Teshio, Mashike and the Rumoi subprefecture: Of palisades and Christ signs

Japan’s population in 2030, at 108mn, is 14.0%—nearly 18mn people—lower than in 2000. In 2050, it has declined 32.4%—nearly 41mn people—to 85mn. Japan experienced population growth unprecedented for a developed nation in the half-century to 2000. From about 83mn in 1950, the population grew more than 50%. It will now shrink by approximately the same numerical amount over the next half-century. And that stunning decline is unavoidable because it will result mainly from the demographic profile of people already alive.

Shrinking-population economics: Lessons from Japan, Akihiko Matsutani (2006)

Except for the elderly, the Japanese population will decrease by almost half in the coming 50 years.

Commentary to population projections for Japan, Ryuichi Kaneko et al, The Japanese Journal of Population, Vol. 2 No. 1 (March 2009)

My portal to the Rumoi subprefecture was the little cow town of Horonobe (1980 population 4,253, estimated 2009 population 2,652, projected 2035 population 1,476). The prospect of finding an inn with a room for the night there seemed dim, and I headed on to Teshio (1980 population 6,281, estimated 2009 population 3,669, projected 2035 population 2,013), which at least was deemed important enough to have an inset showing the town center in detail on the page in my roadmap book on which it appeared. No room was to be found at Teshio Onsen hot springs, full of frolicking kids; instead I was directed to Hotel Ichihana on the fringe of town.

I suppose Hotel Ichihana is a species of business hotel: its stern appearance, lonesome location, locked-up and evidently long-dormant dining room, and echoingly empty strip-lit and linoleum-clad corridors led me to immediately dub the sullen proprietor Bates. In the twin room to which I was assigned, the naked mattress of the bed I wouldn’t be sleeping in was smeared with stains that did not bear too close inspection.

In the shriveled remains of the entertainment district in the town center, there were at least a few sounds of life from inside a yakinuku skewered chicken place, Hamanasu (that flower again…) The party of thirtysomething women that had been the source of the merriment soon departed, however, and the customer count was reduced to two. I sipped away at an icy beer while assiduously writing up the day’s notes. The guy at the counter kept glancing my way until he could contain his curiosity no more and insistently beckoned me over.

Minami-san was caning a bottle of shochu rice spirit while being gently chided this Thursday evening by the ever-smiling proprietor, Nagayama-san, about how hard everyone had to work the following day. They made a fine double act. He turned out to be the town’s garbage collector, on his fourth, childless, marriage. Divorced, died, divorced, survived: he was turning into Henry VIII. He was separated from his last wife, who was Chinese and lived in Sapporo. I asked in mock innocence where they had met and brought the house down. As if she could have arrived any other way but mail order. For reasons that long escape me, I was forced into explaining the constitutional arrangements of the UK for the second night in a row. Minami-san vowed that he would become a human bomb and kill dear leader Kim Jong-il but as Nagayama-san noted drily, he’d be unlikely to get an interview. The conversation ranged widely but was streaked with nostalgia, for days of the coal mines, young people, and the express trains, which once upon a time ran from Horonobe straight through to Sapporo on the long-gone Haboro line. The food—Genghis Khan mutton, beansprouts, and onion on a sidedish, garlic fried potatoes, and tsukune minced chicken balls from the freezer on a skewer—was dire but the crack, the crack was ninety that night.

Having had one too many for even these deserted roads, I spent a restless night at Bates Motel and stumbled back to the center of town early the next morning to pick up the car, passing as I trod the deserted sidewalks Japan Ground Self-Defense Force troops—at first in dribs and drabs and then in torrents—wobbling past on an eclectic assortment of bicycles—fold-up commuters, choppers, charinkos, and real clunkers that looked like they had been pensioned off from the streets of Shanghai a decade ago.

If like Japan you only commit about 1% of GDP to defense, there won’t be much left in the piggybank for bicycles, which are not, I suppose, a vital ingredient of contemporary combat. Even so, they couldn’t help but remind me of Dad’s Army. Is Walmington-on-Sea really safe?

My destination that drab and squally day was the little town of Mashike, at the far southern end of the Rumoi subprefecture, chosen for no other reason than that I had noticed it had one of the three remaining branch lines on Hokkaido, wanted to know why, and suspected there was a story behind it.

But first, what is the Rumoi subprefecture? It’s one of the 14 subprefectures into which Hokkaido is divided and the sixth smallest, covering only around 5% of the island. Nevertheless, it’s larger in area than Nagasaki prefecture and all but a handful of English counties, stretching some 155km up the Sea of Japan coast.  It’s the second least populous subprefecture and the least densely populated. After the annexation of Hokkaido, it was one of the first places outside the southeastern “chicken’s neck” to be settled and one of the first to see its extractive industries go into decline: the herring fishery before World War II, forestry in the 1950s, and coal in the 1960s. About all the subprefecture has left is some dairy farms in the north, vegetable and rice cultivation in the center, and fruit farms in the south, with Rumoi City’s sole claim to fame being that it is Japan’s largest importer and processor of herring roe and cod roe.

Not that you can discover much of this from the Internet, at least not the English language side of it: strip out Wikipedia and its interminable mirror sites and—aside from Spike—Google can unearth just five references to “Rumoi subprefecture”: two from Hokkaido Electric Power, two in reference to a minor 2004 earthquake, and one from the subprefecture’s own website. Five. We might as well be dealing with an obscure New Guinea language, the 18th century genealogy of a New England family, or a long departed Amazonian tribe.

I headed south, drinking in the desolation, down the coastal Ororon line, named after a local word for the Common Guillemots that breed on an offshore island, through Enbetsu (1980 population 5,375, estimated 2009 population 3,139, projected 2035 population 1,943), Shosanbetsu (1980 population 2,444, estimated 2009 population 1,459, projected 2035 population 745), Haboro (1980 population 13,254, estimated 2009 population 8,368, projected 2035 population 5,070), Tomamae (1980 population 6,528, estimated 2009 population 3,776, projected 2035 population 2,169), and Obira (1980 population 6,474, estimated 2009 population 3,843, projected 2035 population 2,661).

Even the most casual reader of this series of posts, skimming and dipping across the text to get to the next photo the way a deftly tossed stone skips across the sea, cannot have failed to be detained by a meticulous—and sometimes intrusive—cataloguing of past, present, and projected population data for the third or so of Hokkaido’s municipalities through which I passed. Initially I started adding in the population datapoints half in jest, to draw attention to emptying out of rural Hokkaido, before realizing that, having started, for consistency’s sake, I would have to continue to the end of the trail.

As there are just two more sets of data to go, let’s now pull them all together for a demonstration of the strange death of rural Hokkaido. I’ve arbitrarily deemed Sapporo and the five biggest cities—Hakodate, Asahikawa, Obihiro, Kushiro, and Tomokomae—and their suburbs, as well as Chitose, the home of the main airport, to be urban Hokkaido and the rest to be rural.

Population 1980         2009        2035

Urban Hokkaido 2.97mn    3.53mn    3.04mn

Rural Hokkaido 2.60mn    2.01mn    1.37mn

Total Hokkaido 5.58mn    5.54mn    4.41mn

So in the space of two generations, rural Hokkaido—and this includes a handful of cities that at least started off with well over 100,000 people—will have lost very close to half of its population. The hemorrhaging will be worse in truly rural districts such as the Rumoi subprefecture, which will lose fully 60% of its population over the same period. Here is the long-term population trajectory for the subprefecture.

1920: 73,287

1945: 122,256

1970: 118,629

1980: 89,554

1995: 70,403

2005: 61,488

2035: 35,825

As is readily apparent from the above, the population is already well below its 1920 level and heading back to where it was, I would guess, at the turn of the 19th century.

But it would be a grievous mistake to write off the Rumoi subprefecture as a forgotten and neglected corner of the land, with little relevance to anyone save its inhabitants, because it is emblematic not of Japan’s past but its future. The arc of its population trajectory from 1920 to 2005, although a little more intense, skewed, and compressed, mirrors that through which Japan will go in the century between 1955 and 2055. Allow the luminaries at the National Institute for Population and Social Security Research (NIPSSR), talking about Japan’s total population, to take over for a moment:

Coincidentally, the population size of 89.28mn in 1955––50 years ago––is approximately the same as the total population of 89.93mn in 2055 obtained by the medium-variant fertility (with medium-variant mortality) projection. That is, in the coming 50 years, the Japanese population will revert to the size it had approximately 50 years ago. However, the median age was 23.7 in 1955 but will be 57.8 in 2055, showing that the age structure will be completely different and that the population composition will definitely not go back to how it was.
(Commentary, Kaneko et al)

Naturally, the further out you project, the more room for uncertainty there is: under the NIPSSR’s high-variant fertility and low-variant mortality model, the population falls to only 99.53mn in 2005, while under the low-variant fertility and high-variant mortality model it slumps to 82.40mn. The key variable in the long run is fertility, and even in the high-variant model it only recovers to 1.55 in 2055, against the replacement rate of just under 2.1, from 1.26 in 2005, while in the low-variant model it slides to 1.06 in 2055. I’ll have more to say on fertility later. The further out you project, too, the more likely an exogenous negative population event (though not a positive one) is to occur: war, disease, or climate change, to name but three.

Let’s leave the last words on what the experience of places such as the Rumoi subprefecture augur for the rest of Japan to Professor Matsutani:

An earlier pattern of population of population outflows will thus soften the adverse economic effects of aging and population decline in several prefectures [such as Hokkaido]. Those prefectures are harbingers for the economic tribulations that demographic change will occasion in Japan’s three largest metropolises. (Matsutani)

I hunted high and low in the subprefectural capital, Rumoi (1980 population 36,626, estimated 2009 population 25,459, projected 2035 population 16,525) for lunch, but aside from a McDonalds, a KFC, and a couple of no-name ramen noodle outlets, none of which were offering fare befitting a holiday, could find nothing, and pressed on, as watery sunshine usurped the rain, to Mashike (1980 population 8,319, estimated 2009 population 5,411, projected 2035 population 3,223).

At its 1897 inception, the Rumoi subprefecture was known as the Mashike subprefecture, with Mashike as its administrative seat; the seat was moved to Rumoi, just up the coast, in 1914, and Mashike’s heyday was all but done. It may well be in obeisance to its modest former glory that Mashike is allowed to keep its branch line. Could there have been a better spot from which to muse on the aging of Japan?

By 1998, Japan had become the oldest society on earth, and it is poised to continue aging faster than any other nation.
(Matsutani)

Let me reiterate that, with emphasis added for effect: in 2010, Japan is the oldest society that has ever existed in the history of humanity on the planet, and will remain so, not remotely challenged, for several decades at least.

The line from Rumoi to Mashike is not, technically, a branch line, but rather a 17km extension of the 50km of the Rumoi mainline, completed in 1910, that runs between Rumoi and a place called Fukagawa; by the time the tracks turn up at Mashike, however, its essential branchiness is plain for all to see.

Even if we were to assume that the fertility rate will increase to a much higher level than today, given the current situation, Japan will be unable to avoid having the highest proportion of elderly in the world; and the trend will continue. (Commentary, Kaneko et al)

In 1921, the last spike in the mainline extension southward from Rumoi to Mashike was driven in, and my best, though architecturally uninformed, guess is that this gorgeous if derelict ryokan hotel, Tomitaya, was put up around the same time, a sublime exemplar of Taisho Romanesque and a nonpareil restoration project for the wild and brave at heart.

Japanese society is aging at a pace unprecedented in any large nation. Demographers characterize societies where people older than 65 constitute more than 7% of the population as aging societies, and they refer to societies where the over-65 percentage is greater than 14% as aged societies. By those criteria, Japanese society went from aging to aged in the stunningly brief span of only 24 years. The over-65 component of Japan’s population reached 7% in 1970 and 14% in 1994. (Matsutani)

How stunning is brief? It took Germany 40 years, from around 1930 to around 1970, the UK 47 years, from around 1930 to around 1975, France a mere 115 years, from 1864 to 1979, and the US, having reached ageing status in 1942, is not yet, although very close, to aged.

And the further aging of Japanese society will proceed just as fast. The over-65 percentage of Japanese will reach 28%—double the “aged” demographic threshold—shortly before 2020. Japan’s over-65 surge to 28% of the population, from 14%, will occur in approximately the same quarter-century span as the increase to 14% from 7%. (Matsutani)

With the over-65s already accounting for over 28% of the population by 2006 and climbing steeply, the Rumoi subprefecture is thus more than a decade ahead of the national curve. Yet by the standards of rural Japan it is not particularly aged: ratios of 40% and 50% are not that uncommon, and Nanmoku, a onetime lumber village scarcely two hours from downtown Tokyo, boasted in 2008 an elderly ratio of 57% and an average age of 62.5, making it the most elderly municipality in Japan and perhaps the world. One of its larger constituent hamlets has 200 people, only two of whom are under 65; one of its remoter hamlets has 20 people, all of whom are over 75.

Much as I am loath to credit Japan with any unique attributes at all, I am forced to concede that it has an extraordinarily distinctive demographic profile, one that grows more unlike those of other developed nations every hour of every day. Until the early 1990s, it superficially resembled those of large Western European nations, but the diligent observer would have been able to detect easily enough what was about to unfold.

Japan’s departure from the demographic norm can be traced back to its postwar baby boom, which was nipped in the bud by the charmingly named Eugenics Protection Law of 1948, which in its amended 1949 incarnation made abortion available on demand. While baby booms elsewhere carried on into the late 1950s or early 1960s, Japan’s birthrate had already fallen below the replacement rate as early as 1957 and remained there until 1965, staging a resurgence to around the replacement rate in the decade between 1965 and 1974, the peak of the “echo boom”, and then falling definitively and permanently below it in 1975. Japan finally entered the netherworld of lowest-low fertility (a fertility rate of below 1.5) in 1995 and the fertility rate bottomed out, at least for now, at 1.26 in 2005. It is this long half-century of near or below replacement rate fertility that is partly responsible for the rapidity of the aging of the demographic picture that Japan continues to undergo.

Also responsible are the astonishing gains in postwar life expectancy, which averaged just 52 in 1947 but 72 in 1970, surging by two decades over a span of a mere 23 years, compared to gains of only 2.8 years in the US and 2.3 years in the UK over the same period. Japan now, of course, has the longest overall life expectancy at birth of any state or territory on the planet, at 82.6 years.

The proportion of the elderly is expected to grow from 20.2% as of 2005 to 25.2% in 2013, already accounting for more than one-quarter of the population of Japan at this stage. According to the medium-variant projection, it will then reach 33.7%, or more than one-third of the population, in 2035. It will reach 40.5% by 2055, which means that 1 out of 2.5 persons will be in the aged category in Japan 50 years from now.

Population Projections for Japan: 2006-2055, Ryuichi Kaneko et al, The Japanese Journal of Population, Volume 6 No. 1 [March 2008]

The incontrovertible axiom that life yields no known survivors still holds, however, and not yet having concocted the elixir that forever wards off death, the Japanese remain mortal. So as Japan becomes the world’s first hyper-aged society in the latter years of this decade, its population will decline at an accelerating rate.

The main school building (above) and gymnasium (below) of Mashike elementary school, founded in 1868; the present buildings date back only to 1936. It can accommodate 1,000 pupils but currently has fewer than 200 on its rolls.

未来に向かってはばたけ

exhorts the slogan: “fly to the future”. Mashike’s sole high school stopped recruiting students in March 2009 and will close its doors forever in March 2011.

In case of Japan…the proportion of the child population will fall below 10% (8.6%) in 2050, the proportion of the working-age population will be c50% (51.8%) and the elderly proportion will comprise c40% (39.6%) of the total population, which means that the Japanese population will show the most advanced degree of population aging in the world, combined with an ever-diminishing number of children. (Commentary, Kaneko et al)

I lunched on the freshest of seafood in an ancient converted kura storehouse and found lodging at the Auberge Mashike, which as its name suggests, had definite pretensions, before retracing my steps north along the coast at a slightly more leisured pace.

Shuttered shops in Shosanbetsu, existing somewhere in the penumbra between shadow and sun, life and death.

While Japan’s overall population has only been contracting since 2005, the working-age population began to fall a decade earlier, in 1995, and this has had and will continue to have repercussions for the vibrancy of Japan’s socioeconomy. The demographic rot set in even earlier though. Take it away, professor:

The inevitable aging of Japanese society was a foregone conclusion by the early 1960s. But that demographic change did not begin to affect the labor supply until 1968. In that year, population growth in the 20-39 age range slowed for the first time in the postwar era. And in 1976, the population in that age range began to decline. (Matsutani)

In the sunset, a long-silenced gravel plant south of Shosanbetsu.

Japan’s working-age population has been declining since 1995, and it will continue to decline for the foreseeable future. … Trends in working-age population suggest strongly that Japan will have the slowest economic growth among the large industrialized nations. (Matsutani)

Bunting and lanterns had been strung out for a neigborhood association festival in the center of Haboro, but the enforced gaiety only made the melancholy of the rusting shutters more poignant.

The economically active population is poised to decline to 54.7mn people in 2030, from 67.7mn in 2000—a decline of 13mn people, or 19.2%. … The decline in working-age population between 2000 and 2030, at 27.8%, is nearly double the 14.0% decline in overall population. (Matsutani)

An Elasmosaurus (and it took some research to bring you that) welcomes visitors to the “Romantic Seaside Town” of Obira, home to more than 17,000 folk in 1950 and fewer than 4,000 now.

The coming contraction in the active workforce, together with the swelling ranks of the elderly, will result in a skyrocketing dependency ratio, with more and more senior citizens reliant on fewer and fewer working taxpayers. Youthful societies also have high dependency ratios because of their abundance of children, but while children can—in theory—be raised and educated relatively cheaply, Japan’s oldest old will need hugely expensive medical treatment to help them negotiate their last few years of life.

The child dependency ratio and the old-age dependency ratio added together is referred to as the overall dependency ratio, and this ratio is used to show the degree of support for the entire working-age population. According to the medium-variant projection of birth, the overall dependency ratio is expected to increase to 70.9% in 2030 from 51.3% in 2005, and will eventually reach 95.7% by 2055. (Population, Kaneko et al)

An abandoned café-cum-restaurant with rudimentary bungalows for tourists, Tomamae. The door to the café was ajar and I poked my nose in: it was like the Marie Celeste, with a half-drained bottle of whisky on the sideboard. I suspect what the Japanese call a “yonige”, a flight in the night to evade the pursuit of creditors, to start afresh somewhere new. Very very bad juju around here.

One conceivable, if only partial, solution to the decline in the working population might be to shepherd more women and over-65s into the workforce. But Japan’s female labor participation ratio is not exceptionally low by international standards and coercive policies to get women working would run the risk of depressing the birthrate further. And why would women want to sacrifice the primes of their lives to the serfdom of the firm in a land where the glass ceiling is made of Kevlar—fewer than 2% of corporate executives are female.

Nor would Japan’s recalcitrant retirees be likely to take kindly to efforts to push them back into offices, shops, and factories, and they would no doubt express their displeasure at the ballot box, where thanks to their sheer numbers, their high propensity to vote, and a electoral system rigged in favor of rural areas where they are—relatively—most numerous, they are more than capable of turning a general election. My professor snorts:

No reasonably imaginable increase in the employment of women and the elderly can prevent the economically active population from declining faster than the overall population. … The notion of resolving the nation’s fiscal problems by mobilizing the population is ludicrous—an anachronous echo of the wartime National Mobilization Law. (Matsutani)

The proud breast of the concrete guillemot proclaims, “Welcome to Haboro, Sunset Kingdom”. In more than one way.

In 2005, Japan became the world’s pioneer in shrinking-population economics. It became the first large industrialized nation to experience a population decline as a result of natural causes. (Matsutani)

Surrender to the inevitable could not be postponed indefinitely, and the rollercoaster that had taken Japan’s population from 56mn in 1920, past 100mn in 1967, and up to a peak of 127,787,000 in 2004 reached the apex of the ride and began to teeter on the rails, rocking back and forth: although the population fell by 19,000, a rounding error, in 2005, it rose again in 2006—by a miniscule 2,000—and again in 2007—by just 1,000, before staging the first truly irrevocable fall of the coming half-century or more of contraction in 2008, when the rollercoaster lunged forward and the population declined by 79,000, or more than the entire population of the Rumoi subprefecture. We’re on the way down now, and we’ve got a long way to go.

I put years on my tires in the Rumoi subprefecture, executing emergency stops as something caught my eye, smoking U-turns off gravel to get back to better vantage points, churning up muddy verges to make a break for it in the face of oncoming traffic. I needed a sticker for my rear window that read not “I stop for children”—there were none—but “I stop for rust”. I’d rather have this photo, blown-up, than any Rothko in my living room.

Who hinged the door? Who stacked the aluminum windows? Who piled the tires? Did they know each other? And who chose the yellow, and who the blue?

What a patchwork quilt of corrugation! Look how delicately the embers of rust lick up and down the ridges and furrows, how the windows shed tears, grow beards of rust.

According to the 2005 Population Census, the base year of this projection, the total population of Japan was 127.8mn. Based on the results of the medium-variant projection, the population is expected to enter a long period of depopulation. The population is expected to decrease to about 115.2mn in 2030, fall below 100mn to 99.4mn in 2046, and drop to 89.9mn by 2055. (Population, Kaneko et al)

Part of a near-abandoned hamlet north of Tomamae, and the first sign of the palisade fortifications built around houses and compounds against the fearsome blizzards that pound remorselessly in off the Sea of Japan through the winter.

An absolute decline in population will accompany the aging of Japanese society. Quite simply, the number of people dying will outnumber the number of people being born. … No developed nation has ever experienced a large, long-term decline in population. Japan is thus destined to be a demographic pioneer in the uncharted waters of negative population growth. (Matsutani)

Another forsaken hamlet north of Tomamae, the wind turbines—the Rumoi subprefecture generates about 40% of Hokkaido’s wind power—a chilling summer reminder of the intensity of the winter gales.

Japan is headed for several decades of population decline. That decline will continue long after the end of the sharp decline associated with the passing of the baby boom generation. In no industrialized nation do we find a birthrate higher than the 2.1 needed to maintain population size, and we have no reason to believe that Japan’s birthrate will return to even that level. … As stunningly swift as the change will seem initially, it will accelerate. The rate of population decline will increase annually. Stopgap solutions will soon prove wholly inadequate. (Matsutani)

A cliff-perched hamlet half-submerged by summer grasses, Tomamae. The badlands of the Rumoi subprefecture are not badlands in the way American badlands are, arid, eroded, and impenetrable, but they feel like badlands in a spiritual sense, a land of false promises, dashed hopes, and broken dreams, a land forsaken.

It can be said with a significantly high degree of certainty that the Japanese population will continue to decrease for a large part of the 21st century.
(Commentary, Kaneko et al)

Even newer houses have succumbed: this one, near Tomamae, has been boarded up with weather-withered plywood that gives it some aesthetic affinity with the barn, almost as if it has been colonized in revenge for its upstart effrontery in presuming that new roots could be put down in a dying land.

Japan will lose some 30% of its population in the coming 50 years. Continuing these extrapolations…the population will have dwindled down to 44.59mn in 2105, 100 years in the future, a mere 35% of the population in 2005. The population has never before in Japanese history shown such a constant decreasing trend for such a long period of time, literally making the 21st century a century of depopulation for Japan. (Commentary, Kaneko et al)

Sundown coming, Tomamae: looking south to the 1,500m peaks behind Mashike, which bear fresh dustings of snow even in midsummer.

As a matter of fact, even in the extremely unlikely case where the fertility rate recovers to the population replacement level in 2005 and onward and maintains that level afterward as well, the population will continue decreasing until the 2070s, at which point it will have shrunk to approximately 87% of the original population before stabilizing. Thus, the conclusion is clear: Japan is facing an inevitable long-term population decrease. (Commentary, Kaneko et al)

A storm-savaged barn in the sunset, Tomamae.

50 years from now, the shape of the population pyramid will be transformed into an inverted triangle with a very high center of mass, completely lacking stability—much like a pyramid balancing on its tip. (Commentary, Kaneko et al)

The skeletal remains of a house and boat, Shaguma, between Rumoi and Mashike.

Demographic innocents—I was one until recently—no doubt fondly believe that if only the birthrate could be magically reset to 2.1 or thereabouts, long-term population decline could be averted. But it absolutely cannot be averted, as we have seen, due to the wonder of population momentum: just as Japan’s population grew for a half-century after the fuel of a replacement birthrate was spent, so it will decline for at least the next fifty years.

Depopulation will first be felt most acutely in the countryside: the average age of Japan’s farmers is already over 65 and by 2030 few will be capable of harvesting rice, herding cattle, or picking apples. More than a tenth of Japan’s paddies, fields, and orchards already lie fallow for want of a farmer and that ratio is destined to surge in the coming couple of decades. Already dependent on food imports for 60% of its calorific intake, Japan will grow ever less able to feed itself. Sound the alarm, professor:

Time is short. Japan can ill afford any further delay in preparing for the inevitable. In the next year or two, Japan will enter a phase of irreversible population decline. The time to act is now. (Matsutani)

Population decline, especially of the economically active population, will be compounded by a continuing fall in the number of hours worked per person, overinvestment in and very low returns on capital stock (a perennial Japanese sin), permanently sub-par productivity (“productivity isn’t everything, but in the long run it is almost everything”, as Paul Krugman quipped) due to a cosseted domestic sector, intensifying foreign competition in areas of traditional comparative advantage such as autos and consumer electronics, an ingrained abhorrence of entrepreneurship, and extraordinarily low levels of foreign direct investment.

Haboro Koyo elementary school, with a setting sun mural.

My estimates are that Japanese real GDP grew by a lethargic 0.8% annually in the “lost decade” of the 1990s and by half that, 0.4%, in the first decade of the 21st century. Professional economists, far more knowledgeable than I, put Japan’s current potential GDP growth rate as high as 1.5%; I cannot but help feel that the last two decades of sub-potential growth suggest that it is far lower, and falling, with the Japan on the threshold of turning into the world’s first post-growth society. The professor, for one, is in no doubt:

The wrenching demographic change in store for Japan will do worse than slow the pace of economic growth; it will shrink the nation’s economy. Negative economic growth will become the norm in the nation that until recently set the pace for the industrialized world. (Matsutani)

Haboro Koyo elementary school closed for good in March 2001.

In [our] downside scenario even more Japanese individuals and companies would turn their backs on their own economy. More people would emigrate, eke out a living as “freeters” and lower their labour input. Japanese companies would continue to scale down their domestic capacities and foreigners would have hardly any interest at all in the Japanese market. In this scenario the growth rate of potential GDP would decline to zero by 2020.
Japan 2020 – the decline in trend growth is home-made, Stephan Bergheim et al, Deutsche Bank research, 2006

While there is no sign of mass emigration yet—nor do I expect to see one—in other respects Deutsche Bank’s downside scenario is proving, four years on, to be uncannily accurate.

Reuke elementary school opened in 1884 and closed in March 2006.

A decline in total wages and salaries caused by a decline in available labor diminishes demand and causes the economy to contract. And that is what is about to happen in Japan. (Matsutani)

Reuke station, midway between Rumoi and Mashike, had a stationmaster until 1984. Six trains a day still ply the roundtrip between Rumoi and Mashike.

At present we see hardly any signs of fundamental social and economic-policy changes that might stem or indeed reverse the demographically induced decline in trend GDP growth. (Deutsche Bank, 2006)

Welcome to Rumoi—first in Japan for the production volume and quality of its herring roe. The signboard on the left warns locals to keep a vigilant eye out for the poaching of kelp, sea urchins, abalone, mussels, whelks, and octopus. Coast-anchored seaweed and littoral-hugging mollusks are about all that is left in the marine larder since the collapse of the herring fishery.

A row of palisades turns a hamlet into a fortress against the enemy snow, Tomamae.

As death’s knock can only be delayed, the natural political response to an aging population is to encourage the birthrate higher, and Japan has for the last decade or so had a minister for “low birthrate countermeasures”, usually coupled with a portfolio for sexual equality (or rather “joint participation in society by men and women”); that the post has without exception been held by a woman, often the token woman in the cabinet, is evidence enough of the significance the political elite attach to the issue. The current incumbent, Mizuho Fukushima, whose 1992 authorship of a book titled “Choosing not to give birth: The joy of not having children” (産まない選択 子供を持たない楽しさ) suggests she might not quite be the right person for the task at hand (could you make this stuff up?), is anyway much more passionate about getting those nasty, nasty US troops out of Okinawa—leftist nationalism masquerading as pacificism—than the nuances of pronatal policies. But at this late stage, a rising birthrate could well be a poisoned chalice, wouldn’t you say professor?

Even a miraculous upturn in the birthrate would not turn the [demographic] tide. A surge of infants would not begin to augment the workforce for nearly a quarter-century. And in the meantime, those infants would aggravate Japan’s economic woes. The burgeoning ranks of dependents would depress Japan’s savings rate further and worsen the economic downturn. They would also undermine the financial capacity of the contributors to Japan’s pension system. (Matsutani)

The palisades of Tomamae in close-up: the most forbidding domestic architecture it has ever been my honor to encounter.

Like it or not, though, an upturn in the birthrate is what Japan is getting: having bottomed at 1.26 in 2005, it rose to 1.37 in 2008 before stalling at the same level in 2009. Some demographers are already declaring an end to lowest-low fertility, although no state or territory, having fallen to the lowest of the low, has yet escaped.

Only time will tell if it was the clement economic winds of 2005-2007 or the last reproductive cry of the echo boomers born in the late 1960s and early 1970s whose clocks are ticking ever louder that spurred the uptick, although a near-concurrent and near-uniform turnaround in the fertility rates across the nations of the lowest-low—Germany, Greece, Italy, Spain, and Portugal in Western Europe, much of Eastern Europe, and Japan, Taiwan, South Korea, and Hong Kong in East Asia—suggests that something more profound is afoot. The biological urge to reproduce is hard to suppress for good.

It might just be, however, that despite recent evidence to the contrary, Japan has embarked on a vicious demographic spiral, in which a variety of complex feedback mechanisms set to work: aging results in declining international competitiveness, which results in greater economic hardship at home, which results in a suppressed birthrate; aging results in ballooning fiscal deficits, which in the absence of debt issuance must result in higher taxes or cuts to government spending, which cause economic pain, driving down the birthrate; aging, as the elderly dissave, results in a decline in the pool of domestic savings on which government borrowing is an implied claim, reducing room for fiscal maneuver and resulting in less ability to withstand exogenous shocks; aging further entrenches conservative attitudes to everything from pension reform to immigration, resulting in greater government outlays and smaller government receipts; aging leads the electorate to fear for the future of the pension system, resulting in more saving by the economically active, depressing consumption, which drives manufacturers offshore and raises unemployment, which is strongly correlated with the birthrate. Time will tell.

A colorful palisade in Shaguma, between Rumoi and Mashike. Again with the blue and yellow.

However the birthrate metrics work out in the long term, Japan is bound to be a lonely place to grow up in the future.

The annual number of births in Japan has declined from 2.09mn in 1973 to 1.06mn in 2005. Consequently, the population of children under the age of 15 has decreased from 27mn in the early 1980s to 17.52mn in the population census of 2005. … According to the medium-variant projection, the population size of this age group will fall to 16mn in 2009. The decline will continue, and the population of this age group is expected to fall below 10mn in 2039, eventually decreasing to around 7.52mn by 2055. (Population, Ryuichi Kaneko et al)

A buttressed stockade, somewhere between Rumoi and Mashike.

Aside from birthrate engineering, the other frequently vaunted—although mainly by foreigners—solution to aging and population decline is to open the immigration floodgates and embrace the tired, poor, huddled masses of Asia’s teeming shores. In this context, the United Nations conducted a fascinating, if other-worldly, study in 2000 of what levels of immigration would be needed under various scenarios, with a generous assumption that the Japanese population only falls to 104.9mn by 2050. (Scenarios I and II are slight variants of a no net immigration hypothesis, so we’ll ignore them here).

Scenario III: According to the medium variant projection of the United Nations 1998 Revision, the population of Japan would reach a maximum of 127.5mn in 2005. If Japan wishes to keep the size of its population at the level attained in the year 2005, the country would need 17mn net immigrants up to the year 2050, or an average of 381,000 immigrants per year between 2005 and 2050. By 2050, the immigrants and their descendants would total 22.5mn and comprise 17.7% of the total population of the country.

Scenario IV: In order to keep the size of the working-age population constant at the 1995 level of 87.2mn, Japan would need 33.5mn immigrants from 1995 through 2050. This means an average of 609,000 immigrants are needed per year during this period. Under this scenario, the population of the country is projected to be 150.7mn by 2050. The number of post-1995 immigrants and their descendants would be 46mn, accounting for 30% of the total population in 2050.

Scenario V does not allow the potential support ratio [of the working-age population to the retired-age population] to decrease below the value of 3.0. In order to achieve this, no immigrants would be needed until 2005, and 94.8mn immigrants would be needed between 2005 and 2050, an average of 2.1mn per year during that period. By 2050, out of a total population of 229mn, 124mn, or 54%, would be post-1995 immigrants or their descendants.

Scenario VI: This scenario keeps the ratio of the working-age population to the retired-age population at its 1995 level of 4.8. In order to keep this level of potential support ratio, the country would need 553mn immigrants during 1995 through 2050, or an average of 10mn immigrants per year. Under this scenario, the population of Japan is projected to be 818mn in 2050, and 87% of them would be the post-1995 immigrants and their descendants.

Replacement Migration: Is It a Solution to Declining and Ageing Populations? Population Division, Department of Economic and Social Affairs, United Nations (2000)

A footnote comments without apparent irony:

*Scenario VI is considered to be demographically unrealistic.

Ya think? In fact, even the most milquetoast proposal, Scenario III, in which the population is merely kept constant, results in Japan, historically mightily hostile to immigration, having a foreign-born population of 17.7% in 2050, versus 12.9% in the US in 2005. File it all under “ain’t gonna happen”.

Here’s official Japan on the subject of immigration:

It is not appropriate to consider using foreign workers to cope with labor shortages. (Ministry of Health, Labor and Welfare, 2005)

Everything in contemporary Japanese dealings with the precious few foreigners in their midst—from the alacrity with which manufacturing towns such as Hamamatsu sent Brazilian descendents of Japanese settlers packing as world trade imploded in early 2009, the brouhaha over the first batch of a couple of hundred Indonesian caregivers and nurses who arrived last year and are expected to both work full-time and reach written fluency in Japanese within three years to pass tough exams that fully half of all native speakers fail, on pain of being deported if they flunk them, and the scandalous treatment of the tens of thousands of overseas “trainees”, many of them Chinese, in Japan on a program organized by the Japan International Training Cooperation Organization, who are essentially used as sweatshop labor—absolutely everything points to a hardening, rather than a softening, of attitudes to immigration.

A comical palisade this time, with the front door and downstairs windows replicated meticulously.

Anecdotal evidence from my own life amply confirms the vulnerability of the foreign-born, be they of high or low station: I lost count of the number of farewell parties I attended last year, as contracts were not renewed and visas expired. My firm recently made its first foreigner analyst hire in nigh on a decade. As he said wistfully over a drink one evening, “The problem is the whole industry’s been ethnically cleansed”. Last week I interviewed a job applicant who works at one of the big domestic brokerages. “There are 65 of us foreigners there in a company of 6,500 people”, he related, “and we’re treated at best as a necessary inconvenience”. And as second-class non-citizens, he could have added; while the bulk of his Japanese colleagues will be full-time employees, who are tricky to dismiss, he and his foreign confreres are—with one exception—on one-year contracts. Career opportunities? Puh-leeze…

One advantage of the palisades is that when the last occupant passes on, only a few planks have to be added to the entranceway to seal up the house.

The number of solitary householders of 65 years of age and over will increase by 86%, from 3.78mn to 7.17mn in the coming 25 years. Moreover, looking at solitary householders of 75 years of age and over, the number will swell by a factor of 2.18 from 1.97mn to 4.29mn households, i.e., the number of solitary households of older elderly people in the Japanese population is projected to more than double. (Commentary, Kaneko et al)

Earlier I fingered farming as the first victim of aging and population decline; next to collapse could well be the giant Ponzi scheme of the state pension, where either contributions need to be doubled or benefits halved to maintain some sort of equilibrium. Over to you, professor:

The all-too-rapid pace of Japan’s aging is also the villain in the nation’s pension system drama. … As things stand, the number of people who pay into the system will decline sharply even as the number of people who receive benefits increases rapidly. Something has got to give soon and in a big way. (Matsutani)

Even the vegetable plots get their own windbreaks.

[At constant contributions and benefits] the funding shortfall in Japan’s social insurance programs will swell annually, to Y33.4trn in 2010 and to Y57trn in 2030, from Y8.7trn in 1998. Contributions would cover only 44.0% of expenditures under the programs in 2030. The cumulative shortfall for the years 1998 to 2030 would total Y1,240trn. To put that total in perspective, it is roughly equivalent to aggregate personal financial assets in Japan today. (Matsutani)

Occasionally the palisades are lopped to more modest, fence-like dimensions.

Being told that pension benefits might decline 10% or 20% during your lifetime is one thing. Hearing that they will decline 50% or more is something else again. (Matsutani)

Sometimes the palisades are removed in summer: here the giant staves are the only year-round feature.

As the pension system collapses, so will the infrastructural fabric of Japan. First to go will have to be investment in new infrastructure:

Japan now has a well-developed infrastructure…and the incremental benefits of new airports and highways are minimal in terms of increasing productive capacity. … An airport or a highway is visible, but the mechanism by which increased public works spending diminishes the economy is invisible, and few people have a grasp of that mechanism. (Matsutani)

The remnants of a truss bridge on the Haboro line, which saw its last train run in 1987.

The professor has done his math:

The amount of public works spending that will be possible in 2030 without compromising economic growth is Y14.2trn. That is down 47.0% from the actual figure of Y26.8trn in 2002. Public works spending needs to decline by nearly one-half. (Matsutani)

A Haboro line tunnel, north of Shosanbetsu.

As public works spending declines, more and more of the available budget will be taken up by infrastructure maintenance and repairs, with spending on refurbishment and replacement of existing infrastructure exceeding the threshold level for total public works spending as soon as 2023, calculates the professor:

Japan’s infrastructure will thus begin to deteriorate, and some of it will fall into ruin. (Matsutani)

A Haboro line viaduct, again north of Shosanbetsu.

One counterintuitive consequence of aging and population decline is that Tokyo and other big cities will see their economies wither and their vitality drain away. The most rural of the hinterlands have already done their aging, and will have demographic profiles in 2030 not dissimilar to the ones they had in 2000. Tokyo, on the other hand, with its huge bulge of twenty- and thirty-somethings, will see its demographic profile converge with those of the provinces; even if people aged 20-39 continue to move to Tokyo at their recent rate, there will be a third fewer people in the cohort in 2030 versus 2000, so they will have a much reduced impact. Meanwhile, although the absolute percentage decline in the working-age population in Tokyo will be only around the national average, the urban workforce will age more rapidly than anywhere else and by 2030 account for a smaller percentage of the overall population than any other prefecture in Japan. One last time, then, professor, if you insist:

Japan’s metropolises face even bigger issues than the nation’s nonmetropolitan areas do in regard to economic growth rates and per capita income. … Per capita income in Japan is likely to change little between 2000 and 2030…but it will decline sharply in the nation’s metropolises. And that will mean a decline in living standards there. (Matsutani)

As I snapped away at the palisades, my attention turned to little metal placards which bear Christian messages, with their familiar white and yellow (for emphasis) fonts on a black background, discreetly placed on the walls of houses, sheds, and even palisades. They are to be found all over rural Japan, but in their apocalyptic menace they seemed perfectly suited to the Rumoi subprefecture. This one reads “God can see your heart”, which it attributes to the Bible.

“Jesus Christ is the Son of God”. This is unattributed. The placards are known as “Christ signs” (キリスト看板) and are the handiwork of the Bible Distribution Cooperation Society (聖書配布協力会), which was founded in 1952—noone knows by whom—and which is headquartered in the tiny town of Marumori in the wilds of the border between Miyagi and Fukushima prefectures.

“God punishes sins, the Bible”. It’s all getting a bit too Old Testament for my taste. The society is purportedly a loose network of like-minded individuals and has no hierarchy or leadership. The consent of the property owner is solicited before a sign is put up; once up, it may endure for decades. No money changes hands, and the property owners are unlikely to be Christian; some welcome the signs in the belief that they ward off thieves, some in the belief that they improve the moral tone of the neighborhood.

“Jesus Christ brings eternal hope”. The society is conservative evangelical in outlook, with its foes the usual suspects: atheists, materialists, defenders of evolution, idolaters, abortionists, homosexualists, adulterers, pornographers, Roman Catholics, and Freemasons. Hey, those are some of my best friends you’re talking about!

There must be some structure to the society, though, because it runs a school in its hometown of Marumori, a school renowned for its enthusiasm for corporal punishment, which it justifies—of course—by appeal to the Bible: “Foolishness is bound in the heart of a child; but the rod of correction shall drive it far from him”.  (Proverbs 22:15)

By 8pm, only one party lingered that evening at the immaculate blue and white tablecloths of the Auberge Mashike as the last navies of the sky were vanquished by the night and the waves of the Sea of Japan fell softly against a rocky shore where noone strolled.

On the left, the father and mother; on the right, farside, their two unmarried daughters, one obscured; nearside center, a friend of the father, perhaps a suitor for one of the daughters; and nearside right, a friend of the daughters, brought along for moral support.

The assembled company, plainly from lands far to the south, exuded good breeding and old money. In another country—mine—they would have been upstanding members of the horsey squirearchy; here I found it hard to place them. They had nothing directly to do with anything as grubby as money or commerce, it might have been land.

My code of photographic ethics would normally have prohibited me from taking this surreptitious shot, but the father started prattling on about how Kenyan environmentalist and 2004 Nobel Peace Prize winner Wangari Maathai had determined that the Japanese expression “mottainai” (“what a waste”) had no equivalent in English and had adopted it in her campaigning, which was so objectionable on so many levels that I, in turn, determined that he, and by extension everyone else, had sacrificed their right not to be photographed.

The princess daughters’ well-groomed hands covered their mouths with every forkful of food and every burst of titters; one apologized deeply for interpolating Maathai’s name when her father simply referred to her as “that Kenyan woman”. How drearily soul-sapping life must be as a fortysomething spinster scion of the elite!

I took a commemorative photo of my last Hokkaido supper: a book of maps, a notebook, a packet of smokes, and a bottle of wine. What more does a rover need?

So whither Japan? The key question is whether it can maintain GDP per capita, or whichever proxy for that you might prefer, at a level that affords most of its citizens bearable, even intermittently enjoyable, lives, as the economy contacts. Professor Matsutani is optimistic that it can. To find out why, I heartily recommend you read his book. I’m not convinced.

The economists of Deutsche Bank offer the following 2020 prescription for Japan’s ills—for what it’s worth, as noone in Japan will be listening to what a bunch of ignoramus foreigners are saying, and no foreigners in their target audience are in the slightest bit interested Japan any longer. My comments, which generally ignore the utility and only consider the plausibility of the measures, are in parentheses:

Japan would need to integrate the female half of its population more actively into working life [didn’t happen in the last decade, won’t happen in the next] and increase the proportion of economically active older people [some progress possible, though not necessarily a positive, as the elderly are less productive and more conservative than the young] again. A managed immigration policy could limit the decline of the population [unthinkable]. In this scenario the innovation system would be altered in such a way as to enable it to produce more radical and path-creating innovations and be internationally networked [extremely unlikely, as the only plausible new entrants in large numbers to the workforce are the elderly]. Fledgling companies could draw on modern capital market instruments for their financing [won’t happen, not with the domestic securities companies under the thumb of the banks]. The education sector would also be given a deliberate international slant with English as the language of instruction in more universities [incremental progress conceivable] and more foreign providers in Japan [implausible in the extreme]. And more branches of industry, particularly in the services sector, would be made fit for competition with foreign providers [doubtful]. Foreign direct investment in Japan would surge [you must be kidding]. Reconciliation with China and Korea [possible, but not with any particularly positive economic consequences for Japan] would pave the way towards an Asian economic union [fairy-tale stuff, they’re all beggar-thy-neighbor mercantilists] and give Japan’s share of foreign trade an enormous boost [all well and good, except the Koreans and the Chinese are desperate to sideline the old enemy any way they can].

I made good enough time through indifferent weather on the 600km hike south to the ferry to stop for lunch outside Hakodate at a burger joint popular with courting couples called Lucky Pierrot.

I do hope he shares his good fortune liberally around.

The skies were mostly stormy grey, but ships anchored outside the port were bathed in shafts of heaven-sent Old Master sunlight; through the binoculars, my last sight of Hokkaido, beneath the crags of Mount Hakodate, was of a house with planks nailed across the eyes of its windows and the mouth of its door.

I pounded down and down Honshu until I could take it no more and sought shelter at a ryokan inn in Ninohe, Iwate (you know the vital statistics drill by now: 37,537, 29,714, 20,388), whose innkeeper was 90 if she was a day. I made it back home for a late lunch the following day, smiling smugly at the Tokyo traffic, 5,500km or so after setting out.

I wouldn’t recommend Hokkaido as a destination for the casual tourist: as we’ve seen, the coast has been beaten black and blue with an ugly stick, much of the inland bears the nastiest scars of extraction and construction, the weather, even in summer’s sweetest months, can turn vicious quicker than a Rottweiler, the natives are by no means always friendly, and much of it lies in ruins. Me? I loved it.

My Hokkaido epitaph was written for me by the hoarding on a pachinko parlor, New Japan (same as the old Japan), in Monbetsu, of all places:

Welcome to the best place
Where makes you happy
It gives you such story

Such story indeed.