20
4)0.34
8(1.65
7)0.96
5(15.01
1)-
-0.23
72
11
Hok
kaid
-0.449
(-6.42
4)-0.194
(-1.36
6)0.41
4(3.93
6)
0.27
1(5.611
)-
-0.031
(-1.03
5)-1.901
11
2Kan
to-0.175
(-2.29
7)0.01
4(0.139
)-
0.67
(18
.616
)-
-1.20
31
10
Kinki
-0.194
(-2.56
5)0.22
8(2.901
)-
0.28
7(7.07
1)-
-126
(-2.83
3)-0.886
11
0Chugo
ku
-0.207
(-2.73
2)0.32
2(3.574
)-
0.34
5(6.70
5)-
--1.55
11
0Shikok
u-0.268
(-3.63
7)
0.21
8(1.891
)-
0.24
5(3.53
8)-
-0.122
(-2.85
9)0.84
41
10
Kyushu
-0.040
(0.467
)0.08
3(0.757
)-
0.51
8(10.98
1)-0.053
(-1.95
9)-
-0.284
11
0
(b)Mutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
-0.38(-3.09
4)1.23
1(5.145
)-
--
--
12
-Toh
oku
-0.523
(-4.19
4)0.44
5(2.95)
-0.88
2(14.15
4)
--0.203
(-2.69
9)1.62
82
10
Hok
kaid
-0.444
-(6.67
7)-0.278
(-1.92
9)0.41
1(2.538
)0.29
2(6.288
)-
--2.302
11
1Kan
to-0.153
(-1.76
7)-0.018
(-0.17
1)-
0.67
1(18.69
4)-
-1.12
71
10
Kinki
-0.194
(-2.54
9)0.21
(2.787
)-
0.29
3(7.174
)-
-0.117
(-2.69
)-0.942
11
0Chugo
ku
-0.306
(-3.86
8)0.33
5(3.591
)-
0.34
6(6.722
)-
--1.231
11
0Shikok
u-0.281
(-3.77
8)0.24
4(2.096
)-
0.24
2(3.498
)-
-0.113
(-2.71
7)1.41
31
10
Kyusyu
0.00
8(0.092
)0.13
9(1.308
)-
0.51
61(0.95
8)-0.051
(-1.86
1)-
0.69
11
10
(c)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
-0.324
(-2.64
7)0.93
7(4.534
)-
--
--
12
-Toh
oku
-0.445
(-3.72
)0.32
4(2.404
)-
0.90
5(14.58
9)-
-0.237
(-3.29
4)0.76
12
10
Hok
kaid
-0.443
(-6.83
7)
-0.188
(-3.68
)0.37
9(2.69)5
0.30
2(6.802
)-
--2.416
11
1Kan
to-0.133
(-1.49
3)-0.622
(-4.31
7)0.55
7(3.412
)0.70
1(19.51
1)-
-0.22
11
22
Kinki
-0.253
(-3.34
8)0.31
4(4.186
)-
0.27
4(6.7)
--
0.01
1.1
0Chugo
ku
-0.276
(-3.58
5)0.30
7(3.276
)-
0.34
6(6.685
)-
--0.729
11
0Shikok
u-0.284
(-3.758
)0.30
2(2.279
)-
0.21
7(3.071
)-
-0.121
(-2.659
)2.08
91
10
Kyusyu
0.02
(0.291
)0.21
4(2.372
)-
0.49
0(10.68
6)-
-0.053
(-1.97
8)1.69
21
10
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
23
Tab
le10:Estim
ationresultsof
region
specificdi↵usion
equationforElectricMachinery(C
hubu)
(a)E
lectricMutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
-1.146
(-10
.199
)1.34
1(6.215
)-
--
22
--
-Toh
oku
-0.266
(-2.884
)0.31
9(2.007
)-
0.18
6(3.349
)2.01
11
10
--0.244
(-3.35
4)Hok
kaid
-0.203
(-2.254
)0.56
3(2.342
)-
0.02
2(0.253
)1.11
41
10
--0.163
(-2.81
2)Kan
to0.09
4(0.704
)0.19
4(1.910
)-
0.02
5(0.701
)1.44
42
10
--
Kinki
-0.281
(-3.59
)0.31
0(2.637
)0.14
7(0.545
)0.15
5(3.260
)0.27
71
12
--
Chugo
ku
-0.231
(-2.647
)0.40
3(3.098
)-
0.10
4(2.229
)0.02
81
10
--
Shikok
u-0.196
(-2.17
9)0.07
8(0.373
)0.40
3(2.976
)0.32
1(4.236
)-0.072
(-2.46
6)-
-0.385
11
2Kyusyu
-0.195
(-2.18
3)0.40
2(1.858
)-
0.03
7(0.462
)1.48
21
10
--0.098
(-2.50
5)
(b)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
-0.992
(-9.88
1)1.29
0(6.486
)-
--
--
22
-Toh
oku
-0.234
(-2.98
1)0.15
2(1.452
)-
0.20
2(3.583
)-
-0.147
(-2.90
1)1.23
11
10
Hok
kaid
-0.146
(-1.90
8)0.51
4(2.917
)-
0.01
7(0.197
)-
-0.107
(-2.46
7)0.96
61
20
Kan
to0.19
2(1.638
)0.06
2(1.175
)-
0.02
8(0.797
)-
-1.04
82
10
Kinki
-0.230
(-3.06
3)0.16
5(1.651
)0.17
8(0.65)
0.17
8(3.810
)-
--0.830
11
2Chugo
ku
-0.117
(-1.48
6)0.12
7(1.346
)-
0.11
(2.298
)-
--0.082
11
0Shikok
u-0.182
(-2.40
5)0.04
6(0.37)
0.41
6(3.348
)0.32
4(4.410
)-0.073
(-2.47
6)-
-0.356
11
2Kyusyu
-0.154
(-1.96
9)0.40
0(2.337
)-
0.02
7(0.335
)-
-0.068
(-2.28
7)1.80
61
10
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
24
Tab
le11:Estim
ationresultsof
region
specificdi↵usion
equationforTransportation
(Chu
bu)
(a)Transp
ortation
Mutual
BuyingMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurren
tEC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
0.12
3(1.606
)-1.318
(-5.75
3)-
--
-0.296
(-5.01
2)-
11
-Toh
oku
-0.059
(-0.485
)0.35
9(1.662
)-
-0.014
(-0.32
2)-
-0.133
(-2.46
3)0.70
71
10
Hok
kaid
-0.297
(-3.35
6)1.37
5(5.319
)-0.205
(-3.29
8)0.04
4(0.904
)-0.1
(-3.76
1)-
-1.089
12
1Kan
to-0.152
(-1.19
1)0.05
7(0.365
)0.00
00.00
0-0.050
(-1.69
1)-
-1.28
11
10
Kinki
-0.496
(-5.151
)0.30
2(2.753
)0.00
00.00
0-0.007
(-0.29
7)-
-0.84
11
10
Chugo
ku
-0.216
(-2.26
2)0.21
2(1.609
)0.00
00.00
0-0.045
(-1.28
7)-0.063
(-2.70
4)-
0.05
21
10
Shikok
u-0.245
(-3.16
1)0.08
2(0.775
)0.00
00.00
0-0.011
(-0.34
2)-
-1.24
01
10
Kyusyu
0.02
1(0.19)
0.02
1(0.1)
0.00
00.00
0-0.035
(-0.74
7)
-0.058
(-2.48
7)-
0.30
41
10
(b)Con
tigu
ityMatrix
OwnLag
Neigh
bL
Kan
toLag
Kan
toCurrent
EC1
EC2
Wu-H
aus
kia
kib
kic
Chubu
-0.059
(-0.80
9)-1.013
(-5.37
0)-
--
-0.215
(-4.543
)-
11
-Toh
oku
0.00
9(0.11
2)0.11
8(1.131
)0.00
00.00
0-0.043
)-0.99
3)-
-2.61
81
10
Hok
kaid
-0.300
(-3.86
1)0.44
5(2.581
)-0.142
(-2.36
1)-0.008
(-0.177
)-0.082
(-2.91
4)-
-0.277
13
1Kan
to-0.128
(-1.58
3)0.04
0(0.748
)-
-0.048
(-1.62
4)-
-0.64
01
10
Kinki
-0.477
(-5.42
5)0.37
9(3.086
)-0.087
(-0.26
2)-0.009
(-0.36
6)-
-1.39
91
11
Chugo
ku
-0.190
(-2.15
5)0.17
7(1.530
)-
-0.047
(-1.37
2)-0.069
(-2.98
1)-
-0.495
11
0Shikok
u-0.216
(-2.87
9)0.08
7(1.043
)-
0.01
3(0.425
)-0.049
(-2.44
5)-
-0.288
11
0Kyusyu
0.04
3(0.463
)-0.055
(-0.28
8)-
-0.039
(-0.83
3)-0.058
(-2.49
9)-
1.32
41
10
Note:
Kan
to’s
lagged
e↵ectareestimated
tobe0an
dthusom
ittedfrom
thereport.
Lag
ordersareselected
separatelyby
SchwarzBayesiancriterionfrom
amax
imum
lagorder
of4.
25
Figure
1:
Japanese
IndustrialP
roduction
Grow
th
(Q
uarterly,sa)
Figure
1:
Japanese
IndustrialP
roduction
Grow
th
(Q
uarterly,sa)
−0.2
5
−0.2
−0.1
5
−0.1
−0.0
50
0.050.
1
2008Q3
2008Q4
2009Q1
2009Q2
2009Q3
2009Q4
2010Q1
2010Q2
2010Q3
2010Q4
2011Q1
2011Q2
2011Q3
2011Q4
2012Q1
2012Q2
2012Q3
Indu
stria
l Pro
duct
ion
Gro
wth
(Qua
rterly
)
7
1
Figure 2: Japanese Regional Map
Figure 2: Japanese Regional Map
8
2
Figure
3:
Tim
eSeries
Plot
ofTotalIndustrialP
roduction
Data
Figure
3:
Tim
eSeries
Plot
ofTotalIndustrialP
roduction
Data
Time
2000
2005
2010
7090110
Tohoku
Hokkaido
Kanto
Chubu
Time
2000
2005
2010
7090110
Kinki
Chugoku
Shikoku
Kyushu
9
3
Figure
4:
Tim
eSeries
Plot
ofE
lectric
Machinery
Industry
Data
Figure
4:
Tim
eSeries
Plot
ofE
lectric
Machinery
Industry
Data
Time
2000
2005
2010
40100
Tohoku
Hokkaido
Kanto
Chubu
Time
2000
2005
2010
60120180
Kinki
Chugoku
Shikoku
Kyushu
10
4
Figure
5:
Tim
eSeries
Plot
ofTransportation
Industry
Data
Figure
5:
Tim
eSeries
Plot
ofTransportation
Industry
Data
Time
2000
2005
2010
40100160
Tohoku
Hokkaido
Kanto
Chubu
Time
2000
2005
2010
40100160
Kinki
Chugoku
Shikoku
Kyushu
11
5
Figure 6: Shock on IP based on Similarity Matrix
Figure 6: Shock on IP based on Similarity matrix
0 10 20 30 400
0.5
1
Kanto90% Bootstrap Bound
0 10 20 30 400
1
2
Tohoku90% Bootstrap Bound
0 10 20 30 400
0.5
1
Hokkaido90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Chubu90% Bootstrap Bound
0 10 20 30 400
0.5
1
Kinki90% Bootstrap Bound
0 10 20 30 400
0.5
1
Chugoku90% Bootstrap Bound
0 10 20 30 40−0.5
0
0.5
1
Shikoku90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Kyushu90% Bootstrap Bound
12
6
Figure 7: Shock on IP based on Mutual Buying Matrix
Figure 7: Shock on IP based on Mutual Buying Matrix
0 10 20 30 400
0.5
1
Kanto90% Bootstrap Bound
0 10 20 30 400
1
2
Tohoku90% Bootstrap Bound
0 10 20 30 400
0.5
1
Hokkaido90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Chubu90% Bootstrap Bound
0 10 20 30 400
0.5
1
Kinki90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Chugoku90% Bootstrap Bound
0 10 20 30 40−0.5
0
0.5
1
Shikoku90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Kyushu90% Bootstrap Bound
13
7
Figure 8: Shock on IP based on Contiguity Matrix
Figure 8: Shock on IP based on Contiguity Matrix
0 10 20 30 400
0.5
1
Kanto90% Bootstrap Bound
0 10 20 30 400
1
2
Tohoku90% Bootstrap Bound
0 10 20 30 40−0.5
0
0.5
1
Hokkaido90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Chubu90% Bootstrap Bound
0 10 20 30 400
0.5
1
Kinki90% Bootstrap Bound
0 10 20 30 400
0.5
1
Chugoku90% Bootstrap Bound
0 10 20 30 40−1
−0.5
0
0.5
Shikoku90% Bootstrap Bound
0 10 20 30 400
0.5
1
1.5
Kyushu90% Bootstrap Bound
14
8
Figure 9: GIRF of IP by 1 unit shock on Tohoku
Figure 9: GIRF of IP by 1 unit shock on Tohoku
(a). Similarity Matrix
Tohoku Kanto Hokkaido Kinki Shikoku Chubu Chugoku Kyushu
0
0.5
1
1.5
GIRF
h=0h=3h=5h=10h=20h=50
(b). Mutual Buying Matrix
Tohoku Kanto Chubu Kinki Hokkaido Chugoku Kyushu Shikoku
0
0.5
1
1.5
GIRF
h=0h=3h=5h=10h=20h=50
(c). Contiguity Matrix
Tohoku Kanto Hokkaido Chubu Kinki Chugoku Shikoku Kyushu
−0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
GIRF
h=0h=3h=5h=10h=20h=50
15
9
Figure 10: Shock on IP based on Similarity Matrix(Tohoku)
0 10 20 30 400
2
4
6
90% Bootstrap BoundTohoku
0 10 20 30 400
1
2
90% Bootstrap BoundHokkaido
0 10 20 30 400
1
2
3
90% Bootstrap BoundKanto
0 10 20 30 400
2
4
90% Bootstrap BoundChubu
0 10 20 30 400
1
2
90% Bootstrap BoundKinki
0 10 20 30 400
1
2
3
90% Bootstrap BoundChugoku
0 10 20 30 40−1
0
1
2
90% Bootstrap BoundShinkoku
0 10 20 30 400
1
2
90% Bootstrap BoundKyushu
Shock on IP based on Similarity Matrix Tohoku
10
Figure 11: Shock on IP based on Mutual Buying Matrix(Tohoku)
0 10 20 30 400
2
4
6
90% Bootstrap BoundTohoku
0 10 20 30 400
1
2
90% Bootstrap BoundHokkaido
0 10 20 30 400
1
2
3
90% Bootstrap BoundKanto
0 10 20 30 400
2
4
90% Bootstrap BoundChubu
0 10 20 30 400
1
2
90% Bootstrap BoundKinki
0 10 20 30 400
1
2
3
90% Bootstrap BoundChugoku
0 10 20 30 40−1
0
1
2
90% Bootstrap BoundShinkoku
0 10 20 30 400
1
2
3
90% Bootstrap BoundKyushu
Shock on IP based on Mutual Buying Matrix
11
Figure 12: GIRF of IP by 1 unit shock on Tohoku(Tohoku)
Tohoku Kanto Hokkaido Kinki Shikoku Chubu Chugoku Kyushu0
1
2
3
4(a) Similarity Matrix
h=0h=3h=5h=10h=20h=50
Tohoku Kanto Chubu Kinki Hokkaido Chugoku Kyushu Shikoku0
1
2
3
4(b) Mutual Buying Matrix
h=0h=3h=5h=10h=20h=50
GIRF of IP by 1 unit Shock on Tohoku
12
Figure 13: Shock on IP based on Similarity Matrix (1998 - 2008)
0 10 20 30 40−0.2
0
0.2
90% Bootstrap BoundKanto
0 10 20 30 400
1
2
90% Bootstrap BoundTohoku
0 10 20 30 40−0.5
0
0.5
90% Bootstrap BoundHokkaido
0 10 20 30 40−0.5
0
0.5
90% Bootstrap BoundChubu
0 10 20 30 400
0.5
1
90% Bootstrap BoundKinki
0 10 20 30 400
0.5
1
90% Bootstrap BoundChugoku
0 10 20 30 40−1
−0.5
0
0.5
90% Bootstrap BoundShinkoku
0 10 20 30 40−0.5
0
0.5
90% Bootstrap BoundKyushu
Shock on IP based on Similarity Matrix
13
Figure 14: Shock on IP based on Similarity Matrix (Chubu)
0 10 20 30 400
1
2
90% Bootstrap BoundChubu
0 10 20 30 401
2
3
4
90% Bootstrap BoundTohoku
0 10 20 30 40−1
0
1
2
90% Bootstrap BoundHokkaido
0 10 20 30 400.5
1
1.5
2
90% Bootstrap BoundKanto
0 10 20 30 40−1
0
1
2
90% Bootstrap BoundKinki
0 10 20 30 400
0.5
1
1.5
90% Bootstrap BoundChugoku
0 10 20 30 40−1
0
1
2
90% Bootstrap BoundShinkoku
0 10 20 30 40−1
0
1
2
90% Bootstrap BoundKyushu
Shock on IP based on Similarity Matrix (Chubu)
14
