Daqo New Energy (DQ US)

Deep Value Oligopolist at the cusp of Industry Inflection...

Introduction

Daqo New Energy (US$1.4b mkt cap, US$2b net cash) is a deep-value/special situation inflection play. It is one of the top polysilicon producers in China (~20% market share). Polysilicon is the high-purity form of silicon used in solar panels to create photovoltaic cells, which convert sunlight into electricity. No matter how you slice and dice it, you are effectively getting paid to own a company that trades at negative enterprise value. Paying $20/share to own a business with net cash of $30/share is what I call a fire sale.

NEGATIVE ENTERPRISE VALUE!!! I imagine Benjamin Graham rolling in his grave, incredibly excited that such a no-brainer market inefficiency still exists in today’s age of quant-based hedge funds and sophisticated trading algorithms. You would think “easily screenable” deep value is dead!

• This brings me to a separate point: AI will never truly disrupt the asset management industry. Alpha is fundamentally a result of behavioral arbitrage, informational edge was long arb-ed away since the turn of 21st century but behavioral remains a constant which is why cycles exist → remember in 2022, 2023 and 2024 on how un-investable China dominated headlines? Just look at how narratives changed this year as the Hang Seng Index made new highs, where suddenly the pitch after a +50% move in the index becomes “valuations in China are reasonable”.

AI solves for analysis, not judgment

As per my investment framework, I always try to identify a confluence in macro and micro for high probability set-ups, and Daqo New Energy is at the cusp of inflection:

• Macro: 1st July marked an inflection point in China’s anti-involution campaign where it is now a central government directive. Anti-involution is about rooting out disorderly competition and race of bottom pricing plaguing many Chinese industries. The Central Financial and Economic Affairs Commission meeting on 1^st July was chaired by Xi himself, with premier Li and other members of the politburo in presence. The significance/symbolism parallels that of supply-side reform in 2016 where the primary focus then was on coal, cement and steel industries. I believe the primary beneficiaries this time around will be new energy industries with the worst over-capacity. Polysilicon prices (albeit up 30% from the bottom) are on average, still at industry cash breakevens, and there simply is no worst overcapacity backdrop in what is fundamentally an oligopolistic industry

• Micro: Best management team and balance sheet ($2b cash, no debt) in the industry, 2nd lowest cost producer of polysilicon and the only one that has a history of and intention to distribute significant cash to shareholders

Thesis/Set-up

(1) Anti-Involution accelerates capacity exit and marks a sustainable inflection in industry cycle recovery

Today, ongoing discussions for polysilicon industry centres around the creation of an industry acquisition fund/vehicle to accelerate the exit of obsolete capacity (a novel solution). This formalises the creation of a “polysilicon OPEC” where 6-7 producers are expected to remain from the current 19, production quotas will be managed and likely overseen by the National Development and Reform Commission (NDRC).

• Industry asset acquisition fund will be ~75b RMB in scale -> 80% funded by debt from major banks and 20% equity funded by industry players

• Goal is to acquire 1.5m tons of outdated/obsolete capacity (vs 3.5m tons industry nameplate capacity and ~1.5m tons of annual end demand) → shut it down permanently + potentially acquire some industry inventory as well (350k tons inventory today vs 200k tons average level)

  • The exact details and work around potential hurdles (eg. if insufficient players choose to be on the sellers’ side and/or do not agree on acquisition prices) are likely to be released come September

• Caveat is NDRC will likely manage industry prices between RMB 60/kg to RMB 80/kg range via production quotas allocated to the remaining cartel

  • Surplus industry profits if polysilicon prices are above >RMB 60/kg would be channelled to repaying bank debt used for the industry asset acquisition vehicle until debt is fully repaid

I believe this policy augments the cycle by accelerating the near-term upside (years 1-2) but caps the medium-term upside (years 3-5) until the debt from the asset acquisition vehicle is fully repaid, where polysilicon prices will likely fluctuate in a managed range of 60-80 RMB/kg in the interim (parallels to OPEC+ & oil in the $60-80/bbl range).

Regardless, this represents a seismic shift to industry supply and price response. Today, spot prices are at RMB 50/kg (+50% from lows of RMB 35/kg in June) but another +20% move is required to reach the new ‘price floor’ at RMB 60/kg.

• Whilst it has been an immensely long and painful wait for industry recovery given the brutality of Chinese capital cycles whereby barriers to exit are a feature, not a bug of state-led capitalism (local government backed SOEs are judged by employment/local GDP growth KPIs which prolongs downcycles even when it makes no economic sense to produce under breakeven profitability), there is light at the end of the tunnel

Polysilicon supply is 2x demand at 3.5m tons nameplate vs 1.5m tons consumption

(2) Daqo is amongst the lowest cost producers with a cash fortress, conservative management and history of shareholder friendly distributions

Daqo is the only pureplay polysilicon player in the industry with a pristine balance sheet to weather the storm. Its $2b cash and no debt position contrasts with peers’ (often fully integrated with lower margin downstream cell/module making components) high gearing at 10-11x net debt/ttm EBITDA.

Moreover, with commodity related industries, assessing the quality of assets (ie. position on the cost curve is also important). Daqo has one of the lowest cost operations in the polysilicon industry, aided by a structural cost advantage with lower grid costs (coal-powered generation) as their operations are based in Xinjiang and Inner Mongolia.

China electricity prices by province, Inner Mongolia and Xinjiang are in the lowest quartile

Industry cost curve, Daqo with 300k nameplate capacity and ~40 RMB/kg all-in cost of production

During the go-go years of 2021-2022 where polysilicon prices reached RMB 160/kg, Daqo returned ~US$500m in cash via buybacks to shareholders between 2022-2023 (-15% share count over 5 years). They are the only players within the polysilicon industry that have returned cash at this scale whilst maintaining a very conservative cash fortress to weather the storm.

• Today, they still have US$150m sitting offshore with a total of US$2b in cash. Whilst management have chosen to withhold cash as the downturn worsened into 2024, they have indicated that should polysilicon prices reflect a sustainable an industry recovery, they would likely resume distributions again

• Incentives of the founding Xu family is clearly aligned with minority shareholders, where majority of family wealth resides in their ~40% ownership of the US-listed Daqo ADR. Today, Xu Xiang (CEO/son) and Xu Xiaoyu (Deputy CEO/granddaughter) manage the business with 82-year old founder Xu Guangfu still sitting on the board

  Financials and Valuation

Historical Financials

From the anti-involution policy response and proposed industry asset acquisition vehicle, we can estimate the enterprise value of Daqo New Energy based on the target capacity the vehicle aims to acquire (75b RMB for 1.5m tons capacity = ~US$11b or US$7.5b per 1m tons capacity)

• Noting that US$7.5b per 1m tons implied is for the most outdated/obsolete capacity

• Newly commissioned capacity has EV/ton replacement costs of ~$12b per 1m tons, and would better represent replacement costs for remaining cartel/top 6 players

Daqo New Energy has 300k tons of capacity, no debt and $2b in cash. Using the industry acquisition fund’s implied EV/ton, Daqo’s capacity + cash translates to $4.2b EV vs its current EV of -$400m ($70/share implied vs $20/share today).

This implies a 3.5x upside minimally from current levels, with further earnings growth and retention from a new polysilicon upcycle 2+ years out and capital return/cash distribution momentum thereafter remaining a fully unpriced optionality/fat right tail

From a holding company perspective, Daqo New Energy (DQ US) also owns a 71% stake in their A-share listed operating subsidiary Xinjiang Daqo (688303 CH) which trades on the Chinese market and is valued at US$9b today. This means their stake alone is worth $6.4b, against US-listed Daqo’s market cap of $1.5b - implying a >4x upside!