LGIH financials

One of the concerns of the market regarding LGI Homes is the debt level. This and “valuation” are “sell” level according to S&P. Valuation would be a topic for another post. What interests me now is the debt burden. I did an amateur stress test and would appreciate feedback.

What is LGIH’s debt burden? They have convertible notes and a revolving credit facility. The convertibles carry 4.25% interest payable semi-annually and are in the amount of $85 million. Every quarter they need to generate $903,125 to service this debt. The revolver carries variable interest rate of 3.5% “plus LIBOR” which was 0.53% at September end. There are many U.S. LIBOR rates, but the one that matches September end is the 1 month LIBOR. The revolver is in the amount of $360 million and as of September end required $3.627 million per quarter to cover interest. The total interest for a quarter totaled $4,530,125.

So one part of the question has to do with the variability of the one month U.S. LIBOR rate. I looked back as far as 1998 and noted the September rate for even numbered years (I know, lazy note taker). Anyway, the rate cycled a couple of times, being high in 2000 (Bernanke throttling irrational exuberance?) and 2006 before the real estate bubble burst. The highs were 5.625% and 6.629%. After 2008 the rates were under 1% and we are now at 0.8%, I think—I didn’t write it down.

My purpose is not to predict the U.S. 1 month LIBOR for 2017-2018. I don’t really see an overheated economy needing the application of economic braking. But, this is a “what if” stress test, so I looked at 5%. Assuming many things for a first cut analysis, including the same level of debt, the quarterly interest burden increases to $8,553,125 (let’s say $8,5 million and get rid of the 7-place approximation).

The second part of the question looks at the sales level LGIH would need to generate the $8.5 million. Assuming they could maintain about the same margins (which the likely could not) and assuming that the sales and SG&A expenses remain the same, I believe that need a bit more than $135 million revenue which would be about a 37% drop from their Q3 sales. It is a “little” more complicated than that. The provisions for tax would drop 3 million but if margins are squeezed those $3 million disappear in a hurry.

But of course, for the LIBOR to go to 5%, mortgage rates would also be high. There is no linear relationship between the 1 month U.S. LIBOR and the 30-year fixed mortgage. The 1 month has changed rapidly, I assume in response to reserve bank overnight changes, whereas the 30-year moves slowly. There has been no “normal spread”. All that can be said is that the 30-year fixed mortgage rate has been higher than the 1 month LIBOR, not less than 1.2% higher, not more than 4+% higher (based on these bi-annual, September data). Don’t know what happened in intervening 23 month periods. I suppose that one could get data on new home sales for these same Septembers. Almost 11 p.m., I don’t think I need to do that right now.

I’ve just deleted a section on cash flow which anyway ended up inconclusive. And other stuff on continued growth through geographical expansion even if the pool of qualified buyers in existing markets shrinks due to higher interest rates. Conclusion…., don’t know. As compared to other homebuilders, I don’t think that the LGIH’s higher debt/equity justifies the p/e disparities.

I didn’t add yesterday, don’t plan to do so today (though I haven’t checked the market yet). My LGIH tank is full.

KC, who wishes he could convert his thoughts and research into a more useful post. Sigh.


One of the concerns of the market regarding LGI Homes is the debt level

I thought you made this case in another post with the same subject. In any case, if you are thinking it is the debt then you should compare the debt vs equity with other industry peer’s. I don’t deny debt servicing matters but that is mostly used for fixed income part of the measurement and has very limited use for equity part of the valuation.

In my opinion debt is not the concern here. Perhaps you may want to compare it with KBH, PHM, TOL, 3 companies with varying degree of marketcap, to see how LGIH debt compares to the industry.


I thought you made this case in another post with the same subject.

I have previously posted on the issue of LGIH debt level. However, the latest post quantified what the interest burden is compared to sales and operating income. As far as comparing debt levels of peers, I didn’t bother to share that information. But, for those interested, here is a list:

Taylor Morrison 37.3 Buy
MDC Holdings 38.2 Hold
WCI Communities 39.2 Strong Sell
Toll Brothers 39.7 Buy
CalAtlantic 42.4 Buy
Meritage 49.2 Hold
Century Communities 49.4 Sell
LGI Homes 51.8 Sell
KB Homes 60.5 Hold
William Lyon 61.6 Strong Sell
Beazer 67.4 Hold

As you can see, there is no cutoff dividing buy/hold/sell based on debt/capital percentages, but there are no buy recommendations with debt/capital percentage over 43%.

I did check monthly new home sales from 1963 to present to see what peak to valley drops have been. In the 60’s and 70’s the cycles lasted roughly 3 to 5 years. Then there was a 10 year low-to-low cycle from ’82 to ’92 followed by an 18 year cycle that bottomed in 2010. Since then we have been in a recovery mode. The low sales levels over the first 30 years were more-or-less 400,000 per month. In the 60’s, the highs were around 600,000 and in ’86 the high was 900,000. The highest high was in July ’05 at 1,389,00 which was followed by the lowest low in February ’11 at 270,000. That was a very low low from which we have recovered to just under 600,000. The average monthly new home sales for the period starting in 1963 is 651,580, so we are still below the average of the last 50+ years.

When there were frequent cycles, the peak to low drops were 33% to 50%. The 80% drop from to the 2010 low is sort of a long tail event.


LGIH has a median debt/capital ratio, but 25% higher than the better companies.
Last century-wise, the percent drop in new home sales that would cause LGIH difficulties were common.
Given the extreme low level of sales from which this cycle started, and given the average monthly sales over 50+ years (and corresponding population increase), a drop to 400,000 seems quite unlikely. And, if something happened to cause such a drop, LGIH would be just one of most companies in my portfolio to suffer stock price declines.

While the debt level is not great, it is a result of high growth. But if this is a reason why the market dislikes LGIH, I do not share the concern.

S&P shares the positive housing industry outlook. From the LGIH report:

In 2017, we think the US housing industry is in a position to show positive growth, based on recent housing starts, sales delivery of existing homes and existing home prices. Comments from managements of the publicly-traded homebuilders, as well as actual recent results, in terms of delivered homes, and average prices per home, support this view. We think rising mortgage rates may curb demand for households where affordability becomes an issue on how much house they can afford. However, the US economy remains strong with employment growth and consumer sentiment high. We see these as key macro drivers for new homes started.

And, from the Toll Brothers report:

Homebuilders see good demand for homes in the more attractive Sunbelt and coastal markets. Demand at the lower end, which often comes from the first time buyer, is beginning to show improvement with new household formation or apartment rental costs rising mid-to-upper single digit each year.
Most of the publicly-traded homebuilders have strong balance sheets with ample liquidity, and we think the companies have been prudent in new land acquisitions. Attractive property lots for future development are increasingly scarce in high demographic markets. Thus, homebuilders that have locked up several years supply of land will be in a good position to develop communities and deliver homes.

(At which point let me vent: “high demographic markets”. If the meaning is markets with highly attractive demographics, why not just say so? Demographics is the statistical data of a population. One assumes that nearly all markets in the U.S. have a quite complete data set on their populations, i.e.,high demographics. End of vent.)

Back to LGIH, the company shares many of the positive characteristics that S&P cherishes. They have several years supply of land, and the demographics of apartment dwellers plays to the strengths of LGIH.

I think what they don’t like, in reality, is that LGIH has a short history as a public company and it is small relative to most of the publicly traded home builders. The breadth and depth of the data and analysis of Toll Brothers compared to LGI Homes is indicative of this. They don’t like the risk they perceive with the lack of liquidity and resulting volatility and lack of track record for the management. On the other hand, young, fast growing, profitable companies are more attractive to most of us here. That makes a market. I’m putting this to rest now. I have 10% more than my target allocation in LGIH and I’ll stick with that. We shall see if the market has mis-priced LGIH or if I have erred in my analysis.



As far as comparing debt levels of peers,

You have drawn conclusion that debt is a concern but the factors that you are looking are not relevant or correlates to that conclusion. The reason I suggested to compare debt levels with peers is to see how it impacts the valuation. You could do that by comparing the above with PE. You would notice companies like KBH which has higher debt to equity still carries higher multiples compared to LGIH.

I did check monthly new home sales

Is this single family residence alone or includes multiple dwellings?

On comparing LGIH with Toll Brothers is not very appropriate. Toll was always land heavy and in recent years some of the purchase they have done also came with land heavy. Lastly, Toll caters to completely different market, i.e., higher end luxury. They will benefit from the new administration policies on cutting the tax rate vs the entry level markets getting hurt by increasing interest rates.

In fact if I have to buy one homebuilder I would buy Toll even ahead of DHI.

I seem to have been unable to frame the broader issue with which I am dealing, at least not sufficiently to communicate it to you.

It is not that I have a concern about LGIH’s debt level. From the beginning I have been trying to determine why the market disrespects LGIH, the stock, as represented by its having had the lowest or second lowest p/e of the 23 publicly traded homebuilders of which I am aware. Therefore, the inquiry has been based on reading the reviews of LGIH and listing concerns that have been raised. If a company is one of my top allocations, I want to know the bear case. I’ve had my head handed to me far too often in the last several years by purchasing shares of companies highly recommended by MF and on this board with high conviction. Trying to understand many of these businesses is like peering into a misty swamp at twilight, but homebuilding is, one would think, more of lemonade stand complexity. I have never been concerned about the debt level. I have been concerned that the market seems concerned and I have been concerned that the market seems concerned and I have not been concerned. One thing I probably have not communicated well is that there is this has been more in the nature of looking for the unknown unknown. After all, besides the crappy p/e, LGIH is heavily shorted. I think I understand the mechanics of the high short ratio, but that would be another post.

As to the housing sales data, that is for single family homes, www.tradingeconomics.com/united-states/new-home-sales.

As to comparing valuation (p/e) and debt level, I did not comment on that because I didn’t see a correlation. However, I do thank you for the comment because I made a chart of the collected data and discover there is a strong inverse correlation between debt level and p/e. There are two outliers, KB Homes (hold, 60.5% d/c, 14.26 p/e) and Willian Lyon (strong sell, 61.65 d/c, 12.4 p/e). But the inverse relationship is generally there and appears slightly exponential. Correlation -.786, if you will excuse data exclusion of the outliers.

Anyway, nice day today. Actually not raining. Time to take the old body out and get it dirty and sweaty.