How Buffett Is Changing The Future Of Berkshire's Float - From Insurance To Uncle Sam

| About: Berkshire Hathaway (BRK.A)


Warren Buffett is generating new sources of float in capital intensive businesses using depreciation.

Thanks to 'Uncle Sam', deferred taxes will soon surpass insurance as the primary float generator.

Understanding how deferred taxes work for fixed assets will be crucial to valuation going forward.

Why Buffett is spending large amounts of capex on BNSF and MidAmerican to grow tax float.

What is the intrinsic value of deferred tax float, and valuation implications for Berkshire Hathaway.

Insurance float is a concept that Warren Buffett has clearly explained as a source of enduring value for Berkshire Hathaway Inc (BRK.A, BRK.B). However, there are other sources of float that are becoming more important to Berkshire over time, but are not well understood by investors.

Deferred tax liabilities (DTLs) are one area worthy of focus, and while I know that tax is boring, you will realize Buffett is truly a genius when it comes to float, so bear with me as I try my best to explain how it works, as it is complex.

The first idea that might spring to mind for many people on deferred taxes is unrealised capital gains - to be clear, the focus of this article extends deeper than just this. Read on...

Uncle Sam (deferred taxes) are a large source of float for Berkshire

So why are we focusing on deferred taxes? Because the Owners' Manual says so...

"Berkshire has access to two low-cost, non-perilous sources of leverage that allow us to safely own far more assets than our equity capital alone would permit: deferred taxes and "float," the funds of others that our insurance business holds because it receives premiums before needing to pay out losses. Both of these funding sources have grown rapidly and now total about $135 billion." Warren Buffet (BRK Owners' Manual)

While insurance float has typically taken the limelight, the more 'boring' deferred tax liabilities have taken a back-seat but are becoming increasingly more relevant, growing rapidly from $11bn to $57bn during 2004-2013 and now comprise 43% of the total $135bn of float.

The $77bn of insurance float is now looking mature with Buffett flagging a softer inflationary-like (perhaps declining) outlook. DTL's look to be the next wave of new float, and it appears likely to overtake insurance float in the near future as its rapid growth continues.

The chart shows the two sources of float over time for Berkshire - insurance (in blue) and deferred taxes (in orange).

(image source: iinvestor)

So where exactly do deferred taxes come from and how is Buffett engineering Berkshire's future around this?

Deferred tax - incur today, pay later

Deferred tax liabilities arise on the balance sheet from temporary differences between taxes that will come due in the future and taxes paid today. Accounting rules require this difference to be recognized as a liability and therefore a deduction in calculating net worth. However, the deferral of this tax payment acts as a source of float, and the cash can be invested elsewhere until it is due to be paid.

The key types of deferred tax liabilities for Berkshire

Specifically for Berkshire, there are two key sources of deferred tax liabilities which are:

  • Unrealized investment gains - this source will be familiar to most readers and is effectively deferred capital gains tax on investments. As Berkshire tends to have long holding periods on stocks these deferred tax liabilities can be large. They act as a free loan from the government which doesn't have to be repaid until the investment is sold/realized (which could be never).
  • Fixed assets (PP&E) - this is a more interesting source of DTL for Berkshire and effectively arises from differences between book accounting and tax accounting of capital assets. For example, accelerated depreciation rates for tax accounting give rise to lower assessable income resulting in low tax payments. However, book accounting with less aggressive depreciation rates gives rise to higher profits and higher tax expense. This mis-match results in a deferred tax liability. Over the life of any fixed asset the tax paid on the IRS basis and the accounting basis should be the same, and only the timing differs. The chart below illustrates by example the higher depreciation for tax purposes versus book resulting in a DTL.

(image source: iinvestor)

Berkshire's Deferred Tax Liability mostly relates to fixed assets

The chart below highlights these two key sources of DTL float with unrealized investment gains (in blue) and fixed assets (orange). There is a third bucket (grey) containing other items net of deferred tax assets which is not material in size.

The key observation from the chart is that the primary source of DTL is fixed assets, with unrealised investment gains playing second fiddle. Fixed asset DTL's at $32bn are greater than the $26bn of unrealized investment gains, and are also growing more quickly.

You will also note the large step change in fixed asset DTL's in 2010 which occurred as a result of the BNSF (railroad) acquisition, and has been growing strongly ever since. This gives us a few clues as to what Buffett was thinking about float when purchasing BNSF and may hint to the nature of future acquisitions for Berkshire.

(image source: iinvestor)

As deferred capital gains taxes are largely self-explanatory I will not discuss them any further, let's hone our focus onto the fixed asset DTL's.

Astute readers will realize that fixed asset DTL's should ordinarily reverse over time as the mismatch between the accounting and tax basis converge for a single asset as it is depleted over its life. Therefore, it might seem that float is temporary in nature and not really enduring like insurance float which is revolving and permanent.

However, I think Buffett appears to have found a holy grail of DTL's - instead of being temporary float, the characteristics of Berkshire's DTL float appear long enduring and will grow strongly over time. There are two reasons why this is the case:

1. Buffett's long-life slowly depreciating fixed assets increase the tenure of DTL float

A typical fixed asset with short to medium term lifespan will have any deferred tax liabilities unwind over time. However, the large part of Berkshire's railroad assets (BNSF's tracks and roadways) have useful lives of up to 100 years. The Utilities and Energy assets (MidAmerican's generation, distribution, transmission and pipeline assets) have useful lives up to 80 years.

(image source: BRK 2013 Annual Report)

For financial accounting purposes, the depreciation rates for these particular assets are provided independently and approved by the regulators. Book depreciation for BNSF/MidAmerican is tracking around 3% of the gross assets - implying an overall average useful life for the book of 33 years.

The IRS tax basis would depreciate these assets at a much more accelerated pace (useful lives for a single asset much less than 80-100 years) resulting in higher depreciation and lower tax payable.

High income taxes in the financial accounts would give rise to long enduring deferred tax liabilities. These railroad and utility acquisitions which Buffett wants Berkshire to own for "the next 100 years" were intended to provide sustainable deferred tax float for Berkshire for decades to come.

2. Buffett is accelerating capital expenditures to fixed assets resulting in rapid growth in DTL float

Having identified a source of float in the capital intensive assets, Buffett's next trick is to make it grow thereby increasing its value. It is therefore no surprise that large components of Berkshires capital expenditures are being directed towards BNSF and MidAmerican - the two assets we identified above as being long-term DTL float generators.

Capex in BNSF/MidAmerican has a number of impacts:

  • It increases the level of fixed assets in long-term DTL float generators
  • Capex above depreciation will cause the DTL float to grow
  • As new capital depreciates quicker than existing capital given accelerated tax depreciation it creates more per dollar DTL float than the existing book.
  • Given the above, ever rising amounts of new capital expenditures will make the DTL float start to balloon (or should i say float?) upward

This is already happening, in 2011 BNSF/MidAmerican capex was $6bn, and Buffett is saying this will be $11.1bn this year in 2014, almost doubled in a few years.

(image source: BRK 2013 Annual Report)

When buying BNSF, Buffett was likely thinking about very long-term DTL 'float' and deploying very large (read 'huge') amounts of capital over time to grow that deferred tax float. He is hungry to deploy rising amounts of capex to these railroad and utility businesses:

"America's rail system has never been in better shape, a consequence of huge investments by the industry. We are not, however, resting: BNSF spent $4 billion on the railroad in 2013, double its depreciation charge and a single-year record for any railroad. And, we will spend considerably more in 2014. Like Noah, who foresaw early on the need for dependable transportation, we know it's our job to plan ahead." Warren Buffett (2013 Annual Report)

"Our railroad, utilities and energy businesses (conducted by BNSF and MidAmerican) maintain very large investments in capital assets (property, plant and equipment) and will regularly make capital expenditures in the normal course of business. In 2013, aggregate capital expenditures of these businesses were $8.2 billion, including $4.3 billion by MidAmerican and $3.9 billion by BNSF. BNSF and MidAmerican have forecasted aggregate capital expenditures in 2014 of approximately $11.1 billion. Future capital expenditures are expected to be funded by cash flows from operations and debt issuances." Warren Buffett (2013 Annual Report)

So watch this space… The pace of capex in the capital intensive businesses should see the DTL surpass the insurance float over the coming years.

So what could Deferred Tax Liability float be intrinsically worth?

As a relative starting point, we know Buffett values $1 of insurance float as having at least $1 of intrinsic value given its characteristics of being (1) enduring, (2) costless, and (3) growing.

It appears unrealised gains should largely fit this criteria over time if long-term investments are not sold and continue to perform. However, shorter term investments would not satisfy this criteria. I would expect that $1 of unrealised investment gains might be worth <$1 of intrinsic value, but probably not by alot given the majority of the DTL would reflect 'legacy' investments.

Fixed asset DTL's appear more akin to having all three characteristics similar to insurance float. The nature of the assets appear very long-term, costless and should grow. Again I would expect that $1 of fixed asset DTL's may be worth close to $1. You would need to build a 80-100 year DCF with the relevant tax and accounting depreciation rates for the Railroad/Energy assets to accurately assess this.

As a final data-point worth mentioning, Charlie Munger has previously commented on the value of Wesco's DTL's in his annual letters (numbers below are per share values).

  • 1997 - $102 of DTL was estimated to be intrinsically worth $25 = 25c in the $
  • 1998 - $127 of DTL was estimated to be intrinsically worth $30 = 24c in the $
  • 1999 - $99 of DTL was estimated to be intrinsically worth $20 or 20c in the $
  • 2000-2007 DTL's intrinsically worth "much less" or "much lower" than its face value

(image source: iinvestor)

While Munger's estimates for Wesco suggest 20-25c of intrinsic value per dollar of DTL, he also makes this comment below (repeatedly each year) which suggest his estimates may be too low in the context of Berkshire's DTL:

"Each dollar of book value at Wesco continues plainly to provide much less intrinsic value than a similar dollar of book value at Berkshire Hathaway. Moreover, the quality disparity in book value's intrinsic merits has, in recent years, continued to widen in favor of Berkshire Hathaway". Charlie Munger (2009 Wesco Annual Report).

So on balance, I think we can discount Charlie's Wesco example when looking at Berkshire given the quality differences, and my feeling is that the intrinsic value of a dollar of Berkshire's deferred tax liability could potentially be worth near a dollar.

Float-adjusted valuation for Berkshire

If we assume that DTL's can be added back to book value at par, the only missing piece is goodwill. Much like thinking about insurance float, there is a cost to acquiring it (ie the goodwill). Therefore, if we are adding back DTL to book value, we should also be subtracting any cost with acquiring that float (mostly related to the capital intensive businesses). So in my calculation I net off the goodwill for the Railroad/Energy division.

The results are indeed very surprising:

  1. Adjusting book value for both insurance and capital intensive float suggests a multiple of 1.35x book is warranted.
  2. This multiple has been amazingly very steady over the last decade - Buffett is carefully managing float over time
  3. Float-adjusted book value stands at $192K and has growth at 10%pa over the last decade
  4. Insurance float net of goodwill has compounded 6% over the last decade
  5. Deferred tax liability (float) net of capital intensive business goodwill has compounded 14% over the last decade

(image source: iinvestor)

(image source: iinvestor)

(image source: iinvestor)

Concluding summary

  1. Buffett has found a very unique way to generate float, which is not widely understood among investors.
  2. Insurance float's time in the sun has passed, the next wave will be about deferred tax liability float.
  3. Buffett uses long-life capital intensive businesses and depreciation to do this
  4. Future acquisitions are likely to be focused on acquiring and generating deferred tax liability float.
  5. Buffett is setting up Berkshire to utilize bigger amounts of float for decades to come - he is a genius.
  6. Investors should consider the value of deferred tax liabilities in determining intrinsic value for BRK.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.