Since the beginning of the Asian financial crisis, Salomon has successively stamped its name as the most successful lead manager in the re-capitalization of Asia inc with highly acclaimed deals for Siam Commercial Bank (SCB) in Thailand and Neptune Orient Lines (NOL) in Singapore. Should it succeed with this new deal, which may result in a 100% re-capitalization of Hynix's equity base, it will have also provided a template for the Korean government to further its painful re-modeling of the whole Chaebol sector.
Hynix is hoping to be the standard-bearer for a new type of Korean company. Where once family ownership held sway, the new Hynix will almost certainly be dominated by foreign institutional investors. Where once the company was run as a personal fiefdom of the Chung family, it is now being operated by professional independent management. Where once the focus was on market share and scale, the emphasis will now be cash flow and EBITDA. And above all else, it is hoping to break the previous owners' reliance on short-term debt and unchecked expansion.
Hynix's re-capitalization will be similar in spirit to both SCB and NOL. Success will, therefore, be contingent on persuading investors that this is a one-off re-capitalization and one that is based on highly conservative assumptions that have passed stringent stress tests. Having accepted that the company will not be making further cash calls over the near term, investors will then need to be convinced that the same mistakes will not be repeated further down the line.
Key to both assumptions will be their confidence in the new management's ability to force painful adjustments on a company that has previously been a recalcitrant participant. This "agent of change" is CS Park, the former chairman of Maxtor, who took over as Hynix chairman last February and has pushed hard for its disaffiliation from the Hyundai group.
Not surprisingly, the focus since then has been very firmly on the company's ongoing liquidity crunch and its battles to gain independence from the Hyundai Group. Bankers will be hoping, however, that with the threat of bankruptcy lifted by the rescheduling agreement and ties to the Chungs almost severed, investors will start to focus on the new company that is emerging from the ashes. Once this focus shifts from negative to positive, they will also be hoping that many accounts will view both the debt and equity transactions as strong buys.
The liquidity crunch
To understand how Hynix is trying to escape its past, it is important to step back a few years. Just prior to the financial crisis in 1996, for example, W2.9 trillion in capex exceeded semi-conductor revenues of W1.8 trillion and amounted to about 50% of W5.9 trillion in consolidated revenue. One year later as the financial crisis bit heavily into the Korean economy, it found itself burdened with W7.4 trillion ($5.67 billion) in long-term debt and a negative equity value of W696 billion ($534 million).
Some effort to reduce this debt load then floundered in October 1999, when the company merged with LG Semicon and financed the acquisition with a sizeable amount of short-term debt. At the end of that year, it was still cash flow negative by about W21 billion, although one year later under the new management, it had been able to turn the figure around to W881 billlion cash-flow positive.
Its debt levels, meanwhile, remained extremely high, with W7.928 trillion of short-term debt and W3.533 trillion of long-term debt outstanding as of end of December 2000. But its ability to re-pay this short-term debt became severely constrained by the collapse of the domestic bond market, which was afflicting the whole of corporate Korea and plummeting DRAM prices, which constrained cashflows of the entire industry.
Yet, the company's supporters have long argued that the problem with Hynix's debt load lies not so much in its overall size - debt to equity was a manageable 206% at the end of 2000, but its maturity schedule. Based on 2000 EBITDA of W4.451 trillion, for example, the company was still able to maintain a fairly comfortable interest coverage ratio of 3.5 times to the end of 2000. However, with W6 trillion ($4.6 billion) coming due in 2001, a liquidity crunch was looming fast.
Creditors re-payment and re-structuring agreement
The first step to return Hynix to financial health came with the company's participation in the government's "Fast Track" bond repurchase programme, which enabled eligible companies to re-finance up to 80% of their principal obligations maturing in 2001. Under the programme, the Korea Development Bank (KDB) purchased W2.9 trillion of Hynix debt, giving the company a one year breathing space before it fell due again.
At the same time, creditor banks led by Citibank arranged a W800 billion one year syndicated loan. Both transactions provided a lifeline, but failed to provide a longer-term solution that would prevent a second liquidity crunch in early 2002.
Prior to the new re-scheduling agreement, this left the company with W9.553 trillion in debt of which: W2.76 trillion would still fall due in 2001; W5.234 trillion in 2002; W522 billion in 2003; W871 billion in 2004 and W166 billion in 2005.
The company therefore needed to persuade its creditor banks to push out the maturity curve, but was unable to rely on KDB for further assistance following pressure from Micron lobbyists in the US, who argued that such an action would be anti-competitive.
In terms of re-financing the KDB bonds, the company has now been able to get a group of 17 domestic creditor banks and Citibank to subscribe to a W1 trillion convertible bond that could give the combined group up to 20% of the company's equity should it be fully converted and should the company complete a 100% equity re-capitalization. However, the company does have a call option on the convertible which could be exercised from surplus funds should DRAM prices recover and non-core assets are sold.
Pending completion of the forthcoming debt and equity offering, creditors have also agreed to re-schedule long-term debt amounting to W1.9 trillion and maintain availability of short-term trade financing of W2.8 trillion for up to two years.
As a result of the these efforts and according to company figures, Hynix's new maturity schedule will consist of: W2.7 trillion of debt falling due in 2001; W2.655 trillion in 2002; W1.034 trillion in 2003; W1.65 trillion in 2004; and W1.454 trillion in 2005.
Originally this re-structuring agreement looked like it was being held up a group of domestic ITC's that were reported to be refusing to purchase Won680 billion in bonds unless they were guaranteed by the Seoul Guarantee Insurance Company. This has not happened and the creditor banks appear to have backed down and allowed the restructuring to go ahead without them.
Disaffiliation from Hyundai
At the end of March, what was formerly know as Hyundai Electronics Industries became Hynix in a highly symbolic separation of new company from old. By this point, Hyundai group affiliates still owned 19.2% of the company, but had resigned all board seats, renounced their voting rights and agreed that any affiliate transactions would be at arms length.
In their place, Park appointed nine out of the board's ten directors and in an effort to improve corporate governance, made sure that seven directors were independent. Foreign expertise was also drafted in with, for example, the appointment of James Guzy, also a director on Intel's board.
In his efforts to change the company's corporate culture, Park also replaced guaranteed bonuses with bonuses directly linked to EBITDA projections and sent all staff on a two day crash course in financial management that emphasized the new reigning principle - "cash flow is king."
Following the completion of the GDR offering and assuming conversion of the convertible bond, it is estimated that Hyundai affiliates will be diluted down to 10%. In the next step, the company has also agreed to dispose of the remaining shares.
The GDR and high yield offering
With any recapitalization no investor is ever willing to put in the first dollar, but most remain very willing to put in the last. Enabling the whole restructuring to run smoothly, necessitates making sure that all the pieces of the jigsaw fall into place at exactly same time. The debt and equity offerings would therefore not have been possible without debt rescheduling and in turn, the restructuring of the balance sheet will only be triggered on successful completion of the debt and equity offering..
As one observer puts it, "The major premise underlying the proposal is simple - it is for the markets to decide whether Hynix is a viable business. The commercial banks simply provided the prospect of a strong financial platform IF new money comes in. If the market does not wish to invest in the company, then the debt restructuring will not crystallise and all creditors will be in the position they were previously in."
Currently, Hynix has around 500 million shares outstanding and a market capitalization of roughly W2 trillion. With the issuance of W1 trillion in convertible bonds and $800 million from the GDR issue, the company will have effected a 100% re-cap and reduced debt to equity ratios from 206% as of end 2000 to about 130%.
Under the current shareholding structure, nearly 80% of Hynix shares are in public hands, of which foreign institutions own 35% and domestic institutions and the retail public the remaining 40%. Hyundai affiliates owning 19.2% number: Hyundai Merchant Marine on 9.3%; Hyundai Heavy Industries 7% and several other Hyundai affiliates 2.9%.
All existing shareholders (except Hyundai) have the right to participate in the GDR offering to avoid being diluted, but will have no pre-emptive rights. Whether the company's largest institutional investors decide to do so is likely to be heavily dependent on their having not already written their distressed holdings down to zero. Either way, observers believe that there will be a noticeable swing in favour of foreign institutional investors on completion. In percentage terms, it will have the largest free float and largest foreign investor base of any Korean company.
The company chose a GDR structure over an ADR because it is currently not US GAAP compliant, but is aiming to become so by 2002. The deal is being marketed in terms of absolute price rather than by number of shares, although it has received shareholder authorisation to issue up to 600 million shares.
The rule 144a reg S deal will be Luxembourg-listed with one GDR representing five common shares. Three is also a 180 day lock-up. Fees total 3.25%, with the final syndicate including Deutsche, SG Securities, ING Barings, and Nomura.
To make sure that it is not hamstrung by the innumerable regulations that have proved the downfall of international equity deals from Korea in the past, Salomon has ensured that it will not be subject to any pricing constraints. Hence, there is no limit on the maximum discount that can be applied and pricing can come below par value of W5,000.
Observers say that the removal of pricing constraints should not be read as an admission that the deal will be priced at a huge discount, simply that when there is a floor, investors tend to gravitate towards it to the detriment of bringing in the deal at more aggressive levels.
A further innovative twist is that the GDR's can be simultaneously placed (and at the same price) with both international and domestic investors. This move, which now applies to any future international Korean equity offering, marks a natural progression from the government's recent allowance of two-way convertibility between the domestic and international markets. There is, however, a 10% ceiling on placement to domestic investors.
The concurrent 10 year high yield bond offering will include JPMorgan as a joint lead and is likely to be closely referenced to two outstandings issues by wholly-owned subsidiary Hynix Semiconductor America. Rated B-/B3, these consist of an 8.25% May 2004 transaction currently trading on a dollar bid/offer spread of 58/63 and a yield of 30%/27% and an 8.625% May 2007 transaction trading at a bid/offer spread of 63/68.
Bankers will be hoping that news of the successful debt restructuring and a build up of momentum with the unfolding of the new credit story, will push these dollar prices higher. The last thing the company wants to face is the prospect of future capital markets fundraising based on new benchmarks that have been set at extremely high yields.
Investor concerns about the company's former propensity to fund itself with short-term debt will also be partly allayed by covenants restricting future re-financing at maturities of less than three years. Proceeds of the debt issue will be used to re-pay short-term debt and extend the maturity profile, while proceeds from the equity issue, alongside free cash flow, will be used to fund capex.
The company's ability to fund W1 trillion in capex during 2001 and keep its technology up-to-date will form one of investors' main considerations. This ability will in turn hinge on a turnaround in DRAM prices. Having been the first to predict the downturn, Salomon is now one of the first to conclude that the bottom has been reached and a third or fourth quarter recovery likely.
However, observers emphasize that one of the key premises underlying the rescheduling is that there might be no improvement in DRAM prices through to the end of the year. Should this prove to be the case, the rescheduling and capital raising will still remove liquidity risk and fund capex in full. Should the company then undertake asset disposals of its LCD and telecom's operations, it will also secure itself a further cash cushion.
During 2000, the company reported EBITDA earnings of W4.451 trillion, with interest expense totaling W1.287 trillion and capex W2.28 trillion, leaving free cash flow of W881 billion. Debt to EBITDA stood at 2.6 times. Because the company wrote off a significant percentage of its non-performing assets to bring their book value down to expected sales prices, it reported an overall loss of W2.578 trillion. Shareholders equity stood at W5.554 trillion.
Despite the depressed state of the semiconductor market, Hynix has still managed to stay EBITDA positive for the first three months of this year, reporting W602 billion at the end of March. Free cash flow is likely to stay in line with 2000 levels through to the end of the year, as will interest expense around the W1.1 trillion mark.
In terms of upgrading its technology, Hynix is rapidly converting all existing processes to 0.18 micron or below and will begin implementing a 0.13 micron roadmap at the end of the year.
Currently, the company is the world's third largest DRAM manufacturer based on revenue (behind Samsung and Micron) and number one in terms of units shipped. Given its substantial wafer capacity, it has started to try and reduce volatile revenues by diversifying into other sectors of the semiconductor market. Where DRAM's accounted for 70% of capacity last year, the company estimates that this will fall to 50% by 2003 as Hynix ramps up SRAM, Flash and System IC production.
Investors are also likely to question whether the company will be able to fund a build-out from eight inch wafers to twelve inch. In this respect, Hynix officials are likely to argue that the downturn in the semiconductor industry has pushed most competitors plans back, giving the company a little more breathing space. Infineon is said to be the only major player with 12 inch capacity. By contrast Samsung is said to have minimal capacity, while Micron is said to have taken a wait and see approach.
Roadshows for the twin transactions will kick off in Seoul on Monday, moving to Hong Kong on Tuesday and Wednesday, Singapore Thursday and Friday, then Europe the week beginning May 28 and the US, the week beginning June 4. This coming Friday, there is also likely to be a conference call for non-syndicate analysts to explain the rescheduling and capital markets transactions.
Few would argue that it won't be needed given the large number of doubters out in the market. Salomon is trying to bring an ambitious deal for a misunderstood and distressed credit at an extremely unfavourable juncture for the straight equity markets right across the globe. Investors have shown that they have little interest in high capex semiconductor companies and even less willingness to forsake a conservative investment policy for the sake of what some will view as a punt.
Nevertheless, most have come to the conclusion that Hynix is not going to be allowed to fail. Accounting for nearly 5% of Korean exports and its fortunes directly linked to at least three banks that would almost certainly fall with it, the company has the full backing of the Korean government. Having forced creditor banks to accept the rescheduling plan, the government has now pushed Hynix out on its own in the hope that with a clean bill of health, it can stand on its own two feet going forwards.
Currently trading at only W4,065 (Wednesday's close), there is certainly plenty of upside relative to comparables such as Samsung and Micron. At these levels Hynix is trading on a firm value to 2001 estimated wafer units of only $1,644, compared to a $3,903 level for Samsung and $12,016 level for Micron.
Once investors get past their initial cynicism and start to factor the numbers into their models, observers say that attitudes shift fast. In a recent credit view, for example, Income Partners - one of Asia's long standing fixed income investors - says that it's positive view of the re-capitalization is based on a belief that Hynix has a viable business model constrained by a weak capital structure and skewed debt profile.
"We believe that current prices (or slightly below) could offer an attractive entry point to a potential turnaround credit," it wrote late last week. We see potential for both Hynix bonds to trade to the mid-70's following final closure of the re-cap efforts, with returns in excess of 30% in a six-month period."
However, it also notes, "that there are considerable risk factors....Additionally, further declines or a delayed recovery in the DRAM cycle could pose additional strain on Hynix cash flow."
And its advice: "Hynix bonds should be considered in terms of a small allocation to a diversified portfolio."