Monte Carlo Money: Modelling the Dangote Fertilizer Plant
Dr. Kaase Gbakon
Business Analytics|Financial Analytics|Data Science and Strategy Development|Economic Modelling|Commercial Intelligence
Key Takeaways
INTRODUCTION
In a 2016 article posted here, Sara Menker CEO of Gro Intelligence pointed out that large-scale fertilizer production investments are rare in Sub-Saharan Africa. While 20% of the world's arable land is in SSA, the region consumes around only 2% of fertilizers. This disconnect is seen in the intensity of fertilizer use: globally, the average fertilizer use is 135kg/Ha, compared to 17kg/Ha in SSA.
Against this backdrop, the Dangote fertilizer project was crystalized to address the issue of food security in Nigeria, which is implied by the low fertilizer usage highlighted. In the discourse on the “Dangote projects” in Lekki, so much focus is placed on the refinery, with relatively little mention of the fertilizer project.
This article will draw attention to the Dangote fertilizer plant, and I invite you to join me in assessing its value by the numbers.
PROFILE: DANGOTE FERTILIZER
Dangote Fertilizer Plant is situated on a parcel of land approximately 500 Ha in size in the South-East Quadrant (SE) of Lekki Free Zone which lies in the South-East portion of the LFZ. The product from Dangote Fertilizer Project is 7700 MT per day of Granulated Urea.
The Complex is identified by the following functional blocks:
1. Process Units:
2. Utility Units:
3. Offsite Units:
4. Bulk Urea Handling Units: Unit Identification Capacity
MODELLING ASSUMPTIONS…
To proceed with the financial modelling of the fertilizer plant, I will set out some assumptions on urea price, costs, and the applicable fiscal terms.
Urea Price
The price of urea is an important driver of fertilizer plant economics as it forms the source of revenue for the plant. I will be running two kinds of models here:
a.??? Deterministic
b.??? Probabilistic (Monte-Carlo simulation)
Under the deterministic model, I have assumed a urea price of $400/mt as seen in this graphic from tradingeconomics .
For comparison, I found that the average urea price in Nigeria was estimated as $0.99/kg ($990/mt) in a paper by Camila Bonilla Cedrez , Jordan Chamberlain , Zhe Guo and Robert J Hijmans. ?Of course, this is the market price, and thus my $400/mt assumption as the fertilizer plant gate price passes the reasonableness test. However, the 2020 paper emphasizes that fertilizer price in SSA is far more expensive than in any other region… that is another fish to fry!
For the probabilistic modelling, urea price exhibits volatility, and I use time series probabilistic models in Figure 2 below to capture that behaviour for forecasts.
Now that is something my data science folks can cheer about!
The graphic shows the historical Black Sea urea prices (fob) from 1999 to 2023. I have relied on the @RISK excel add-in from Lumivero to model probabilistic forecast of urea prices to 2034.?
Fiscal Regime
Like the Dangote refinery in my previous article, the location of the Dangote fertilizer plant in the Lekki Free Zone (LFZ) implies that it is exempt from taxes. According to Section 8 of the Nigeria Export Processing Zones Act (2004):
“…approved enterprises operating within Free Zones shall be exempt from all Federal, State and Local Government taxes, levies and rate.”
Further, Section 18(1) of the NEPZA provides that
“… legislative provisions pertaining to taxes, levies, duties and foreign exchange regulations shall not apply within the Zones…”
However, if the fertilizer plant were not located in a trade-free zone, it would be treated according to the provisions of Section 302(6) of the Petroleum Industry Act (2021), Section (39) of the Companies Income Tax, as well as relevant provisions in the Finance Act (2023).
The relevant fiscal provisions are codified in Table 1.
While the location of the fertilizer plant in the LFZ exempts it from taxes, the estimation of government receipts allows me to estimate the cost of the incentives enabled by the NEPZA and enjoyed by the project.
Financing
In my assessment of the Dangote refinery, I assumed it was 50% debt-financed according to the erstwhile governor of the Central Bank of Nigeria. The 2015 fertilizer plant ESIA expected a 60% debt financing on a $1.50 billion CAPEX. The actual CAPEX turned out to be $2.50 billion.
I will align with the words of the erstwhile Governor.
Further, I will assume an interest rate of 8% as my estimate of the blended rate and tenor of 10 years. This is summarized in Table 2.
领英推荐
Fertilizer Plant Attributes
Table 3 summarizes the attributes of the fertilizer plant relevant to the financial model. Key attributes such as the CAPEX are already known. According to this article in African Business Magazine by Dulue Mbachu , construction of the fertilizer plant commenced in 2014. However, I have modelled its construction start date as 2017 and production start as 2022.
I called on the S-curve method in cost engineering to estimate the CAPEX schedule.
I assume a long-term capacity utilization of 90% and an annual on-stream day of 350 days.
I assume variable OPEX of $30/mt and fixed OPEX of $30 million.
Natural gas is a key feedstock for the urea plant, and I estimate that the process requires 175 mmscfd of gas.
Even though foreign exchange is an important factor in the fertilizer plant’s assessment, I have not explicitly modelled its impact. I have assumed revenues and costs in a single USD currency.
Natural Gas Price
Table 4 summarises the natural gas price assumptions.
My gas price assumptions draw from the consideration that the fertilizer plant is classified as a strategic industrial sector, to which gas feedstock pricing is defined in the Petroleum Industry Act, 2021. I have assumed what, in my estimation, is the higher end of gas prices to the strategic sector as articulated in the PIA.
Relevant sub-sections of 167, 168 and the fourth schedule of the PIA are key to the gas pricing to fertilizer plant.
I have drawn from ENR Advisory that the tariff currently applicable to the natural gas transmission network is US$0.8/mmscf. This is the cost of transporting gas to the fertilizer plant.
As for the gas processing fee of $0.54/mmscf, I took my best guess, to be honest!
RESULTS…
Drum-roll … and now for the outcome!
Deterministic Results
I have provided the fertilizer valuation results with and without the LFZ's tax exemptions. My deterministic modelling indicates that the Dangote fertilizer plant delivers an NPV10 of $3.29 billion with the LFZ's tax exemption benefits. Without the tax exemptions, its value would be down to $2.17 billion.
Table 5 provides some more detail.
Note that if the fertilizer plant were subject to tax, I estimate that total government receipts over the 20-year time horizon of the model would amount to $2.61 billion. The fertilizer plant’s total cash flows would be $6.65 billion instead of the $9.27 billion estimated in its current tax-exempt jurisdiction.
I've included for you a graphic showing the project's net cash flow with and without the tax exemptions.
Without the benefit of the FTZ, the fertilizer project net cash flow (represented as the blue line) would decline from 2027 due to the commencement of tax payments, five years after production start in 2022. However, because taxes do not encumber the project, a higher level of net cash flows, represented by the red line, is possible.
Monte Carlo Results
I focused my uncertainty analysis on the urea price as I consider this to be the most significant uncertainty in the fertilizer project. Certainly, other uncertainties could have been modelled such as the OPEX, gas feedstock price, gas processing fee, and fertilizer demand – but not today, folks.
Below is a plot of the distribution of the fertilizer NPV with and without the tax benefits of the EPZ.
The histograms show the distribution of project NPV with the benefits of the EPZ compared to without the benefits of the EPZ. Under the EPZ terms, there is a 90% likelihood that the fertilizer plant will return an NPV10 between $449 million and $9,620 million.
If the plant were situated outside the EPZ, the likelihood of achieving an NPV10 in the above range drops slightly to 88.4%. There is only a 2% likelihood that the fertilizer project delivers an NPV10 of more than $9,620 million under its current fiscal provisions.
I have introduced Table 6 to enable us to quickly compare the project values with and without tax and their likelihoods of occurrence.
Based on this table, there is a 50% likelihood that the fertilizer plant returns an NPV10 of less than $3.78 billion without tax. However, if the plant were taxed, there would be a 50% likelihood that the NPV10 would be at most $2.50 billion.
In probabilistic terms, there is ~ 43% likelihood that the fertilizer plant will return an NPV lower than the deterministic NPV10 of $3.29 billion. In other words, there is a 57% likelihood that the fertilizer plant will exceed the deterministic NPV10.
IN SUMMARY…
Financial modelling of the fertilizer project yields an NPV10 of $ 3.29 billion, which would have been $ 2.17 billion if the project were not in a tax-exempt jurisdiction. Meanwhile, under NEPZA, the government has given up $ 2.62 billion in lifecycle receipts from the project, which has propped up the fertilizer plant’s cumulative cash flows over the 20-year model horizon to $ 9.27 billion.
Over the period I modelled, the Dangote fertilizer plant will utilize 992 bcf of gas, 40 times the amount of gas utilized by the neighbouring refinery. Additionally, I reckon that the fertilizer plant will emit an average of 9 million tCO2eq/annum, which, together with the 8 million tCO2eq/annum from the neighbouring Dangote refinery is about 15% of Nigeria’s total emissions as of 2019.
Over the life of the model, the fertilizer plant will emit 162 million tCO2eq. If carbon pricing were implemented at $30/t, it would add $ 4.85 billion to the coffers of the Climate Change Fund (Section 15(1-e) of the Climate Change Act) – only if the fertilizer plant were not in a tax-exempt jurisdiction. However, a carbon price at this level, even if implemented in the EPZ, with all the tax incentives, will depress the economics of the fertilizer plant - reducing its NPV10 from $3.29 billion to $0.35 billion - a 90% drop in valuation!
PLEASE TAKE THIS AWAY
The multi-billion-dollar Dangote refinery, petrochemical and fertilizer complex represents a significant investment and a value proposition difficult to ignore.
I have provided both a deterministic and probabilistic view of the fertilizer plant’s valuation. With the benefit of the EPZ, the fertilizer plant can be expected to yield an NPV10 of $3.29 billion. However, when viewed through a Monte-Carlo lens, there is a 57% likelihood that this value will be exceeded.
While there is a 90% chance that the fertilizer NPV10 will lie between $0.45 billion and $9.62 billion, the most likely valuation of the Dangote fertilizer plant is $3.78 billion. How is that for a $2.5 billion facility?
Finance Professional
1 个月Wow...a detailed and easy to follow analysis. Welldone. It would be interesting to know how you arrived at a forward discount factor of 10. Also, what the WACC of the project funding is and if it's higher than 10%, whether the project will still yield a positive NPV.
University of Cambridge Graduate
8 个月This is a very Insightful and comprehensive yet easy to digest study. Thank you for sharing this!
Building policies and partnerships for clean energy deployment.
8 个月This is interesting Kaase. Coming in from a climate and overall policy perspective (which I know are not the key foci of the article but important for us to consider given the impact this project could have. I can't help but wonder if a tax break is a good approach for Dangote considering the business' monopolistic hold on key sectors and the Nigerian government's struggle with tax collection given a highly informal economy. In the absence of taxes levied, a carbon tax is surely appropriate to build adaptation infrastructure.
Geoscientist || Project Manager || Microsoft Certified Data Analyst
8 个月It would be interesting to see how raw material costs impact the plant's economics given its relationship and proximity to the refinery. Really good read as usual Kaase Gbakon PhD, especially with the probabilistic approach.