Amid turbulent markets, high yield bond volume plummets to $12.8B in June

high yield bond volume

High yield bond issuance in June slumped to $12.8 billion, the lowest monthly figure since June 2012, from a whopping $43 billion in May, as issuers struggled against overall market volatility and word from the Fed that it might reduce buying activity later this year. The second quarter saw $82.5 billion in U.S. high yield bond activity, bringing the year-to-date total to $172.9 billion (note that these numbers are updated/final from preliminary numbers published on June 30).


High yield bond prices rebound; post largest gain in one year

The average bid of U.S. high yield bonds in the secondary trading market, as measured by LCD’s flow-name high-yield bond sample, surged 171 bps over the past two trading sessions, to 101.39% of par, yielding 7.71%, according to LCD, a division of S&P Capital IQ.

This is the single largest advance in the sample this year, and, in fact, is the largest gain dating back just over one year, to a surge in early June 2012. Moreover, the advance nearly wipes out Tuesday’s big decline of 188 bps, so the average is off just 17 bps week over week.

However, the largely technical correction in recent weeks still wields plenty of influence. The average is lower by 146 bps over the past two weeks and down 393 bps in a trailing-four-week reading. And the overarching, aggressive repricing of the market since the May 9 peak is significant, as the average is down 625 bps since that time.

Recall that before this latest rebound rally, the week-over-week tally as of Tuesday, at negative 390 bps, was the largest one-week decrease in average price dating to the outset of October 2011. At that time, the market was roiling over U.S. deficit woes and rising sovereign debt yields in Europe. In contrast, the current inflection, which has essentially been in place for six weeks, is tied to rising U.S. Treasury rates as the market considers tighter government policy and outflows of cash from the asset class, including the largest on record, at $4.6 billion in the week ended June 5.

With a large gain in price today, it’s no surprise to see that all 15 constituents rose, with the majority advancing one point or more. That’s a sharp reversal from the prior two readings, which saw all 15 credits fall, with a majority lower by one point or more.

Tuesday’s reading at 98.68 is the new low for 2013, and the year’s first sub-par average price. With the latest rebound, the average is in the red by 499 bps in the year to date.

With a large increase in average price, the yield-to-worst on the sample falls 67 bps, to 7.71%. The prior reading, at 8.38%, was the highest since June 2012. Recall that last month the yield to worst on the sample fell below 6% for the first time ever, including a low at 5.86% at the May 9 peak.

In contrast to the huge decline in yield, the average spread to worst cinched inward to a lesser extent, tightening 48 bps, to T+618, or L+601, swap-adjusted. The undersized reading as compared to the large fall in yield is tied to U.S. Treasury strength of late, as an outsized decline of underlying rates encourages spread widening.

For reference, the respective averages at the 2013 market peak were 5.86%, T+510, and L+494. In contrast, at the prior cycle, the April 2007 market peak at 104.85 offered respective yields of 7.69%, T+290, and L+237. – Staff Reports


Orchard Supply unsecured creditors’ committee appointed, contacts listed


The U.S. Trustee for the bankruptcy court in Wilmington, Del, on June 26 appointed a creditors’ committee in the Chapter 11 proceedings of Orchard Supply Hardware Stores, according to a filing with the court.


The members of the committee and contact information is as follows:


  • Eleven Western Builders (Attn: Richard Huey, (760) 796-6346)
  • Jordan Manufacturing Company (Attn: David Jordan, (574) 583-6008)
  • Kawahara Nurseries (Attn: Josh Kawahara, (408) 779-2400)
  • National Retail Properties (Attn: Christopher P. Tessitore, (407) 650-1115)
  • Gina Rondone on behalf of Certified Class (Attn: Laura M. Cotter, Esq., The Markam Law Firm, (619) 399-3995)

– Alan Zimmerman


Valeant Pharmaceuticals completes 2-part bond offering, backs Bausch & Lomb buy; terms

Valeant Pharmaceuticals today completed a benchmark two-part offering of senior notes via bookrunners Goldman Sachs, Bank of America, Barclays, J.P. Morgan, Morgan Stanley, and RBC, according to sources. Both tranches were priced in-line with official talk, although that had widened from early whisper levels after the deal was announced last week as market conditions weakened. Also, the five-year tranche was added to replace an initially proposed 10-year (non-call five) issue. Note that both tranches will have a first call price at an investor-friendly par plus 75% of the coupon, rather than the usual 50%. Proceeds from the 144A-for-life offering will be used to partially finance the Canadian pharmaceutical company’s $8.7 billion acquisition of Bausch & Lomb, which was announced last month. Financing also includes a $4.05 billion loan package consisting of a $3.2 billion B term loan and an $850 million A term loan as well as an offering of common shares that netted gross proceeds of $2.3 billion. Terms:

Issuer VPII Escrow / Valeant Pharmaceuticals
Ratings B/B1
Amount $1.6 billion
Issue senior notes (144A-life)
Coupon 6.75%
Price 100
Yield 6.75%
Spread T+529
FRN eq. L+512
Maturity Aug. 15, 2018
Call nc2; 1st call @ par +75% of coupon
Trade June 27, 2013
Settle July 12, 2013 (T+10)
Joint Bookrunners GS/BAML/Barc/JPM/MS/RBC
Px talk 6.75% area
Notes w/ two-year equity clawback for 35% @ 106.75; subject to T+50 make-whole call; w/ change of control put @ 101; replaces 10-year tranche
Issuer VPII Escrow / Valeant Pharmaceuticals
Ratings B/B1
Amount $1.625 billion
Issue senior notes (144A-life)
Coupon 7.50%
Price 100
Yield 7.50%
Spread T+531
FRN eq. L+505
Maturity July 15, 2021
Call nc3; 1st call @ par +75% of coupon
Trade June 27, 2013
Settle July 12, 2013 (T+10)
Joint Bookrunners GS/BAML/Barc/JPM/MS/RBC
Px talk 7.5% area
Notes w/ three-year equity clawback for 35% @ 107.5; subject to T+50 make-whole call; w/ change of control put @ 101



AMF nets reorg plan confirmation, to merge with Bowlmor

The bankruptcy court overseeing the Chapter 11 proceedings of AMF Bowling confirmed the company’s reorganization plan on June 25, following a hearing in Richmond, Va., according to a court order filed in the case.

As reported, under the plan AMF would merge with New York-based Bowlmor, a high-end bowling operator. The plan would convert $80 million in second-lien debt into 77.53% of the equity in the merged company, while Bowlmor founder and CEO Tom Shannon, who will serve as the chairman, CEO, and president of the new company, will receive 20.69% of the new equity, and Bowlmor CFO Brett Parker will receive the remaining 1.78% of the equity.

Shannon and Parker own Bowlmor, formed in 2000, along with Goode Bowling Corp., a subsidiary of private equity firm Goode Partners.

As also reported, first lien claims are to be repaid in full, in cash. According to an amended reorganization plan the company filed on June 24, the company resolved a final dispute with an ad hoc group of first-lien lenders, which had argued that they were entitled to payment of additional default interest of about $2.7 million.

According to the amended plan, first lien claims will be allowed in the amount of about $213.8 million in principal, plus interest at the non-default rate through June 8, and interest at the default rate from June 9 through the effective date of the plan. The amended plan did not specifically state what amount of additional monies the default interest would comprise.

AMF filed for Chapter 11 on Nov. 13, 2012, with a pre-arranged plan that would have exchanged first lien debt for equity in the reorganized company, subject to better offers. Second lien lenders challenged that plan from the get-go, however, claiming that the company’s bidding procedures were onerous and designed to insure that no rival offer would be submitted (see “AMF 2nd-lien lenders blast company’s plan, aim to make own offer,” LCD, Dec. 5, 2012).

Meanwhile, unsecured creditors also signaled their intention to challenge the pre-arranged plan on valuation grounds, arguing that unsecured creditors would fare better under a liquidation than under the proposed plan (see “AMF discovery fracas may preview valuation challenge to plan,” LCD, April 15, 2013). The pre-arranged plan would have paid unsecured lenders, with claims the company estimated at $30-35 million, $300,000, a recovery of 0.1% or less.

Under the confirmed plan, unsecured creditors are to receive a pro-rata share of $2.35 million. – Alan Zimmerman


Dish Network withdraws tender offer for Clearwire shares

Dish Network this afternoon withdrew a tender offer for Clearwire shares at $4.40 per share after abandoning its bid for Clearwire’s majority shareholder Sprint Nextel, according to a company statement. Recall that last week Clearwire’s special committee evaluating the rival takeover plan recommended that shareholder accept Sprint’s offer to buy the shares of Clearwire it doesn’t already hold at $5 per share.

The cancelled takeover plan ends roughly six months of jockeying by the company’s just as Sprint was targeted by Japan’s SoftBank. That deal is going through after Sprint shareholder approval yesterday, filings show.

In a related move as reported last week, Dish extended a special redemption feature on recent new corporate issuance since bailing on the Sprint takeover effort. Under terms of the special provision baked into the indentures, the company will redeem a $1.25 billion issue of 5% notes due 2017 at par, and a $1.35 billion issue of 6.25% notes due 2023 at 101, according to a company statement.

Dish issued the paper last month via issuer subsidiary Dish DBS to potentially support the takeover in the face of a rival bid for Sprint by Japan’s SoftBank. Issuance was at par apiece, the BB-/Ba3 paper was steady on the break, and levels hovered around the redemption price in recent weeks despite the market turmoil. Indeed, market players peg the 5% notes at either side of par and the 10-year tranche at 100.5/101.

The debt was issued on May 15 via bookrunners Barclays, Jefferies, Macquarie, and RBC, and settlement was May 28. As for the special call provision, settlement was Monday, June 24, according to SEC filings.

SoftBank itself issued $2.845 billion of 4.5% notes due 2020 in April to support the takeover of Sprint. The paper was rated BB+/Ba1 at offer, but has since been upgraded to BBB/Baa3. Issuance was par, the tight end of talk, but the touch market climate has taken it down to a 94 context in recent weeks, offering about 5.6%, according to sources. – Matt Fuller


Fridson Monthly: What if the 10-year Treasury yield hits 4%?

Note: The following is a special monthly report from Marty Fridson. For Marty’s latest weekly column, for LCD News subscribers, please see “Security selection – improving yet wide,” published on June 25, 2013.

This report addresses the potential impact on major asset classes of a rise in the 10-year Treasury yield to 4%, assuming no change in short-term interest rates. Our analysis points to drops of 17.5 and 14 points, respectively, from present levels of the high-yield and investment-grade indexes. The S&P/LSTA Leveraged Loan 100 Index’s three-year yield would increase from 4.86% currently to 7.81% under our assumptions, while the P/E multiple on the S&P 500 would rise moderately, from 15.62x to 17.36x. An important caveat is that we base our analysis solely on the yield curve implied by the 4% Treasury assumption. The actual outcome would depend importantly on economic conditions at the time of the Treasury rate increase. We believe our projected outcomes are most consistent with stagflation.

The question

A leading question on investors’ minds at present is the expected impact on major asset classes of a sharp rise in Treasury bond yields. The question has become especially pertinent in view of suggestions that the Fed may soon begin tapering its purchases of long-dated obligations. Those who ask the question generally assume that the Federal Reserve will keep short-term interest rates low for an extended period. We therefore formulate the question as follows:

“How will high-yield bonds, investment-grade corporate bonds, leveraged loans, and equities react if the 10-year Treasury rate increases to 4% with no change in the three-month T-bill rate?”

We plan to address this question as forthrightly as possible. For the record, however, the question is technically unanswerable. No conditions other than the three-month and 10-year Treasury rates are specified. Expectations for numerous other factors perennially influence the valuations of the major asset classes, e.g., economic output, the inflation rate, the speculative-grade default rate, and corporate earnings. A wide range of values for bond yields, loan yields, and stock multiples could be consistent with a yield curve of 4.00% (the 10-year rate assumed in the question) minus 0.03% (the present three-month yield) equals 3.97%, depending on where the other valuation factors stood.

In particular, any answer we provide to this study’s question will be challenged on grounds that we have presumably interpreted the yield curve’s message. A 397 bps slope from three months to 10 years might reflect a booming economy that has driven up demand for capital and, as a consequence, long-term bond yields (see note 1). Alternatively, the greatly elevated long-term yield could indicate that inflationary expectations have taken hold, while the still-low short-term yield could mean that credit demand remains weak. Those two contrasting scenarios, among others one might devise, would lead to very different valuations of our four asset classes.

The results presented below are probably most consistent with the latter scenario, which can be aptly termed “stagflation.” A stagnant economy combined with high inflation is the worst of all worlds for investors. That dire scenario is in line with the very wide credit spreads that our analysis produces. Other analysts will get a different answer if they assume that a different set of conditions will accompany the 397 bps yield curve implied by the question. In short, we do not represent that we have come up with the only possible answer to the question. We do believe, however, that our approach will help investors assess the risks associated with a drastic steepening of the yield curve.

The yield curve and valuation

There are three classic interpretations of the yield curve, all of which have a bearing on asset valuation. The liquidity preference hypothesis holds that investors inherently value liquidity and demand a premium for sacrificing the superior liquidity of short-dated instruments to move into longer-dated instruments. Clearly, an intensification of the demand for liquidity, as occurs most dramatically in a financial crisis, adversely affects the valuations of longer-dated corporate bonds.

The expectations hypothesis maintains that a positively sloped yield curve results from expectations of a rise in short-term interest rates. Therefore, if the expectation sets in that the Fed will boost short-term rates, longer-dated corporates will suffer. Leveraged loans, on the other hand, should benefit on a relative basis from expectations of rising short-term rates.

Finally, the segmented market hypothesis contends that each investor confine concentrates fixed-income investments in a certain maturity range. In this view, the yield curve at a given time reflects the sum of investors’ maturity preferences. If the segmented market interpretation is correct, longer-dated corporates will be disadvantaged by a migration of investors toward shorter maturities, based on factors such as their evolving liability-matching needs.

Based on the expectations view that the yield curve reflects expected future increases in short-term interest rates, which are a function of the expected future demand for credit, a positively sloped yield curve is commonly interpreted as a predictor of a pickup in the economy. If investors expect the economy to improve, they will tend to shift assets from corporate bonds to stocks. Widening of credit spreads and escalation in P/E multiples will probably follow.

The present, highly assertive U.S. monetary policy is not covered by any of the hypotheses enumerated above. In the past, financial economists did not consider the possibility that the Federal Reserve would ever deliberately keep short-term interest rates below zero in real terms for an extended period. If long-term rates, which are less controllable by the Fed, rise in response to inflationary fears, the result will be a steep yield curve created under circumstances unlike any in the past.

This is all to point out that history may not guide us precisely in predicting the financial markets’ response to the 397 bps yield curve implied by the question we seek to answer. Keeping that caveat in mind, we proceeded as follows: We determined the historical relationships between the yield curve and valuations in our four major asset classes. Based on those relationships, we calculated expected valuation indicators under conditions of a 397 bps yield curve.

Historical relationships

To determine the historical relationships, we ran simple regressions using monthly data from December 2002 to December 2012. We did not use available data going back to December 1996 because P/E multiples appeared to be in a different (higher) regime in the earlier period.

In each case the explanatory variable (x) was the Treasury yield curve from three months to 10 years, based on the respective BofA Merrill Lynch indexes (note 2). The response variables were:

The correlations (R) between the yield curve and the valuation measures were as follows:

These correlations were strong enough to justify analyzing the valuation measures as a function of the yield curve, even though valuations were also greatly influenced by other factors. (Not surprisingly, the debt instruments were more closely related to the yield curve than stocks were.) As discussed above, we did not attempt to predict how other variables (GDP, CPI, the speculative-grade default rate) would look when the 10-year Treasury rate hit 4%.

Our analysis produced the following regression formulas:

Applying the analysis

We ran these regressions using the present yield curve (2.48) as x and derived the predicted valuation factors shown below, along with the actual values.

Consistent with the findings of our five-factor value model (note 3), the present high-yield spread is considerably tighter than its estimated value. At the other extreme, the P/E multiple on stocks is less than one point above its predicted value. On the whole, the yield curve by itself does a respectable job of explaining how the major asset classes are currently valued.

Encouraged by these results, we next used 3.97 as the x variable (yield curve), based on the premises of this study’s question, namely, the 10-year yield at 4.00% and the three-month yield unchanged from the June 21 level of 0.03%. Running the regressions produced the following results:


Our analysis is based exclusively on the yield curve. It predicts that if the 10-year Treasury yield rises to 4% while short-term interest rates remain unchanged, the option-adjusted spreads on high-yield bonds and investment-grade corporates will widen to levels last seen in September 2009. That was just three months after the end of the most recent recession, as determined by the National Bureau of Economic Research.

The predicted spreads represent widening of 342 bps for high-yield and 88 bps for investment-grade. The spread for each issue in the corporate indexes is calculated off its corresponding Treasury maturity, but the weighted-average underlying Treasury yields (calculated as yield-to-worst minus option-adjusted spread) closely approximate the five-year rate for high-yield and the median of the five-year and 10-year yields for investment-grade. Based on the present curve, this exercise’s assumed 4% 10-year Treasury yield can reasonably be taken to imply a 2% five-year yield and a 3% median between the five-year and the 10-year.

Adding the estimated Treasury yields and predicted spreads produces yields of 10.51% on high-yield and 5.54% on investment-grade. Those figures represent yield increases from present levels of 6.62% and 3.43%, respectively. Based on the two indexes’ effective durations of 4.50 and 6.54, respectively, the move to a 4% 10-year rate will inflict price drops of 17.5 points on high-yield and 14 points on investment-grade. Total returns, under these assumptions, will depend on how long it takes the 10-year yield to reach 4%.

As a major caution, the high-yield spread was only 437 bps and the investment-grade spread was only 98 bps on May 13, 2004, when the yield curve reached its observation-period maximum of 389 bps, just six basis points shy of the level assumed in our analysis. This reinforces our earlier point that factors other than the yield curve can make a big difference. By contrast, six months after the end of the last recession, on Dec. 31, 2009, the yield curve was 378 bps and the high-yield OAS was 639 bps. Keep in mind that we envision the combination of a 0.03% three-month yield and a 4.00% 10-year yield as a highly inhospitable, stagflationary climate. Others may make a different assumption and therefore reach a different conclusion.

The leveraged loan market’s floating-rate nature will represent a comparative advantage, as its three-year yield rises by 295 bps under our assumptions. Note that as recently as October 2011 the yield was as high as our yield-curve-based analysis predicts under conditions of a 4% 10-year Treasury yield.

Stocks actually improve from a 15.62x to a 17.36x multiple, according to our analysis. Remember, though, that equities are the least sensitive of the four asset classes to changes in the yield curve. By extension, the P/E ratio will be more influenced by non-yield-curve factors than will be the case for fixed-income spreads and yields.

The level of the S&P 500, under our assumptions, will depend on the E in P/E – that is, the earnings of the constituent companies at the time that the 10-year Treasury yield reaches 4%. As a point of reference, however, consider the case in which earnings remain flat between now and then. Under those conditions, the S&P 500 will rise from its June 21 level of 1,562 to 1,736. Stagflation could be consistent, however, with a slight increase in the P/E ratio that coincides with a sharp drop in earnings. We note that the S&P 500 plummeted by 42% between 1970 and 1972 as the original U.S. stagflation got underway.

Martin Fridson, CFA

CEO, FridsonVision LLC

Research assistance by Xiaoyi Xu


1. A yield curve of 397 bps would be slightly steeper than the steepest curve in our observation period (Dec. 31, 2002 to Dec. 31, 2012), which was 389 bps on May 13, 2004.

2. Source: BofA Merrill Lynch Global Research, used with permission.

3. See “Determining fair value for the high-yield market” (Nov. 13, 2012).


4th HY deal struck from calendar this month, American Equity postpones $250M senior offering

American Equity Investment Life Holding has postponed its proposed $250 million senior notes offering, citing market conditions. The SEC-registered deal was originally expected to price late last week.

The West Des Moines, Iowa-based annuities and life-insurance firm was looking for funds to purchase its 5.25% convertible notes due 2029 tendered in connection with an offer to exchange the debt for cash and shares, to repay all amounts outstanding under an RC, to pay related fees, and to fund general corporate purposes.

The eight-year (non-call four) notes had netted issue ratings of BB+/BB from S&P and Fitch. J.P. Morgan was sole bookrunner on the deal.

This is the fourth high-yield deal that has been struck from the calendar this month due to the deteriorating market conditions. Yankee Candle canceled a dividend recapitalization transaction and first-timer Warren Resources opted to shelve a deal to refinance bank debt. Last week Federal Mogul canned a broad refinancing effort that included $750 million of senior notes, although that deal wasn’t yet in the market. – Jon Hemingway


Tenet Healthcare nets financing for Vanguard acquisition

Tenet Healthcare has committed financing from Bank of America to support its proposed acquisition of Vanguard Health Systems in a $4.3 billion transaction, including assumption of $2.5 billion in debt. The Dallas-based hospital operator will acquire Vanguard for $21 per share in an all-cash transaction.

Financing includes a $1.8 billion senior secured term facility as well as a $2.8 billion senior unsecured bridge loan, all or a portion of which will be refinanced in the high-yield market, according to an SEC filing. Proceeds will be used to fund the merger consideration, to refinance Vanguard debt, and to pay fees and expenses.

Outstanding debt includes $1.175 billion of 8% notes due 2018 and $725 million of 7.75% notes due 2019. Trading in the 2018 paper this morning has been up over two points on the day, in the 106.5-106.75 range, trade data show. In the loan market, Vanguard’s existing TLB due 2016 (L+275, 1% LIBOR floor) is quoted at 100/101 this morning on the news, sources said. By comparison, the loan was bid at 100.5 on Friday. The existing term loan totaled about $1.09 billion as of March 31, according to an SEC filing.

Shares of the Nashville-based operator of acute-care and specialty hospitals, which closed Friday at $12.37, surged on the news this morning, gaining approximately 65% in early trading, to $20.74.

The acquisition is subject to customary closing conditions and regulatory approval and is expected to close before the end of the year. Boards of both companies have approved the deal. With the acquisition, the combined entity will operate 79 hospitals and 157 outpatient centers in 16 states. The combination will create $100-200 million of annual synergies, according to Tenet. Management expects to lower leverage to 4.75-5x within the first year after closing.

Blackstone in 2004 purchased a majority equity interest in Vanguard from Morgan Stanley Capital Partners and management. The company completed its IPO in 2011 and Blackstone retains an approximate 37.7% stake in the company, according to S&P Capital IQ.

Tenet is currently rated B/B1 and Vanguard has a B/B2 profile. Outlooks for both companies are stable from both agencies. – Jon Hemingway/Kerry Kantin


European high yield funds see €251M cash outflow; YTD inflow: €3.75B

J.P. Morgan’s weekly analysis of European high-yield funds shows a €251 million outflow for the week ended June 19. Of that total, €51 million is attributable to ETFs. The reading for the week ended June 12 was revised to a €366 million outflow, from a €360 million outflow. The provisional reading for May is a €722 million inflow. That compares with a €555 million inflow in April, a €579 million inflow in March, a €156 million outflow in February, and a €1.8 billion inflow in January. The latest estimate for total inflows this year through May is €3.75 billion.

A total of €927 million has flowed out of the market over the past three weeks (by the weekly reporters), which represents the longest consecutive stretch of outflows since May/June last year. Bond prices are under pressure as government bond yields rise on the back of Bernanke’s comments that QE will begin to be tapered before year-end. Falling prices are hitting returns – with high-yield now showing a 1.75% positive return for the year, down from nearly 8% in mid-May, according to sources – and as a result money is leaving the asset class. At the same time primary issuance is struggling, highlighted by a second deal being postponed – Intralot – in a fortnight.

In the U.S. there was a net outflow of $333 million from high-yield mutual funds and exchange-traded funds in the week ended June 19, according to Lipper, a Thomson Reuters company. However, the reading is based on $512 million of mutual fund redemptions against an inflow of $179 million to ETFs, for the first inverse reading of this type for four weeks. Net outflows are significantly down from the $3.3 billion outflow the week before and the prior week’s record $4.63 billion withdrawal, but the latest figure is the fourth consecutive negative reading, for an outflow of $9.1 billion over that period. The year-to-date outflow is now $6.7 billion.

Retail-cash inflows into bank loan mutual funds and ETFs swelled to $1.4 billion for the week ended June 19, according to Lipper FMI, a division of Thomson Reuters. The inflow builds on the prior week’s $1.35 billion infusion, but most notably it extends the streak to 53 weeks for a whopping $33.1 billion of inflows over the one-year span.

As reported, J.P. Morgan only calculates flows for funds that publish daily or weekly updates of their net asset value and total fund assets. As a result, J.P. Morgan’s weekly analysis looks at around 50 funds, with total assets under management of €10 billion. Its monthly analysis takes in a larger universe of 90 funds, with €27 billion of assets under management. For a full analysis, please see “Europe receives HY fund flow calculation.” – Luke Millar