Japanese Regional Banks Non Performing Loan Time Bomb – Horseman Capital
Whilst I have recently highlighted many serious issues at Japanese regional banks. It is the vast amount of hidden non-performing loans (NPLs) that worry me most. The ignition point for this time bomb will likely come with the mass retirement of bankrupt “baby boomer” Small to Medium size Enterprise (SMB) owners that have just started turning 70.
Japan’s long history of underreporting NPLs reached a new level in 2009, when the then Financial Services Minister stated:
Seth Klarman: Investors Can No Longer Rely On Mean Reversion
"For most of the last century," Seth Klarman noted in his second-quarter letter to Baupost's investors, "a reasonable approach to assessing a company's future prospects was to expect mean reversion." He went on to explain that fluctuations in business performance were largely cyclical, and investors could profit from this buying low and selling high. Also Read More
“We’re going to get financial institutions to provide firms with more loans”. “Banks won’t have to treat debt on which they provide a moratorium as bad… I ’m not going to leave small companies in the lurch unable to get loans. If a bank takes that approach, I’ll hit them with a business improvement order. ”
Soon after, the Small to Medium Size Enterprise Financing Facilitation Act (SMBFFA) was passed. Which KPMG suggests “provided many borrowers (including SMEs with financial difficulties) with almost automatic extensions or loan rescheduling opportunities”. At the same time the Japanese Financial Services Authority (J FSA) changed its financial supervision measures, to ensure that banks when dealing with SME requests to modify loan conditions: “contribute to reducing the burden relating to its repayment of debt” (i.e. reduce the SME’s monthly payments) and “grant credit appropriately to SME managers who have had changes” (i.e. give more money to SMB’s that can’t meet their monthly payments). As long as it can be “expected that an SME will formulate a business improvement plan” the loan “would not fall under the category of a restructured loans”. (i.e. if the SME says it will come up with a business plan the bank should not record the loan as non-performing).
The following National Institute for Research Advancement (N IRA) article provides a more readable outline of the policy from page 3. The author of the NIRA report felt “it is probably safe to assume that almost all of the loans for which terms have been changed would have been classified as non-performing under the previously enforced standards”. The NIRA note refers to J FSA data which at the time, October 2011, showed that loans modified under the SMEFFA amounted to ¥38.9bn, of which ¥25.3bn was loans from Japanese regional banks. This J FSA data was sporadically updated until March 2014, at which time loan modification approvals under SMEFFA had risen to ¥142.8tn, of which Japanese regional banks accounted for ¥67.3tn.
It is important to clarify that this data does not reflect the outstanding loan amounts, as the data double counts loans when additional modifications are made and/ or when loans with modifications in place are rolled-over. We can however get a rough idea of the outstanding loan amount, by estimating the amount of completely new approvals given in each year and adding those amounts up. The Research Institute of Economy, Trade 8: Industry completed a survey in October 2014 of SMEs that had “modified loan conditions” under the SMBPPA. One of the survey questions asked in which year the SME received its first loan modification under the SMEFFA. As we can see below just under half (48.4%) of these SMEs received their first loan modification in 2009-2010, soon after the policies began. More companies applied for their first loan modification in 2011, 2012, 2013 and 2014.
By using the data in Chart 2 and the data in Chart 1, we can estimate that in October 2014 the total of loans outstanding with modified conditions under the SMEFFA amounted to ¥93.9tn, of which ¥50.3 was provided by Japanese regional banks. This is 2.73x the total net assets at Japanese regional banks (¥ 18.4tn) on their combined balance sheets in that year (Source: Japan Bankers Association). And is 8.5x larger than the non-performing loans officially stated by Japanese regional banks in that year (¥5.9tn) (Source: J FSA).
“Although the SME Financing SmoothingAct expired at the end of March 2013, the JFSA has not reversed the rule that allows banks to classify the restructured loans to SMEs as normal”. “Even after the expiration of the Act on Temporary Measures to Facilitate Financing for Small and Medium-Sized Enterprises, more than 90% of applications for changes in loan terms are accepted by financial institutions”. Whilst the J FSA seems to have stopped reporting the total value of loans modified it does still report the number of loans modified. The number of loans has continued to grow, again this includes further modifications and roll overs of previously modified loans. It is also important to note that applicants can and do apply for modifications to loans at multiple banks.
The previously mentioned RIBTI survey makes for interesting reading. Of those SMEs that received modified loan conditions:
*52.8% said they would have gone out of business without the loan modifications.
*54.7% had their monthly repayments deferred and 37.9% received grace on principal repayments.
*Only 43% said that the business plan they submitted provided a clear image of the future for their company.
*And 24.4% did not even bother to submit the required business plan.
*23.6% say their business conditions haven’t changed since then, whilst 17.3% say their conditions have deteriorated further.
*Of those that modified conditions more than once 58.3% always expected to change repayment terms again from the beginning.
The outlook remains poor for SMEs in Japan with the Teikoku Credit Databank Survey in June 2015 suggesting “47.5% of companies anticipate their “core business” market to shrink in the future, while 19. 7% expected an expansion “. Whilst a weaker Yen has lifted larger corporate export earnings, it has hurt SMEs who largely import for domestic demand “46.2% firms in Japan stated that the yen’s depreciation had done more harm rather than good to them, while only 8.2% considered the yen’s depreciation an advantage”.
Banks meanwhile have been forced to continue to lend to troubled SMEs: “If we only keep looking for healthy companies for lending money, there won’t be much left for us,” (Hideki Yoshida, a chief loan officer at Hachijuni Bank). “We won ‘t flatly say no just because a company has a poor credit rating or bad earning figures” (Shigeharu Sanada of Bank of Tokyo Mitsubishi UFJ)
It is relatively easy to identify the industries where SMEs are in deep trouble. The following data comes from the March 2014 aggregated SME results database. Japan’s real GDP and its SME business confidence have both declined since then (Source: Ministry of Finance and Shoko Chukin). I have filtered the data down to the smallest SMEs, those with less than 5 employees, for industries I feel face the most oversupply versus dwindling demand. In my travels around regional Japan I saw many restaurants, service providers and retail stores operating with a few staff and no customers.
The key question from a market timing perspective is how long can SMEs go on making losses with negative equity. It seems that demographics will soon bring this to a head. 48% of managers at micro corporates are aged over 60. “20% of micro business owners have resigned to the possibility of going out of business during ones own generation”. The rate of business closures has been rising, with 86% of business closures coming from those above 60. “Baby boomers”, the largest group of Japan’s population, are start turning 70. Business closures will accelerate as business owners are not physically able to carry on.
The J FSA is certainly aware of these issues “Japan is faced with structural issues in the form of population decline and hyper-aging. The hyper-aging of the population will continue, accompanied by a steep decline in the working-age population. The exit rate of firms and self-employed businesses is much higher than the entry rate. These factors mean that Japanese financial institutions will face a very difficult business environment in the long term”. (Source: J FSA report 2015)
Recent policy will also see banks take a larger hit for businesses closing with negative equity. In 2013 the Abe cabinet established guidelines for treatment of personal guarantees, “such that upon business termination consideration should be given to setting aside a certain amount for things such as lifestyle expenses”. SMEs in Japan employ 75% of workers, according to the Ministry of Health Labour and Welfare, and as such they are an important voter base for Abe’s Liberal Democratic Party.
It is also likely the asset and collateral values at SME borrowers are overstated. SME holdings of buildings and land account for 31% of total assets, or equivalent to 90% of Net Assets. Much of this is still held at book value despite prefectural land prices falling 65% since 2000 and 81% since 1990 (Japan real Estate Institute). Overvaluation of collateral on bank balance sheets is highly concerning in itself.
SME borrowers account for 66% of corporate borrowing in Japan and over 30% of regional bank total lending (Japan Bankers Association). Despite the dire outlook for many of these borrowers the FSA and government policies have meant that provisions at Japanese regional banks have fallen significantly. The credit cost ratio for Japanese regional banks in Japan are now 7bps (Japan Bankers Association), significantly lower than US bank credit costs of 44bps (Federal Reserve) which are at a cyclical low. Many Japanese regional banks have actually been temporarily boosting earnings by writing back the reduction of prior provisions.
“Moody’s report notes that even a small uptick in credit and investment losses to 20 bps or a level equivalent to the 10-year average net credit cost for the Japanese regional banks, would severely damage the profitability of the average Japanese regional bank. As for credit and investment losses of 40 bps – a level which has previously been reached – such a loss would wipe out most of the banks’ earnings”.
The 11 largest Japanese Japanese regional banks trade on an average of 10.5X price to last year’s earnings. As we have previously highlighted their core recurring earnings are significantly overstated as they account for one-off equity gains as interest income, these equity gains would have now reversed into losses. Their mortgage books face a demographic decline tipping point at the same time as the Japan Post Bank is attempting to enter that market. And the Bank of Japan has now crossed the Rubicon of negative rates which it will likely cut further.
The 11 largest Japanese regional banks trade on an average of 0.8x price to book (latest company statements). That book value is highly questionable with hidden non-performing loans likely accounting for 2.73x of book value. As many of those SME borrowers hit retirement age we will get a better idea of real bank balance sheets. I would not be surprised if significant capital raising and dilution of existing shares is required.