At the Digital Watershed: Terrestrial Television Broadcasting in Japan

The switch to digital terrestrial broadcasting on 24 July 2011 marked a watershed for the broadcasting industry in Japan. Digitalisation is the single largest industry-wide event since the advent of alternative distribution technologies, satellite and cable, in the 1980s. Preparation for the switch to digital, known as chideji-ka, has put existing business arrangements under pressure and has led to a renewed focus on the future shape of the industry. There is increasing acknowledgement that change, especially in the relationship between central and local broadcasters, is inevitable. This paper summarises the position of the industry at the beginning of its digital age, arguing for a new view of broadcasting in Japan that recognises the two-tier reality behind industry rhetoric. It also summarises the major options open to the industry as it looks to redefine itself in a much-changed media environment.


Introduction
Television broadcasting in Japan, in terms of its industrial structures, legislative fundamentals, financial underpinnings and major players, has continued, barring the introduction of satellite distribution in 1989, without major upheavals between its inception in 1953 and the 2001 decision to switch to digital. 1 Media firms have enjoyed sustained stability; only one company of the 127 that now operate has come close to going out of business, 2 and there have been no company mergers or takeovers since the very early years.
Apart from satellite, the state of broadcasting is very much as it was in the early 1960s. Communications technology, on the other hand, has changed beyond recognition. New distribution possibilities have thrown the broadcast industry's problems into stark relief, the external world has overtaken the television business, and 'broadcasters must evolve or die '. 3 The summer 2011 switch from analog to digital terrestrial television (DTTV) 4 broadcasting -a rare and significant industry-wide event -is an appropriate point from which to sum up the situation of the commercial television broadcasting industry in Japan and to address the changes that may take place in the coming years.

Sources
Recent English-language works looking at Japan's mass media from a business perspective are rare. The standard introductory text is outdated considering recent changes in technology, 5 and while Valaskivi's review updates much of the above, as a research report it lacks the depth of a book-length work. 6 The edited volume Media and Politics in Japan is also showing signs of age, much of the content looking back to the upheavals of the 1993 general election, though Westney's chapter on media businesses covers valuable background and highlights many issues which are still relevant. 7 More recently, Kwak's work on satellite broadcasting and regulation offers a clear explanation of the regulatory tangle that governs broadcasting in Japan. 8 More works are available in Japanese. Industry handbooks, often aimed at students looking for a career in the media, offer factual overviews with minimal interpretation; 9 industry yearbooks and handbooks, such as those produced by Minkan H os o Renmei (Minp oren, the Japanese commercial broadcasters' industry association) and national broadcaster Nihon H os o Ky okai (NHK), are also useful data sources and present the industry view of developments. However, for this reason they cannot always be relied on as a basis for interpretation.
Recent writing on the television industry has naturally concentrated on digitalisation, with works either extolling the (potential) virtues of the digital society or offering sensationalist portrayals of a nation of blacked-out television screens as the digital promise fails. Although the move to digital, which started in 2001, has already had some material effects on the broadcast industry, it is still early for academic analysis, though a small number of works do exist which offer an industry-wide media-as-business approach. 10 Throughout this study I refer to sources sometimes justifiably deemed less than reliable, such as the sh ukanshi weekly magazines Friday and Sh ukan Gendai. The nature of the mass media in Japan, the integration of the large television and newspaper companies and their reliance on government sources for information, tend to mean that critical reporting on the media in the mainstream press is uncommon, thus my reliance on the less inhibited magazine and industry press.

Varieties of Television Broadcasting
This section outlines the various forms of television available to viewers in Japan. My focus is terrestrial broadcasters, as the dominant suppliers of television programming and the main delivery system; alternative delivery methods are mentioned here only insofar as they have an impact on their conduct.

Terrestrial Broadcasting
Terrestrial broadcasting can be viewed as split into two fundamentally different types of business: . Larger stations based in Tokyo (known as 'key stations ', kii-kyoku), Osaka ('sub-keystations', jun-kii-kyoku) and Nagoya; . Smaller local broadcasters outside these areas, known as keiretsu-kyoku. 11 In licensing and coverage terms there are three wide broadcast areas (WBAs) -Kant o (around Tokyo), Kinki (based on Osaka) and Ch uky o (around Nagoya) -and 29 smaller areas, mostly congruent with Japan's prefectures. 12 In this study, while I treat the broadcast industry as divided into two tiers, a first tier of firms within the WBAs and a second tier of smaller non-WBA firms, regulatory structures do not make this formal distinction. These tiers are linked in 'informal' networks which grew out of news exchange agreements initiated in the late 1950s. Network arrangements have gradually led to mutual reliance, of local broadcasters on key stations for programming and revenues, and of key stations on local broadcasters for distribution. 13 Over the last two decades this latter dependency has been eroded by changes in distribution technologies; the former, however, has continued intact. The resulting imbalance has led to questions about the future of the broadcasting system in its current form, especially the role of the second tier. The combination of the economic effects of investment in digitalisation and falls in advertising revenues since the 2008 world-wide economic downturn has accentuated tensions between the tiers and highlighted the shortcomings of a system in need of renewal.

Network Infrastructure
Broadcasters are arrayed in networks materialised through program, news and information sharing agreements; these consist of a number of hierarchically arranged stations with a Tokyo station at the hub (see Table 1). There are four genuinely national networks, JNN, NNN, FNN and ANN; the fifth, TXN, consists of six stations strategically placed to maximise coverage. Each network has ties of various kinds and strengths with one of Japan's large newspapers. Signal distribution is via some 2300 installations housing relays for each station, some 11,000 pieces of transmitting equipment. 14 On 24 July 2011 the analog television service distributed via this 11 Terrestrial broadcasters are represented at national level by their industry association Minp oren, known in English as The National Association of Commercial Broadcasters in Japan (NAB). To avoid confusion with the parallel US organisation, also the NAB, I refer to this organisation as Minp oren throughout. equipment since 1953 was replaced through nearly the whole of Japan by a digital equivalent. 15 A number of cross-network stations, taking programming from more than one network, exist in the less well provisioned areas. They have become rarer since the last round of company formation in 1991-97, having been replaced by these networkaffiliated 'New Heisei Stations' (Heisei shin-kyoku). 16

Geographical Licensing
Broadcasters are licensed by the Ministry of Internal Affairs and Communications (MIC) on the basis of geographical areas, limited (theoretically) by either prefectural or other administrative boundaries. Reality is more complex; television signals do not stop at administrative boundaries and terrestrial broadcasters' signals are often redistributed outside the licence area by cable operators. With many viewers accustomed to the internet as an effectively boundary-less distribution channel, the current geographic/ administrative-based licensing scheme for television looks outdated.

Alternative Distribution: Satellite, Cable and the Internet
At the end of the analog era there were four types of satellite broadcasting in Japan: analog and digital services distributed by broadcasting satellite (BS); and digital signals distributed by two different communications satellites (CS). 17 All these services require appropriate receiving and decoding equipment. BS ceased analog broadcasting (12 channels) at the same time as terrestrial television, but the number of digital channels on offer had increased to 31 by August 2012. 18 In March 2010 27.5 million households, just over half of all households in Japan, were equipped to receive BS broadcasts. One of these satellites is in the same position as the BS satellite, thus subscribers can receive both BS and CS broadcasts on one dish, though they still need separate tuners.
The complex history of CS broadcasting reaching back to 1990 has resulted in the emergence of a dominant player, Sky Perfect JSAT (SPTVJ), an amalgam of Sky TV, PerfecTV and JSAT. As of August 2011, 3.77 million subscribers took at least one of its four services. 19 The business and licensing structures legislated to manage CS broadcasting created two separate categories of firm, program supply (itaku h os o jigy osha, consignor broadcasters) and program distribution (jutaku h os o jigy osha, consignee broadcasters). In this respect it is fundamentally different from terrestrial broadcasting where 'vertical integration' is the norm. Relations between these two types of business and their customers are managed by another type of business, the 'platform'. However, the exact role of the 'platform' company (SPTVJ) whilst key to the operation of CS services, is legally ambiguous. 20 All SPTVJ-provided channels are subscription, either bundled or as individual channels; four separate services provide access to up to 350 channels. Channel content is a mixture of international brands (Discovery, Disney, CNN, Fox), content specifically oriented towards Japanese viewers (e.g. Takarazuka Sky Stage, a jidaigeki samurai drama channel) and general channels provided by the major terrestrial broadcasters.
Despite initial hopes that satellite broadcasting might encourage participant diversity, BS was, until the expansion of channels possible after the switch to digital in July 2011, dominated by the big terrestrial broadcasters, which for commercial reasons are unenthusiastic about splitting the advertising pie further. On the CS side, SPTVJ provides access to a broader range of channels though viewership is small by comparison with terrestrial services.
At the end of 2010 Japan's cable television companies (530 large-scale, 45,000 'village-based', regulated by the 1972 Cable Television Broadcast Law) provided services for 32.6 million households. 21 Of these, roughly 25 million receive, as well as terrestrial channels, programming produced by the cable operator, often local news and information programs. Just under 8 million use cable solely to receive terrestrial broadcasts. Many operators are offering a continuation of analog services (supported by public spending) after 24 July 2011 by providing customers with an analog signal backconverted from digital. This will allow households to carry on using older non-digital receivers until this service is discontinued in spring 2015.
Net distribution, both simulcasting and video-on-demand-style program replay, has so far failed to make an impact in Japan. A number of causes have been cited: . Television, whether license fee-funded PBS or advertising-funded commercial, has been free at time of viewing, so audiences are resistant to pay-per-view systems; . Traditionally vague handling of copyright means that re-airing of back-catalog programs can be difficult in the face of protests from program participants; . For commercial channels, coming to arrangements with the original sponsors/ funders for recycling of their programs has proved problematic. 22 Media commentator Ikeda Nobuo describes broadcasters in Japan as actively attempting to obstruct the use of alternative, specifically internet, distribution of television programs. This he attributes to a determination to preserve the value of 19 SPTVJ, 'News Release: 2011nen 8gatsumatsu genzai'. 20 Kwak, 'Restructuring the Satellite Television Industry ', 73. 21 Data for this section is taken from MIC, 'K eburu terebi no genj o'. 22 'Terebi gy okai: 10 no gimon'. investments in physical distribution infrastructure made since the 1950s. 23 Certainly Hirose Michisada, Minp oren Chairman and CEO, has made the industry position regarding internet distribution very clear: 'We already have an infrastructure for free delivery of television programs across more or less the whole country, it's called terrestrial broadcasting, so we are not considering adding net simulcasting to this.' 24

Digitalisation
The benefits, social and commercial, of integration between television broadcasting and IT/computing that digitalisation seems to offer have been discussed, and enthused over, since the late 1980s. 25 However, the power balance between broadcasting and computing in the intervening years has shifted. In an environment where a computer that shows television programs (and does all the other things a computer is expected to do) is a reality, television has to rethink its place.
The question of the future role of broadcast television is in many sets of hands, including those of station executives, program-makers, advertisers with a choice of competing media, and, perhaps primarily, administrators, whose control of the legal structures is pivotal. The process of interaction between these parties will be complex, and with the added unknowns of developments in technology and trends in media use, any attempt to predict the mid-to long-term for the television industry is rash. This section is therefore limited to summarising forecasts made for the direct financial effects of Japan's introduction of DTTV.
Government estimates that the total economic impact over the two decades 2001-2021 will be some ¥249 trillion. 26 This includes the effects of the actual shift to digital technology for broadcasters and consumers, estimated effects on other related services industries, DVD manufacturers, internet and mobile distributors, data broadcasters, and finally, the broader effects of the freeing up of spectrum space. However, benefits will not be distributed evenly across all interested parties. While there are evident and immediate gains for some industries, others must feel themselves manifest losers. Potential large-scale beneficiaries of the move to digital will include: Mobile telecoms operators who will gain access (potentially) to the spectrum freed up by reducing broadcasting use. This is vital for mobile operators if they are to be able to sell the next generation of 4G services, dataoriented and bandwidth-hungry, to consumers; Manufacturers of electrical goods: the four largest domestic home electronics companies (Sharp, Sony, Panasonic and Toshiba), with near total dominance of the domestic market (and major exporters), have seen demand for their products soar since 2005. In 2000-06, yearly sales of television sets were just under 8.9 million units; in 2009 the figure was 13.6 million and in 2010, 25.2 million. Sales for the months of the first half of 2011 were up yet 23 Ikeda, 'Rieki o umanai chideji-ka', 18. 24 'Tomaranai ''make no rensa '''. Some companies do, however, make their broadcasts freely available online, at a very reduced audio and video quality, via the KeyHoleTV P2P service; see KeyHoleTV website: www.v2p.jp/video/index.html (accessed 11 May 2011). 25 Brinkley, Defining Vision, Sec.2. 26 'Chij o dejitaru h os o e no ik o ni tomonau keizai-k oka nado ', 55. again on the previous year. 27 While it is hard to separate out and quantify the specific effects of digitalisation on their overall performance, this stimulus to domestic demand may have helped insulate companies at a time when worldwide economic downturn has threatened overseas markets.
Some gains may filter through to broadcasting by way of an increased willingness to advertise on television. Both mobile telecoms and electronics manufacturing are largescale advertisers; the telecommunications sector, which includes mobile operators, is the nation's fifth largest advertiser by spend (¥203 billion, 7.2 percent of national annual total in 2009), and two-thirds of this goes into television advertising. 28 On the other hand, broadcasters and viewers must have an ambivalent view of the value of digitalisation. Neither has much initially to gain from digitalisation; they will both make significant investment and get in return (technically) better audio and video quality. It is decidedly unlikely that broadcasters will be able to command a better price for advertising broadcast in HD given that it is the de facto DTTV standard. Viewers who have invested in a new receiver, and possibly also an antenna, will, for that price, be able to carry on watching television, albeit less will have been spent on producing the content they have available as broadcasters are forced to divert funds from programmaking to spending on new digital infrastructure.
Broadcasting, while it has the potential to influence public opinion, thus allowing it to punch above its weight politically, is relatively insignificant in economic terms; estimates of the broadcasting market size are in the region of ¥3.9 trillion, one-tenth that of electronic manufacturing. 29 Any one of the electronics firms mentioned above employs more full-time workers than the whole of the broadcasting industry put together. 30 In decisions on DTTV policy, the broad financial interests of the electronics sector seem to have trumped the latent trouble-making capacity of a disgruntled broadcasting industry.

Public Spending
Going digital has involved significant costs; total government spending, on support and call centres, provision of grants for upgrade and renewal of reception equipment, provision of digital tuners to low-income households and financial support for broadcasters, is expected to be in the region of ¥66 billion in FY2012 alone, 31 down on the previous year's ¥87 billion. 32 While not exactly giving sets away free, central government has committed considerable resources to ensuring the number of 'radio-wave refugees' is minimal. 33 The FY2010 supplementary budget includes funding for the purchase of digital receivers by public offices, hospitals and schools (¥150 billion) and additional funding (¥75 billion) for the 'Eco-Point' scheme, originally intended to promote energy-efficient refrigerators and air-conditioners, to extend it to digital televisions. In total, combining 27 Japan Electronics and Information Technology Industries Association (JEITA) shipments data: www.jeita.or.jp/english/stat/shipment/2011/ (accessed 3 October 2011).  'Viewers mustn't end up as ''radio-wave refugees ' ''. spending from the main and supplementary budgets, measures to ensure progress toward a trouble-free switch-over in summer 2011 came to ¥420 billion. 34 In 2010 DTTV was reported to have achieved 98 percent coverage, but during this process a proportion of areas have been designated 'new difficult-viewing areas' (arata na nanshi chiiki); these approximately 3000 locations contain roughly a third of a million households. 35 Measures to deal with signal provision to such places -setting up additional relays, shared antennae or cable facilities -mean more spending as subsidy applications come in.
There may also be gains for the public purse. In February 2011 an MIC committee was formed to consider how bandwidth freed up by the switch from analog to DTTV, already earmarked for 'multimedia broadcasting' and mobile use, might be auctioned off to interested parties. 36 Of course the results of any auction depend on how bidders assess the value of the product; in this case, how mobile operators assess the gains to be made from 4G services. In December 2010 mobile operator DoCoMo commenced provision of a 4G service (named Xi, pronounced 'Crossy') in urban areas across Japan, 37 and the take-up of this bellwether service, which reached five million subscribers on 22 August 2012, 38 is likely to influence industry perceptions of the value of bandwidth at auction.
Costs to the public purse might be deemed negligible by comparison to the projected economy-wide long-term gains; however, they are real whereas the benefits are largely still speculative. Furthermore, whilst broadcasters have been forced to invest in renewing their entire infrastructure, they stand to gain little if anything from digitalisation; it is unlikely advertisers will pay more to have their commercials distributed digitally, or indeed to have them distributed in HD rather than SD.

Costs to Broadcasters
For broadcasting, digitalisation has meant a significant investment and the replacement of equipment throughout the production process: the renewal of innumerable cameras, video-routers, transmitters, editing machines, satellite trucks, etc. This is on top of the renewal of the entire broadcast infrastructure required to distribute these digital signals. In 2010 NHK expected to invest a total of ¥400 billion in its move to digital over the period 2011-13. The total for all commercial broadcasters was ¥1.04 trillion, and for second-tier companies average investment was expected to be approximately ¥5.4 billion per company. In 2006 these same companies averaged a yearly profit of ¥0.35 billion. 39 Even taking into account the extended ten-year period over which investments have been made, this scale of investment cannot but impact the financial health of the majority of broadcasters.
The majority of investments are now made and the drop in production budgets, predicted by many in the industry, do indeed seem to have materialised; 40 the situation 34 'Hosei-yosan de k og u sareru terebi no chideji-ka'. has been exacerbated by decreases in advertising revenues after the post-2008 worldwide financial downturn.

Costs to the Audience: Receivers and Consumers
For viewers digitalisation primarily means the replacement of analog television sets, roughly one per person in Japan, or the purchase of digital decoders which attach to their current analog sets. Many households may also require a new antenna. For those living in housing with a shared antenna (roughly a quarter of all households) this generally means paying a share of the replacement costs. 41 A March 2010 MIC survey of digital receiver penetration showed a range of values for Japan's prefectures, from Toyama, Saitama and Gifu, all around 88 percent, to the lower levels found in Okinawa (65.9), Iwate (66.7) and Nagasaki (72.9). Mean penetration was 82.4 percent. Typically then, 14 months before the end of analog television services, nearly one household in five was without digital receiving equipment. By September, surveyed penetration had increased to just over 90 percent; 70 percent of those surveyed as to why they had not yet purchased digital receiving equipment gave reasons amounting to 'there's still plenty of time left', but just under 40 percent blamed the price of a new set. 42 However, MIC's survey methodology, and thus also the optimistic conclusion it draws, have been repeatedly called into question. The survey tends to reflect the position of those households willing to talk about their move to digital; and, more critically, householders aged over 80 are excluded. According to a 2009 report from the Nomura Research Institute, the switch to digital may lead to a loss of 10 percent of current viewers; this 10 percent have little intention of 'going digital', at least in terms of buying a new television set or digital tuner. 43 Kawamura Makiko of Shufuren (the Housewives' Union), a member of an MIC digital take-up related consultative committee, estimated that the true level of digital viewership was closer to 60 percent. 44 Roughly one-quarter of those households who cannot currently receive digital television, either because signals do not reach them or they do not have an appropriate set, said they did not intend to make arrangements to watch digital television as they 'don't watch that much'. It has also become clear that the penetration rate of digital sets varies greatly with household income. Policies are in place to provide financial assistance in purchasing new sets, or digital tuners that connect to analog sets, to low income households (perhaps confirming the cynical observation that 'if people didn't buy TV sets, the government would give them away for free' 45 ). Take-up on this scheme has been slow; of the 2.7 million eligible households, 0.63 million applied in 2009, and the first half of 2010 saw just 0.22 million applications. 46 The necessary corollary of the mass purchase of new sets is the disposal of roughly the same number of old cathode-ray tube (CRT) sets; for 2011 alone, the Japan Electronics and Information Technology Industries Association (JEITA) projected the need to 41 Kimura, 'Dai-30kai NHK jushin jittai ch osa'. 42 'Topics: Results of Terrestrial Digital Broadcasting Penetrance Survey' (multiple answers recorded). 43 'Tomaranai ''make no rensa ''', 46. 44 Kamei, 'Chideji no detarame. Dai-2kai'. 45 Barnhurst and Nerone, The Form of News, 25. 46 Kamei, 'Chideji no detarame. Dai-1kai '. recycle approximately 13.2 million sets. 47 However, domestic production of CRT sets, which had previously taken up much of the recyclable waste from television set disposal, has now ceased and given that the glass used in CRTs contains lead, the disposal of such large amounts may be problematic. Since 1998 the costs of recycling have been offset by charges paid by the owner, along with manufacturers and vendors, under the Household Electricals Recycling Law. 48 A reluctance to pay these charges (in addition to the cost of purchasing a new digital-compatible set) may worsen the rate of illegal disposal of CRT sets already visible in data for 2007. 49 Key Stations: First-tier Broadcasting

Changes in Key Station Finances
The majority of commercial broadcasters' income in Japan, 75.9 percent in 2009, 50 comes from selling advertising, or, perhaps more accurately, selling audiences to advertisers. This section looks at recent trends and the effects on broadcasters' finances.
Since 2005, television's share of total national advertising spend has stayed roughly constant at between 28 and 30 percent, but in cash terms it has declined from ¥2.04 trillion to just above ¥1.71 trillion in 2009. 51 Traditionally advertising income has made up roughly 80 percent of broadcasters' income but recent declines have meant they have had to look to other areas of their business to cover costs. In 2009 TBS Holdings' income from its property business brought in almost as much as domestic program sales. 52 Television advertising in Japan comes in two types; 'time' (taimu, also known as 'program advertising', bangumi CM) and 'spot' advertising. Time advertising is effectively program sponsorship and involves the sponsor in funding production and distribution of a program which acts as a vessel for their advertising. Spot advertising goes in the gaps between programs. A time advertiser is locked into a particular financial commitment and will pay significant sums up front. This can involve a degree of risk as new programming may not gain sufficient ratings to be viable. The typical ratio between spot and time income for the key stations is roughly 70:30. 53 Between 2009 and 2010 time advertising revenues for the key stations, which account for over half of all television advertising sales, fell by 15.4 percent, the largest single-year fall since FY1993's 5.2 percent fall following the collapse of the bubble economy. Spot advertising fell by 10.6 percent. 54 Industry opinion is that television advertising overall is undergoing a process of 'spot-isation', perhaps accounting for the disparity in rate of decrease in the two types. 55  Kimura, 'Chideji-ka to uriagegen de'. 55 Kimura, 'Gy oseki-kaifuku ni tenjita minp o-terebi ', 46. fewer businesses are prepared to take on the medium-term commitments which come with time advertising. Time contracts are also tending to shorten; whereas previously a six-month contract period was the industry standard, now periods as short as one month are beginning to appear. This change has apparently also been fuelled by the mandatory quarterly reporting requirement introduced in 2008. 56 Overall this has meant a move toward more spending on spot advertising, which allows advertisers more budgetary flexibility and the possibility of refining advertising strategies to target specific periods and audiences. However, from the broadcasters' perspective 'spot-isation' is equivalent to an increase in instability of income. 57 Total advertising spend, particularly the more liquid spot market, is strongly and directly related to conditions in the broader economy, especially corporate earnings, 58 and there is little broadcasters, as an industry, can do to increase the absolute size of the advertising pie. Self-preservation means therefore competing for a larger slice against other media, chiefly the internet, whose share continues to grow.

Shrinking Production Budgets
How key station advertising is priced depends on the viewership a particular time slot draws, which is in turn dependent on the scheduled programming. Commercial broadcasters rely for their continued existence on drawing in audiences with attractive programs, all of which must be conceived, funded, planned and produced. While the first three of these functions generally take place within the key station, the final step (barring actual broadcast) relies on outside production companies and their workers. 59 These companies are generally small-scale and often staffed by temporary workers earning fairly low wages (¥2-3 million / AU$24-36K p.a.), and poor working conditions lead to high staff turnover. A 2005 Minp oren survey suggests there are in the region of 1100-1200 production companies in Japan, though the Japan Fair Trade Commission (JFTC) points out that precise numbers are unknown. 60 Seventy percent of firms employ fewer than 50 people, and over half are based in the Tokyo area.
The edifice of national commercial broadcasting is built upon a singularly unstable foundation. Recent years have seen production budgets cut and there are signs that this foundation may have reached its load-limit. As a rule only 20-30 percent of 'time' spending, program sponsorship, reaches program producers, the rest being spent on administration and management fees. 61 First, the large advertising agencies that stand between broadcaster and sponsor take fees of 10-15 percent, key stations take another five percent, and half the remainder goes towards payment of network distribution fees (NDF). 62 The remaining one-third or so is passed down to a primary production contractor (moto-uke) who typically sub-contracts to a number of smaller companies (mago-uke). The amount reaching these mago-uke companies may be less than 10 percent of the sponsor's original spend. 56 'Oikomareru terebi no setogiwa'. 57 Ogawa, 'Terebi- Between 2006 and 2009 the key stations reduced production spending by 16 percent, the Osaka stations by over 20 percent. 63 This level of cuts is confirmed in interviews with staff from production companies. Cuts may indicate a trimming of unnecessary 'fat', but if the result is further falls in program quality the long-term effects may be diminishing viewership figures and reduced viewer satisfaction ratings. And, at the other end of the process, they may lead to a loss of production capacity as small production companies either close, under financial pressure, or find themselves unable to hold on to workers due to a reduced ability to pay competitive wages. 64 There is little independent producers can do to ward off cuts, as they are relatively powerless in regard to the broadcasters that commission their work 65 and who seem willing to use this advantage to squeeze smaller companies for whatever savings can be made. Gendai Bijinesu quotes a key station sales staff member as saying: For example, for a two-hour Saturday 'wide-show' the sponsor used to put up [¥]35 million, now that's been pushed down to [¥]26 [million]. Cuts in production costs of 20-30 percent are normal. We're forcing subcontractors into cuts, 'Make 50 percent cuts or we'll take our business somewhere else'. Of course program quality is going to fall. 66 Hayama Hirotaka worked as an 'assistant director' (AD, mostly 'runner' or 'gofer' in the English-speaking world) on a prime time variety show in 2005-06. His working hours for the latter half of 2005 ranged between 382 and 509 hours per month, typically 14-17 hours per working day with 1-4 days off per month. The AD is, in production terms, 'at the bottom of the heap', and Hayama's story illustrates the long-hours culture and poor conditions taken for granted in television production in Japan. 67 Even with five years' experience many ADs may earn as little as ¥100,000 (approximately AU$1250) per month. 68 Ultimately, broadcasters rely on this unstable base of labourers for program production. Workers in small production companies may be initially attracted by the 'glamour' of television, 69 but if production budgets continue to be cut in order to prop up broadcaster profits they may well trade the glamour for more stable and better paid employment in other industries.

Key Stations' High Wage Culture
Typical wages for a married, 30-year-old key station employee with one child are in the region of ¥350-480,000 per month, while ten years later they can currently expect to earn ¥420-920,000 per month. On top of this bonuses, paid quarterly, may total up to half the annual wage; and additional payments, such as allowances and overtime, may more than double the nominal wage. In 2008 a married, 35-year old TBS employee with 63 Kimura, 'Gy oseki-kaifuku ni tenjita minp o-terebi', 51 (based on Minp oren data). 64 'Seisaku genba de nani'. Also, Kubota, 'Jakusha no gisei de'. 65 Nakano, H os o gy okai no d ok o to karakuri, 112. 66 'Uriage-hangen, seisakuhi ohaba katto '. 67 Hayama, AD zankoku monogatari.

68
Tanami, Gen'eki terebiman ga akasu!, 85. 69 Shimazaki, Ikeda and Yonekura, H os o-ron, 51. two children working 50 hours overtime per month could expect to take home in the region of ¥13 million p.a. (around AU$160K). 70 Compared to Japan's average wage, just over ¥4 million p.a., an average wage earner at one of Japan's top-paying 30 television companies is comfortably off on about ¥10 million p.a. Amongst key stations, TBS is making efforts to rein in wage spending; from April 2011 it is implementing a revised system aimed at limiting overtime payments. An insider quoted by T oy o Keizai magazine states that typical monthly overtime is around 90 hours and that some workers notch up as many as 200 hours. 71 The move to limit overtime payments to a maximum of 34 hours per month will no doubt reduce the company's wage bill, but the effect this will have on staff morale and eventually actual production is unknown. TV Asahi, NTV and Fuji are making similar moves, though NTV's planned introduction of a new wage structure in March 2010 led to strike action (carefully planned not to upset output in any way) by 600 union members the following September. 72

Relationship with Local Stations
April 2008 adjustments to the Broadcast Law allowed broadcasters to form holding companies. Two key stations, TBS (April 2009) and Fuji (October 2008) have made the change. These holding companies are, within existing regulatory restrictions, permitted to own and invest in up to 12 other broadcasters. This legal shift, and the possibility of investment limits being raised from 20 to 33 percent in the near future, may be the groundwork which enables key stations to take financially weak local broadcasters under their corporate wing. Whether the hopes embodied in these changes materialise is down to the actions of broadcasters but it seems unlikely given the unstable condition of key station finances.
In early 2010, Ujiie Seiichir o, then president of NTV, admitted '[l]ocal broadcasters' financing/management is a big problem [. . .] More than half of them are in the red. [. . .] Certainly we have to do something but how we help them, that's a difficult problem '. 73 He also acknowledged that it is impossible to help local stations in the numbers that currently exist and suggested either network consolidation, where the 'one station per prefecture' model 74 changes to a 'one station per region' model or where one local prefecture-based station broadcasts all networks. Although the overall financial situation of the non-WBA stations had improved by the end of 2010, 39 companies still ended the financial year in loss, 25 making losses of ¥100-400 million (see below for more on mergers). Actually, local broadcasters' fates may already be sealed. Ikeda Nobuo suggests that for the key stations now, local stations are little more than 'baggage'. 75 He also claims that the MIC's thinking is similar, quoting a senior bureaucrat speaking 'about ten years ago' as follows; 'Squash the local broadcasters, leave just the easily 70 'Terebi gy okai: 10 no gimon ', 48. 71 Ibid. 72 'Nittere-r okumi, 24jikan suto e', 'Nittere-r okumi, 24jikan suto no totsuny u '. 73 Interviewed in T oy o Keizai, 20 February 2010, 70-71. 74 This model was influenced by the World War II ikken-isshi ('one prefecture one paper') system for newspapers; see Kasza, The State and the Mass Media in Japan, 187-188. 75 'Uriage-hangen, seisakuhi ohaba katto '. controllable key and sub-key stations. One of the aims of digitalisation is to get rid of the unprofitable local stations.' 76 On the other hand, writing at the beginning of 2010, Kimura Mikio, of the Minp oren Kenky ujo research centre, suggested that while many broadcasters were facing immediate troubles, an optimistic mid-term outlook was warranted as growth in new areas of income would make up for overall minor shrinkages in income from advertising. Having said this, he considered the strategy of cutting production costs to maintain profitability would soon reach its natural limit, and that this would mean more companies would have to turn their attention to other areas, specifically the 'holy ground' (seiiki) of wages. 77

Local Television: The Future of the Second Tier
The 99 keiretsu-kyoku local broadcasters are prefecturally based organisations with roots in the locality they serve. Although their actual programming output is limited, they are the workplaces for most 'TV people'. Their future, both as a source of informational diversity diluting Japan's tendency to Tokyo-oriented centralisation and as local employers, is far from assured. Whilst in the long term it may prove impossible to maintain the status quo in terms of number and size of stations, it is important that the principle of diversity, a theoretical mainstay of any 'free' media system and here embodied in the form of alternatives to the Tokyo media, be supported.

Second-tier Television: Scale and Markets
Key stations can be, at least superficially, respectful of the wishes of their network partners, a significant number of whom are senior in the sense of 'having been founded first '. 78 However, despite the image of cooperation projected by networks, divergence between the tiers is great. Below I describe actual disparities in business potential, and again demonstrate that terrestrial television broadcasting in Japan is essentially an amalgam of the activities of two qualitatively different industries, the first-tier contentproducing key stations and the second-tier content-distributing local stations.
Local broadcasting employs just under 80 percent of those who work for television firms in Japan, about 17,000 people. 79 It is thus a relatively small industry in absolute terms, though these figures do not count workers who rely on broadcasting for contract and part-time work or in related service industries. There are 500 or so small television production companies outside the Tokyo area who rely heavily on local broadcasters for commissions. 80 The 99 non-WBA companies employ four-fifths of workers yet account for just under 30 percent of total industry sales (around ¥600 billion), the other twothirds or so (around ¥1,475 billion) being made by the 15 largest WBA-based 76 Ibid. 77 Kimura, 'Chideji-ka to uriagegen de', 19. 78 The Minp oren yearbooks include an appendix listing broadcasters in order of their founding; see e.g. Minp oren nenkan 2010, 731-733. Such information is possibly useful to industry historians, but should primarily be seen as a confirmation of the industry's seniority structure. companies. The 13 cross-network independents (or rather 'less-dependents') account for just two percent. 81 The Kant o WBA, the licence area of the key stations, is made up of Tokyo Metropolitan Area and the surrounding prefectures (Chiba, Kanagawa, Gunma, Saitama, Tochigi and Ibaraki). It is home to 42 million people (17 million households); calculated on figures for disposable income per household, 82 there is annually ¥85.2 trillion to be disposed of in this area. Compare this licence area with another, Fukui prefecture. Served by affiliates of just two of the networks, Fukui is home to 0.81 million people (0.27 million households). 83 Even though typical disposable income in Fukui is greater than that in the Kant o area (¥6.45 million per household per year as opposed to ¥5.10 million), annual total disposable income there is just one-fiftieth, at ¥1.72 trillion. While Fukui's case may be extreme, many other prefectures are similar -for example, Iwate, where potential audience is under half a million households and annual total disposable income around one-fortieth of the Kant o area.
In terms of Gross Prefectural Product (kennai s oseisan) the story is equally clear. The prefectures comprising the Kant o WBA account for 37 percent of GDP, those in the Kinki WBA 14 percent and those in the Ch uky o WBA another 10 percent, Hokkaido and Fukuoka contributing another four percent or so each. The rest of Japan's contribution is just under 30 percent, spread over the broadcast areas of four-fifths of broadcasters. Even for local broadcasters to find local funding for the 10 percent or so of total programming they typically produce seems a tall order, given the limitations imposed on them by a licensing system which restricts market size.
Digitalisation, it has been suggested, once the pain of infrastructure investment has been survived, offers local broadcasters a way out of their financial reliance on the key stations. Nishi contends that the ability to broadcast in the 'one-seg' format to mobile phones will mean new business opportunities for local broadcasters. 84 These will emerge from the interactive features of DTTV, leading to more efficient tie-ups between advertisers and shoppers, and from exploiting the possibilities of linking local programming and advertising with local businesses and online services via 'one-seg'capable devices. While many broadcasters are very proud of their local connections and their community involvement, it seems doubtful that they would be willing to rely on inevitably limited income from exclusively local sources. Such reliance would mean local broadcasters having to further cut their coats to suit their cloth, and content further in thrall to commercial interests.

Role of Local Broadcasters
Since the advent of digital BS broadcasting in 2000 -for local broadcasters a veritable sword of Damocles -the key stations, originators of the overwhelming majority of programming, have gradually gained the ability to broadcast directly to all of Japan 81 Dents u S oken, J oh o media hakusho, 132. 82 Used here as a measure of how appealing a market is to advertisers in search of sales. 83 FBC in Fukui prefecture is a cross-network broadcaster offering programs from both TV Asahi and NTV. Its sole competitor is Fukui TV Broadcasting, an FNN affiliate. 84 Nishi, H os o gy okai handobukku, 106. Under the ISDB-T digital television standard (see above n. 4), each 6 MHz digital channel is divided into 13 segments. 'One-seg' broadcasting uses one of these to broadcast a much-compressed signal. HD-DTTV occupies 12 segments and an SD channel requires four.
(whether all reachable homes are equipped to receive these signals is for the time being another matter, though many are). Key stations have decided, primarily out of consideration for their local affiliates, to not simulcast terrestrial programming on their BS channels. They are not legally compelled to differentiate programming on terrestrial and satellite channels. For local broadcasters this is a matter of deep concern; as Kwak has noted, 'the concurrent broadcasting of programs in both BS and terrestrial service would [. . .] jeopardize local terrestrial broadcasters country-wide, whose analog programming is generally less popular '. 85 Specifically, BS undermines their role as relay for key station programming and threatens their income from network distribution fees and 'network time' advertising. The key stations' 'no simulcast' decision could be reversed or amended at will, providing they are prepared to live with the consequences. BS therefore remains a threat to the existence of local broadcasters in their current 'distributor' role. Furthermore, an ever increasing proportion of Japan's population has been raised in a world where borderless, and much of the time gratis, communication is the norm. Local broadcasting, rooted in geographical self-restriction, has to deal with a paradigm shift in its viewers' expectations. The future would seem to offer a choice between painful change or a gradual fall into irrelevance. The record of institutional inertia makes any more creative outcome much less likely; regulators have tended to restrict themselves to small-scale policy 'adjustments', and the industry's primary instinct is to maintain the status quo.

Financial Links between Key and Local Stations
Relationships between tiers are financial as well as informational. Money flows in both directions, local stations making nominal contributions to production costs and newsgathering services and key stations paying for local distribution. How these financial linkages work in practice is far from transparent. Local broadcasters' income can be broadly classified as follows: (1) Broadcasting income (h os o sh uny u) (a)) Advertising sales (b)) Other payments (2) Non-broadcasting income The overall proportion of income a local broadcaster derives from non-broadcast activities varies greatly with individual circumstances and is not dealt with here. Within income from broadcasting, generally headed h os o sh uny u on balance sheets and financial reports, two separate elements are subsumed. 'Advertising sales' comprises sales direct to advertisers by the local station (all 'spot' and a small proportion of 'time') plus the portion of 'time' allotted to the station from the programming provided by key stations; the latter is referred to as 'network time', and makes up 30-40 percent of local broadcasters' advertising income.
'Other payments' refers to the rather opaque NDF revenues, for which a number of alternative terms exist: denpa-ry o, h os o-ry o, nettow aku-hi, nettow aku hosh o and nettow aku haibunkin. The last of these seems most prevalent and I use the English equivalent here. 85 Kwak, 'Restructuring the Satellite Television Industry ', 70. These payments have been described as local television 's 'lifeline'; 86 in this section I look at how the lifeline functions and review recent changes. Their rationale is as follows. Each company has exclusive use of the airwaves in its own licence area, so if a key station wishes a local station to pass on its programming to that locality's audience (which it generally does, thereby being able to sell a nationwide audience to its advertisers) it must 'borrow' the airwaves of the local station and, in effect, pay a rental for each program broadcast -the 'network distribution fee ', NDF. The rates for these transactions are, formally at least, fixed by the companies involved, based on the 'population and economic power' of the locality; 87 the actual amounts, however, seem to depend very much on the specific financial situations of those concerned. 88 Hanzawa and Takada refer to the networks as providing 'operating aid' (keiei hojo) through NDF payments. 89 An interviewee from the accounting department of one large local broadcaster in southern Japan suggested that the basis for calculation of NDF is the local broadcaster's potential audience (license area population), and that for key stations to make yearly adjustments to the payments on the basis of individual local stations' financial situations would be simply too much of an undertaking, and that he doubted whether this happened. Another analyst claims that Osaka, with 20 million viewers, receives only double the amount (five percent) of NDF received by Sendai (2.5 percent) with three million, suggesting that population is by no means the whole picture. 90 Key station comment, such as the statement made by Fuji's Toyoda K o (see below), on reductions in NDF payments 91 would make little sense if the amounts involved were not to some degree discretionary. Observer consensus seems to be that while the objective measure 'area power' (calculated factoring in local demographics and economics) is the ostensible basis for apportioning NDF payments, they are in fact used to prop up local stations who find themselves in straitened circumstances. 92

Networks as Financial Entities
Given this inter-tier linkage the networks themselves can be considered financial entities with analytical significance. As Table 2 illustrates, only FNN (Fuji Television) maintained both tiers in profit in 2009-10. The contribution of local broadcasters to total network profits for the FNN network is in the region of 25 percent, far higher than any of the other networks. 93 To what extent local stations are financially dependent is unknown (due to the lack of clarity in the NDF system), and, as long as the key stations feel the need to keep them in business they are safe. However, by gradually allowing their own local production capacity to dwindle, they have painted themselves into a corner and unless radical costcutting can be implemented the move from independent broadcaster to relay station with occasional local news injects cannot be too far away. At the end of 2009, president of Fuji Media Holdings (Fuji Television's parent company), Toyoda K o, saw a troubled future for local broadcasters: We are reducing NDF payments [nettow aku-hi]. We intend to make reductions in the latter half of FY2009 and through FY2010. However, local broadcasters are currently in a very difficult position with the advertising slump and investment in digitalisation. There's always the problem of how far we should go but I think we're heading into a new era and we have to go on reducing NDF payments. 94

Industry Restructuring Options: Mergers and 'Block-isation'
The main pressure on commercial television broadcasting's ability to maintain the status quo is reductions in advertising income. If the predicted fall in ratings outlined earlier is realised then incomes will fall post-digitalisation. Of course, this assumes that ratings reflect viewership accurately. Video Research, Japan's sole ratings provider, is one-third owned by Dentsu and has recently been led by ex-Dentsu men; 95 it has a vested interest in propping up ratings numbers. If ratings are seen to fall this may act as catalyst for restructuring.
The form a restructuring will take is currently moot, and the various players have different preferences. The first possibility is the maintenance of the current region-based system with a reduction in the number of companies within each region as firms merge. The second involves a change in the nature of the regions, the market within which the broadcasting companies must do business, possibly into larger 'blocks' along the lines of the cross-prefectural 'block' newspapers. 96 Either of these two may be accompanied by a shift to formal division between companies undertaking broadcasting (distributing the television signal) and program-making, the jutaku-itaku arrangement seen in CS. All the above would necessitate some degree of change to existing regulations; so, again, institutional inertia will be a significant factor. 93 The performance of the ANN network in this particular year was adversely affected by the very large one-off loss of ¥4.9 billion posted by its Ch uky o WBA affiliate MeiTV (Nagoya H os o). 94 Hasegawa, 'H os o sh uny u no suii ni ojite '. 95 See the Dentsu FY2009-10 accounts, www.uforeader.com/v1/se/E04760_S00066R3_4_0.html# #E0002 (accessed 12 June 2011). Current president Wakabayashi Itsuma was with Dentsu 1969, and his predecessor Kimura Takehiko in the period 1964 There is no formal definition of a block newspaper, but they tend to be larger than city papers and not national.

Merger Restructuring
In 2003-04 the broadcasting industry at all levels agreed that the number of companies involved in terrestrial television would fall in the future. Debate centred around the questions of who and what would follow, and when. While the outlook seems, even now, to be less than optimistic, the company failures predicted as a result of the burden of investment in digitalisation have not materialised as of (northern) autumn 2011.
Financial analysts, while placing the benefits of any mergers in the medium-rather than short-term, have argued for the necessity of industry restructuring and suggest similar alternatives: 'block-isation' (see below), or mergers, perhaps led by local or block papers, local industries, or national papers and broadcasters. 97 Given networks' identifications with large newspaper publishers, intense rivals all, it is hard to see how mergers could come about without parallel moves in the press. Firms in the TBS-Mainichi system seem the most likely to have to make the first move. Nearly half of the JNN non-key stations failed to return a profit in 2009, and in the same year Mainichi Shimbun-sha ended in the red for the first time since 1994. 98 Mergers would also mean, throughout most of Japan, scrapping a significant proportion of local broadcaster investment in DTTV infrastructure as transmitter facilities are consolidated.

Block Restructuring
This would involve the expansion or merging of broadcast areas to create larger markets, and a reduction in the number of full-scale broadcasting firms operating within those areas. 'Block-isation' raises the question of where viable blocks could be formed. Television broadcasting is both a cultural and an economic activity, and so ideally blocks should be meaningful to audiences as well as of sufficient economic potential to maintain the broadcaster(s). There are also the faits accomplis of the WBAs and other layers of administrative regionalisation to take into account. The strength of regional identities varies greatly; one of Suzuki's informants bluntly observed, '[t]here's no block identity in Ch ubu ', for example. 99 Recent changes to media company regulation may hint at one possible future; media holding companies have been allowed to invest in up to 12 other media companies. With key stations counting as seven stations, the remaining five may be the number of network areas envisaged by regulators, who invariably form policy in discussion with interested parties.
Block-isation seems to be the future preferred by 'the centre', that is, by those involved in forming policy and by the key stations. 100 The result would be four nationwide networks centred on the Tokyo stations, with regional centres at the socalled kikan-kyoku (nucleus stations) Sapporo (Hokkaido), Sendai (T ohoku), Shizuoka (Ch ubu), Hiroshima or Okayama (Ch ugoku-Shikoku), and Fukuoka (Kyushu). 101 The smaller local stations would become branch offices of these regional centres performing news-gathering functions and maintaining local distribution facilities.

Summary and Conclusions
Japan's advertising pie is getting smaller, and terrestrial television's portion of that pie has been shrinking for several years. Television companies have been impoverished by the government policy of digitalisation. Spending on content production has fallen, industry insiders acknowledge a corresponding fall in quality. Both ratings and figures for the number of 'households using television' (HUT) continue their slow decline. 102 Nothing happens quickly in public policy in Japan, so the broadcasting industry can expect a few more increasingly painful years of status quo as the various industry players build consensus. This will inevitably involve pain for someone; key stations must share the income they have access to more equitably, meaning a reduction of fixed costs such as wages; weaker network stations must increase efficiency, either by prioritising distribution or production but not both, or by finding themselves local or regional partners and looking for economies of scale.
Given what seems to be key stations' and administrators' preference for the 'blockisation' option, smaller local broadcasters face an uncertain future. If the number of broadcast areas is reduced from the current 32 to 10-12, this will mean that two-thirds of the firms in each national network could be reduced from broadcasters with the full complement of functions -sales, marketing, personnel management etc. -to branch offices with limited information and news-gathering capabilities. Staff numbers at these firms could be cut by 80-90 percent.
Block-isation would mean the loss of a significant number of television jobs and media production would become yet more concentrated in regional centres. This may nevertheless be a price worth paying for stronger regional media and the preservation of informational diversity at the national level. It should be remembered that the level of diversity exhibited by Japan's current television system is to some extent artificial, emerging from a bureaucratic 'one-size-fits-all' frame of mind which regards fairness as the equal treatment of all regardless of individual differences. Thus, all prefectures, regardless of their population or 'economic power', have stations affiliated to each of the four national networks. Block-isation may offer a way of preserving a proportion of this diversity while placing broadcasting businesses on a sounder financial footing, thus securing a degree of independence from Tokyo key stations.