Deferred Revenue and Profitability in Tesla’s Software-Integrated Model: A 
Case Study under ASC 606 
 
 
 
 
 
  
Liufei Li 
  
  
  
  
  
 
  
 
 
 
  
Supervisor: Ted Lindblom 
Master’s thesis in Accounting and Financial Management  
Graduate School, School of Business, Economics and Law 
University of Gothenburg, Sweden 
Abstract 
This thesis examines financial reporting considerations related to the Accounting Standards 
Codification (ASC) 606, which governs revenue recognition under U.S. GAAP. The study 
focuses on software-bundled business models in the electric vehicle (EV) sector, with Tesla 
Inc. as the primary subject of analysis. With growing vehicle software offerings—such as 
Full Self-Driving (FSD), premium connectivity, and over-the-air updates— traditional 
revenue recognition under a bundled sale is complicated by the fact that the company’s 
performance obligation is spread over time. 
Testing Tesla’s annual reports from 2013 through 2023, I use a quantitative approach to 
analyze the impact on reported profitability and earnings volatility of deferring revenues. 
The results reveal that Tesla’s increasing amount of deferred revenue, which primarily comes 
from software sales, is a short-term obstacle to profit but acts as a smoothing mechanism on 
earnings between periods. While software-driven revenue is still a small portion of the top 
line it's vastly out-sized in its strategic influence - contributing to margin expansion as well as 
financial predictability. 
Also, this study adds to the body of literature by uniting technical accounting guidelines with 
industry-specific strategic considerations in new sectors. It also highlights the requirement for 
ongoing academic and regulatory focus on how emerging business models interface with 
legacy financial systems. 
Keywords: ASC 606, revenue recognition, deferred revenue, Tesla, electric vehicles, 
software-integrated business models, profitability, earnings volatility, financial reporting 
 
 
 
 
 
 
 
Acknowledgements 
 
First of all, I would like to give a long due thank to my supervisor, Professor Ted Lindblom, 
who has provided me with excellent supports and much valuable comment during the time of 
the thesis. His intellectual gifts and encouragement have greatly influenced this research both 
in its direction and in its depth. I also appreciate the lecturers and administration staff at MSc 
in Accounting and Financial Management, University of Gothenburg. Critical 
thinking, analysis and research abilities have been developed in a stimulating academic 
environment with a supportive academic community. 
My deepest gratitude to my family and friends, for their unconditional emotional support, 
trust and understanding that have given me the strength to keep going. They have gotten me 
through my time of doubt and struggle. I would also like to thank myself — for enduring the 
uncertainties, staying committed, and bravely facing the challenges of living alone in a 
foreign country for the first time. This experience has not only tested my resilience but has 
helped me grow in ways I could never have imagined. 
Last, but not least, I am grateful to live and work in Sweden. This has been an invaluable 
experience; not just from an academic and professional perspective but on a personal level 
too. 
 
 
 
Liufei Li, Gothenburg, 2025-06-01 
 
 
 
Table of Contents 
 
1. Introduction 1 
1.1 Background 1 
1.2 Problem Discussion and Research Question 2 
2. Literature Review 5 
2.1 Accounting Standards of Revenue Recognition 5 
2.1.1 Five-step Model for Revenue Recognition 5 
2.1.2 Deferred Revenue Under IFRS 15 & ASC 606 6 
2.2 Revenue Recognition Challenges in Similar Industries 7 
2.3 Bundled Sales and Pricing Strategies in the EV Industry 8 
2.4 Revenue Treatments in Mature Markets and The Financial Impact of Revenue 10 
Recognition on Profitability 10 
3. Methodology 15 
3.1 Case Selection 15 
3.2 Data Sources 16 
3.3 Analytical Framework 17 
3.3.1 Revenue Composition Shifts 17 
3.3.2 Deferred Revenue Development 17 
3.3.3 Profitability Impact 17 
3.4 Simulation Design 18 
3.5 Limitations 18 
3.6  Statements on the Use of Artificial Intelligence (AI) 19 
4. Results and Analysis 20 
4.1 Overview of Tesla's Revenue Composition 20 
4.1.1 Automotive Sales Revenue 22 
4.1.2 Services and Other Revenue 23 
4.1.3 Energy Generation and Storage Revenue 23 
4.2 Development of Deferred Revenue 24 
4.3 Tesla’s Real and Simulated Financial Performance 29 
4.4 Results of the Simulation Cases 34 
4.4 Tesla as a Representative of Emerging Hybrid Models 37 
5. Conclusions 39 
Reference 42 
 
 
 
 
 
 
1. Introduction 
1.1 Background 
Revenue is a fundamental element of a company's financial statements and a critical indicator 
of financial performance. The International Accounting Standards Board (IASB) and the 
Financial Accounting Standards Board (FASB) have both revised revenue recognition 
standards by issuing IFRS 15 (Revenue from Contracts with Customers) and ASC 606 
(Revenue Recognition – Revenue from Contracts with Customers) to enhance transparency in 
financial reporting. These standards establish a structured five-step process for recognizing 
revenue, requiring firms to allocate transaction prices based on separate performance 
obligations and recognize revenue as obligations are fulfilled (IASB, 2022; FASB, 2016).  
IFRS 15 was issued by the IASB in May 2014 and became effective for annual periods 
beginning on or after 1 January 2018, while ASC 606 was issued by the FASB in May 2014 
as part of the convergence project and was similarly effective from 1 January 2018 for public 
entities with earlier voluntary adoption permitted  (IASB, 2018; FASB, 2018). At the same 
time, Electric Vehicle industry has been growing exponentially over past decades. Not only 
did the technology of EV companies reach new heights but higher government support and 
wider adoption of sustainability initiatives led to more flexible business models like sales 
bundling in EV companies. However, these new models present complex revenue recognition 
challenges, as companies must determine whether revenue should be recognized upfront (at 
the point of sale) or deferred over time. For example, Tesla offers its Full Self-Driving (FSD) 
software on a subscription basis, leading to deferred revenue recognition (Tesla, Inc., 2023), 
whereas Polestar follows a direct-to-consumer model, primarily recognizing revenue at the 
point of sale (Polestar, 2023). The choice between upfront vs. deferred revenue recognition 
has direct implications for profitability, earnings volatility, and financial transparency. It is 
also important to note that under both IFRS 15 and ASC 606, while revenue may be deferred, 
certain costs related to software development, such as general R&D expenditures, are 
typically expensed as incurred. However, specific costs that qualify as contract acquisition or 
fulfillment costs—such as incremental costs to deliver Full Self-Driving functionality or to 
support future over-the-air updates—will be capitalized and amortized over the same period 
as the related deferred revenue.  
1 
To go further, while the EV industry is currently in its growth phase, the long-term 
sustainability of these revenue models remains an open question, especially when the markets 
become mature. But for now, as we can see, industries that initially relied on hardware sales, 
have transitioned towards subscription-based revenue streams as their markets matured 
(Brynjolfsson & McAfee, 2014).  
Tesla’s so-called Full Self-Driving software is arguably one of the company’s most ambitious 
and possibly profitable innovations. As an industry report reveals, the figure shows Tesla 
potentially earning up to $75 billion from its FSD software alone by Goldman Sachs ( 2023) 
by 2030,  Which is for Tesla—not only as a technology but as an integral part of the 
company's long-term business model and valuation. A software-integrated model combines 
hardware products with software features, creating a hybrid offering that generates both 
upfront and recurring revenue.Tesla is changing and so is the nature of the revenue it reports 
over time as it moves towards a software based future, understanding how this plays a role, 
and how Tesla recognizes software generated revenue via the new accounting standard - ASC 
606 and reports it on the line by line basis will be an important part of the evaluation of its 
future based financial status, transparency, and investor focus.  
Finally, given the relative newness of the revenue recognition standards and the evolving 
nature of business models in the EV sector, this intersection represents a timely and 
underexplored area for academic and professional inquiry. A deeper understanding of how 
revenue recognition frameworks such as ASC 606 are applied in emerging 
industries—particularly in software-enabled EV business models—is essential for ensuring 
consistent financial reporting, enhancing investor confidence, and supporting long-term 
strategic planning. 
1.2 Problem Discussion and Research Question 
As the electric vehicle industry undergoes a rapid transformation, an increasing number of 
manufacturers—such as Tesla, NIO, and XPeng—are shifting from purely product-based 
models to integrated offerings that include software-based services such as autonomous 
driving, vehicle connectivity, and over-the-air updates. While these innovations offer new 
revenue opportunities, they also pose significant accounting challenges under the current 
standards. 
2 
While certain EV companies have begun exploring service-based revenue sources in their 
bundled sales contracts, the income that comes from non-hardware sales bundled with 
vehicle purchases remains a fairly minor source of overall profits. For example, for the year 
2023, Tesla reported total revenue of $96.77 billion, with deferred revenue of $3.54 billion 
from software subscriptions like Full Self-Driving, accounting for only 3.7% of total revenue 
(Tesla, Inc., 2024). Similiarly, in the fiscal year 2023, NIO’s total revenue was $7.83 billion, 
made up of $6.93 billion (88.6%) from vehicle sales and $900 million (11.4%), derived from 
other sales concessions (NIO Inc., 2024), including services like Battery-as-a-Service (BaaS), 
extended warranties, and charging solutions.  
However, while hardware revenues still dominate at the moment, Tesla’s revenue 
methodology shows a slow transition toward software revenues over time. Tesla was in 
hardware sales mode at first and then started adding software as a bundled package over 
time. Make Autopilot a standard feature in 2019 and change the pricing model (Tesla, Inc., 
2019). As the EV industry evolves from hardware-dominant revenue models to increasingly 
software-integrated business strategies, financial reporting for software-related income has 
become a growing area of concern and complexity. These offerings introduce significant 
challenges under ASC 606, which mandates the separation of performance obligations and 
the recognition of revenue when, or as, obligations are fulfilled. 
Specifically, when software functionalities like autonomous driving are continuously updated 
or gradually delivered, ASC 606 requires revenue to be deferred over the period of service. 
This deferral affects the timing of revenue recognition, influencing not only short-term 
profitability, but also earnings volatility and the transparency of financial reports.  
Moreover, although software-related revenue currently represents a small portion of Tesla’s 
total income, the company—and CEO Elon Musk in particular—has repeatedly emphasized 
that autonomous driving will be a cornerstone of its future value, with some forecasts 
projecting tens or even hundreds of billions in potential revenue from robotaxi services and 
Full Self-Driving software (Goldman Sachs, 2023). 
To be concluded, these facts have raised my interest in the fact that Tesla’s accounting 
choices for software-related revenue may have material consequences for how its financial 
performance is perceived. Against this backdrop, this thesis seeks to answer the following 
research question: 
3 
How do revenue recognition strategies under ASC 606 impact profitability and earnings 
volatility in software-integrated electric vehicle business models? 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4 
2. Literature Review 
2.1 Accounting Standards of Revenue Recognition 
IFRS 15 and ASC 606 are chosen as the primary accounting standards for this research 
because of their global significance and widespread adoption. Both standards were developed 
in a joint effort between IASB and FASB to improve comparability in financial reporting 
across industries and jurisdictions. They share the same five-step model for revenue 
recognition.  
Given the EV industry’s global nature, analyzing both IFRS and US GAAP ensures a 
comprehensive understanding of how revenue recognition policies impact financial 
performance and investor decisions across leading EV companies in different regions. 
2.1.1 Five-step Model for Revenue Recognition 
The Standards, IFRS 15 and ASC 606, are both applicable for electric vehicle companies that 
provide bundled sales, outlining a consistent set of principles to recognize revenue in diverse 
industries. These frameworks present a five-step model for revenue recognition： 
To begin with, a contract must be identified between the company and the customer. 
According to IFRS 15 (IASB, 2014, para. 10) and ASC 606 (FASB, 2016, ASC 606-10-25-1), 
a valid contract must create enforceable rights and obligations, be approved by the parties, 
and have commercial substance.  
Secondly, once a contract is established, it is crucial to identify the performance obligations 
within it. IFRS 15  (IASB, 2014, para. 22) and ASC 606 (FASB, 2016, ASC 606-10-25-14) 
define a performance obligation as a distinct good or service that the customer can benefit 
from independently or together with other resources.  
Thirdly, another key consideration is determining the transaction price, which represents the 
total consideration an entity expects to receive in exchange for delivering goods or services, 
as defined in IFRS 15 (IASB, 2014, para. 47) and ASC 606 (FASB, 2016, ASC 606-10-32-2).  
This price may consist of fixed amounts, variable consideration (e.g., performance bonuses, 
discounts), and any significant financing components.  
Fourth, once the transaction price is determined, it must be allocated to the individual 
performance obligations within the contract. According to IFRS 15 (IASB, 2014, para. 73) 
5 
and ASC 606 (FASB, 2016, ASC 606-10-32-28), the allocation should be based on the 
standalone selling price of each distinct component. If direct prices are not observable, they 
must be estimated using industry-accepted methods such as adjusted market assessment, 
expected cost plus margin, or residual approach.  
Finally, revenue is recognized when (or as) the entity satisfies a performance obligation, as 
stated in IFRS 15 (IASB, 2014, para. 31) and ASC 606 (FASB, 2016, ASC 606-10-25-23). 
This occurs either at a point in time or over time, depending on when control of the goods or 
service is transferred to the customer.  
The centrality of revenue recognition to both the timing of profit realization and the visibility 
of firm performance has been emphasized in prior research. Wagenhofer (2014) argues that 
the adoption of ASC 606 and IFRS 15 represents not merely a technical adjustment, but a 
fundamental shift in how financial outcomes are communicated to investors—particularly in 
sectors undergoing structural transformation, such as the electric vehicle industry. 
2.1.2 Deferred Revenue Under IFRS 15 & ASC 606 
In cases where revenue is recognized over time, specific criteria must be met to classify it as 
deferred revenue. Revenue is recognized over time if one of the following conditions is met: 
(1) the customer receives and consumes benefits as the entity performs, (2) the entity creates 
or enhances an asset controlled by the customer, or (3) the entity has an enforceable right to 
payment for completed performance (IASB, 2014; FASB, 2014). When revenue is recognized 
over time, progress toward fulfilling performance obligations is measured using either output 
methods, such as milestones reached, or input methods, such as costs incurred (IASB, 2014; 
FASB, 2014). 
For contracts involving subscription services, maintenance, or software updates, revenue 
must be deferred and recognized over the service period (IASB, 2014; FASB, 2014). The 
determination of whether revenue should be recognized at a point in time or over time is 
based on the transfer of control to the customer (IASB, 2014; FASB, 2014). 
This is especially relevant to the EV sector which include over-the-air updates and software 
subscription services as the control is not changing in one transaction but in serial feature 
activations. But where delivery is modular and has evolved, the current literatures give 
6 
limited direction on how to evaluate performance obligations – suggesting to me that there 
may be a gap that this paper attempts to fill. 
2.2 Revenue Recognition Challenges in Similar Industries 
The implementation of IFRS 15 and ASC 606 has introduced significant challenges for 
companies across various industries. Firms need to exercise considerable judgment in 
identifying performance obligations, estimating variable consideration, and determining when 
control of goods or services is transferred to customers (EY, 2020; Deloitte, 2016). These 
complexities are particularly pronounced in industries with bundled offerings, subscription 
models, and high variability in pricing, making revenue recognition a critical issue. 
Challenges in the Technology Industry: Complex Bundled Sales and Pricing Models 
With its new standard, ASC 606/IFRS 15, the Technology Industry faces the most challenges 
around revenue recognition in revenue allocation, and transaction pricing.  One of the main 
difficulties is appropriately identifying all different performance obligations in a contract. 
Revenue allocation challenge is particularly critical in firms that have bundled software with 
other services, such as design, hosting, post-contract customer support, and software 
upgrades (Deloitte, 2016; PwC, 2022). Under the new standard, companies must apply a 
robust revenue allocation such as Standalone Selling Price (SSP)-based model (PwC,2022). 
In addition, the estimation of the transaction price has become even more complex as 
businesses need to figure out what is the total consideration expected from the customer at 
the inception of the contract and reassess it throughout the contract period (Deloitte, 2016). 
These adjustments can be to account for discounts, rebates, performance bonuses and sales 
returns – they all imply clearly articulated methodologies for revenue attribution(Deloitte, 
2016). 
Challenges in the Automotive Industry: Variable Consideration and Warranty 
Obligations 
Revenue recognition complexities in the Automotive Industry stem from various 
performance obligations, sales incentives, and warranties (EY, 2020). One major problem 
identified in EY’s automotive sector review of IFRS 15 implementation is the specifics of 
original equipment manufacturer (OEM) sales, where the OEM business model typically 
includes free services (i.e., maintenance), extended warranties and price incentives that may 
7 
be considered as additional goods or services. Under IFRS 15, an assessment is carried out 
on whether these elements are separate performance obligations, influencing the timing of 
revenue recognition (i.e. either upfront or deferred).  Although EY’s report is primarily 
directed at traditional automotive companies, the themes related to variable consideration, 
product warranties and repurchase agreements identified here are relevant to the sector 
concerning electric vehicles.  
Supporting this view, Caylor (2010) examines strategic revenue recognition behaviors within 
the software industry, showing that firms may deliberately adjust revenue timing to meet 
earnings benchmarks. This highlights the potential for revenue recognition to serve not only 
transparency purposes but also earnings management objectives, adding another layer of 
complexity to the adoption of ASC 606 in software-integrated automotive models. 
In conclusion, previous reports highlight the challenges companies encounter when adopting 
IFRS 15 and ASC 606 (EY, 2020; Deloitte, 2016; PwC, 2022), especially in industries where 
sales are bundled together, where consideration is variable, and where long-term service 
commitments exist. Software related contracts remain application ones for the technology 
industry, and are also involved in the automotive industry with long term warranties and sales 
functions (Deloitte, 2016; EY, 2020). These complexities are further magnified in the EV 
sector, where companies not only sell vehicles but also generate revenue through software 
subscriptions and leasing models.  
2.3 Bundled Sales and Pricing Strategies in the EV Industry 
Bundled Sales at Different Phases 
According to Guiltinan (1987), there are two types of price bundling, upfront bundling and 
subscription bundling, and they impact revenue recognition differently. Upfront bundling 
applies to one-off sales, where you recognize revenue when each transaction is completed. 
On another hand, subscription bundling relies on services delivered over a period of 
time, which leads to revenue being deferred until it can be recognized as performance 
obligations are met. 
Moreover, as Guiltinan (1987) points out, price bundling has unique functions in market 
engagement and customer loyalty. This has the effect of lowering entry barriers for 
consumers and accelerating market adoption in the growth phase through the use of bundling 
8 
strategies. With maturity in the market, businesses focus on keeping their current customers 
engaged through a bundled product offering available on a recurring payment basis, allowing 
the companies to capture predictable revenue, impacting long-term financial viability. 
Subscription bundling is especially relevant for the purpose of this research, given the fact 
that Tesla provides software-related functionalities, for example the FSD, through 
subscriptions and bundled sales, resulting in the deferred revenue as per ASC 606. Price 
bundling is hence not only a marketing instrument but also a significant accounting issue that 
has a direct impact on the timing of revenue recognition and financial reporting. 
Pricing Strategies in the EV Industry 
Pricing strategy of a company can profoundly affect accounting treatment (including profit 
and revenue recognition) and financial performance. According to Kaplan and Atkinson 
(2015), pricing in a competitive market context is not only a marketing tool, e.g., a 
competitive weapon but also a determinant of a company’s revenue structure and accounting 
treatment. Companies need to ensure that their pricing strategies are align with the accounting 
standards to present accurate financial reporting and avoid financial misrepresentation due to 
pricing errors. 
Gupta and Lehmann (2003) discuss the importance of Customer Lifetime Value to price 
strategy, arguing that many firms focus on maximizing revenues over a longer term, as 
opposed to profit maximization over a shorter period. This long-term perspective aligns with 
modern business models, such as subscription-based pricing models and long-term contracts, 
which require a change in the way revenue is recognized, moving from one-time recognition 
to revenue recognition over a longer period of time. Such deferred revenue recognition is 
consistent with IFRS 15 and with ASC 606 (IASB, 2022; FASB, 2016). 
In the Electric Vehicle industry, many companies adopt bundled sales and subscription-based 
services, such as Tesla’s Full Self-Driving subscription and NIO’s Battery-as-a-Service model 
(Tesla, Inc., 2023; NIO Inc., 2023). Regarding this, companies need to allocate revenue using 
the SSP method, as required by IFRS 15 (IASB, 2014), ensuring that revenue is distributed 
based on the relative standalone selling price of each performance obligation. For instance, a 
company that offers discounts on bundled sales of vehicles and software must allocate those 
discounts proportionally across different performance obligations—including the vehicle, 
software, and warranty—according to IFRS 15 (IASB, 2014). Failure to properly allocate 
9 
revenue could distort profit margins and reduce financial transparency (Kaplan & Atkinson, 
2015), as improper revenue allocation practices may not accurately reflect the economic 
substance of the transaction. 
Despite the significance of this issue, existing research rarely investigates the financial 
reporting complexities that stem from dynamic pricing strategies in bundled EV offerings. 
This study contributes to the literature by linking pricing decisions with their downstream 
effects on revenue allocation and earnings presentation.. 
2.4 Revenue Treatments in Mature Markets and The Financial Impact of Revenue  
Recognition on Profitability  
Revenue recognition policies play a crucial role in shaping a company’s financial 
performance, earnings volatility, and investor confidence (Healy & Palepu, 2001). The 
adoption of IFRS 15 and ASC 606 has significantly changed how firms report revenue, 
affecting both short-term and long-term financial stability (IASB, 2022; FASB, 2016).  
As a result, companies must decide between immediate revenue recognition (upfront 
recognition) or deferred revenue models (subscription-based revenue streams), each of which 
has distinct implications for profitability and stock price stability. 
Revenue Treatments in Mature Markets 
As industries mature, companies must adjust their revenue recognition strategies to maintain 
financial stability and transparency (Healy & Palepu, 2001; PwC, 2022). 
During the growth phase, firms often rely on upfront revenue recognition to accelerate 
profitability (Lev & Thiagarajan, 1993; Dechow et al., 1995). However, as markets become 
saturated, an increasing number of companies shift toward subscription-based models and 
deferred revenue recognition to ensure sustainable profitability and enhance investor 
confidence (Gupta & Lehmann, 2003; Brynjolfsson & McAfee, 2014).  
Moreover, Lev and Zarowin (1999) argue that traditional financial reporting frameworks 
often lag behind the economic reality of firms transitioning toward service- and 
technology-driven business models. They call for an extension of reporting boundaries to 
accommodate hybrid revenue structures, which is increasingly critical as EV companies 
integrate software and subscription services into their core business models. Multiple studies 
10 
emphasize the importance of revenue recognition methods in mature markets and their impact 
on financial reporting and valuation (Healy & Palepu, 2001; PwC, 2022). 
Yet most of the works frame the changeover of revenue strategy as binary transitions between 
stages and cannot explain the hybrid situations in the modern industries. In the case of 
EV companies, they work with a dual logic: they need to indicate short-term profit, while 
creating long-term service-based income flows. This type of hybrid revenue model is not so 
well covered by existing literature, which often study traditional manufacturing or pure 
digital subscription firms in isolation. 
Profitability Predictability 
 Wagenhofer (2014) indicates that as a result of the new revenue recognition rules, more 
revenue is now recognized in the quarter in which the sales transaction occurs instead of 
being deferred to future periods. This change could increase earnings volatility, making a 
clear line between high-revenue and low-revenue quarters.  
Similarly, PwC (2022) discusses the complexity that arises when the contract includes 
multiple performance obligations and the transaction price must be allocated. With the new 
standards, a more accurate distribution of the recognition of revenue is done according to 
stand-alone selling prices as opposed to legacy accounting that can drastically affect the 
timing of revenue recognition. According to PWC(2022), this process has an impact on the 
predictability of profitability, where in the past revenue was recognized in a consistent way, 
now revenue recognition changes with the terms of the contract and deliverables. 
In industries with bundled services, the challenge is even greater. In the life sciences industry, 
Deloitte (2023) discusses how contracts containing multiple elements require careful 
assessment of whether bundled components qualify as distinct performance obligations. This 
determination directly influences when and how revenue is recognized, often leading to 
nonlinear revenue patterns across financial periods (Deloitte, 2023). Such revenue 
recognition challenges extend to other industries that rely on subscription-based, recurring, or 
deferred revenue models, including the electric vehicle (EV) industry. Companies like Tesla 
and NIO, which bundle software, maintenance, and hardware, face similar revenue allocation 
complexities (Tesla, Inc., 2023; NIO Inc., 2023). 
11 
Previous research findings highlight the increasing difficulty companies face in ensuring 
profitability predictability under the new revenue recognition standards. As industries shift 
toward bundled and subscription-based models, firms must carefully balance compliance 
with IFRS 15 and ASC 606 while maintaining financial stability and investor trust. 
The Importance of Transparency in Financial Reporting 
The research underscores the need for transparency in the financial reporting process, as it 
reduces information asymmetry, which is an essential determinant for the investor 
confidence. As from Healy and Palepu (2001), who contend that financial disclosures impact 
market expectations. EV companies would have to make their accounting practices clear, as 
essentially the industry is largely based on deferred revenue models for software 
subscriptions and battery leasing services, and investors would have to enter with predictable 
revenue streams. 
Piosik (2023) empirically demonstrates that the adoption of IFRS 15 has provided enhanced 
convergence of analysts' forecasts for revenue, operating income and net income. This 
implies that transparent revenue recognition in IFRS 15/ASC 606 can mitigate investors’ lack 
of confidence and information asymmetry, especially, if bundled sales and deferred revenues 
are more frequent in some industries. 
In addition to the impact on transparency and timing, deferred revenue recognition has been 
empirically linked to earnings management behaviors. Stubben (2010) finds that firms often 
manipulate the timing of revenue recognition—especially when approaching earnings 
targets—by recognizing revenue prematurely. Such strategies temporarily inflate earnings at 
the cost of future performance and reporting credibility. This reinforces the importance of 
examining Tesla’s deferred revenue treatment not only as a compliance issue under ASC 606 
but also as a potential instrument for earnings smoothing. 
Investor Confidence and Market Reactions 
While a company's revenue recognition policy impacts its financial performance, it can 
also have effects on investor trust and market valuation. According to Healy & Palepu 
(2001), because information asymmetry hinders investors from correctly estimating the 
actual profitability of a firm, the nature of financial disclosures directly influences market 
expectations.  
12 
Revenue recognition policies significantly affect a company’s profitability and affect 
investor confidence. Based on Dechow, Sloan, and Sweeney (1995), this paper discusses 
earnings management using accrual manipulation that misleads investors over the longer 
period while increasing the volatility of earnings. The results of their study indicated that 
companies undertaking aggressive revenue recognition, like early recognition of goods or 
services received prior to fulfilling performance obligations, could show a temporary 
improvement in earnings at the cost of long-term weakening financial situation and lowering 
of investor credibility. Applying this framework to the EV industry, they must carefully 
balance upfront revenue recognition (e.g., immediate software sales) versus deferred revenue 
models (e.g., subscription-based services like Tesla's Full Self-Driving, FSD). Similar to how 
Dechow et al. (1995) discussed that aggressive earnings management can mislead investors 
in the short term while reducing long-run stock price stability, EV companies that prioritize 
short-term revenue boosts over sustainable revenue streams may experience market distrust 
and valuation fluctuations. 
Just as an observation in the EV industry, what that means is companies like Tesla are 
increasingly going to have to figure out how to weigh the short-term financial attractiveness 
of recognizing revenue upfront (e.g., recognizing revenue on software all at once) with the 
long-term predictability of a more traditional deferred revenue model (e.g., 
subscription-based features like FSD). Although financial markets reward short-term 
behavior, the full costs of overly aggressive revenue recognition practices and the loss of 
investor trust and the negative valuation consequences outweigh any perceived advantages 
for EV rms. This research suggests that the scrutiny of revenue recognition is most 
pronounced in innovation-intensive industries where business models shift quickly and 
investor attention is intense. 
Overall, prior literature and consulting reports highlight the growing complexity and 
strategic importance of revenue recognition in hybrid product-services industries. Though (or, 
more accurately, because) both IFRS 15 andASC 606 offer an industry agnostic, principles 
based model, application of the standard remains a current issue, especially in the context of 
the EV industry, due to the nature of bundled offerings, variable consideration, and 
performance obligations of long term. In addition, prior research shows that aggressive or 
smooth revenue recognition distorts measures of profitability and erodes investor confidence, 
particularly in fast-changing, innovation-based markets. 
13 
These considerations call for investigating the gap between accounting standards and 
behavior in the real financial world. Drawing on Tesla as a specific example, the study seeks 
an understanding of the utilization of ASC 606 in the context of software-related revenue for 
EV business models and to explore the implications where deferred revenue strategies 
influence earnings volatility. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14 
3. Methodology 
This study adopts a single-case design because it enables an in-depth, longitudinal 
examination of a complex accounting issue—how software-integrated revenue models are 
recognized under ASC 606 in the electric vehicle sector. As Yin (2018) argues, a  single case 
study is particularly effective when the objective is to understand a real-life phenomenon 
within its actual context. Given the research question’s focus on the financial reporting effects 
of ASC 606 in software-integrated EV models, a simple but deeply contextualized case study 
is methodologically well-suited and provides rich empirical evidence for answering the 
research question with clarity and precision. As Siggelkow (2007) emphasizes, well-chosen 
case studies are not only valuable for illustrating complex mechanisms, but also persuasive in 
theory-building when the case is particularly revealing or representative. Tesla’s strategic 
positioning and transparent disclosures make it a compelling and informative example, 
capable of offering broader insights into how ASC 606 is applied within emerging hybrid 
business models. 
Given this study’s focus on the real-world financial reporting practices of Tesla, a single-case 
approach is methodologically appropriate and allows for a detailed, longitudinal analysis. 
Moreover, this research follows an inductive case study logic (Eisenhardt, 1989), whereby 
patterns and theoretical insights emerge from empirical observations rather than being tested 
through predefined hypotheses. This approach is particularly appropriate for industries 
undergoing structural transformation, such as EV firms integrating software-based services 
into their traditional product offerings. 
3.1 Case Selection 
According to IEA (2023), Tesla, together with BYD, accounted for over 30% of global EV 
sales in 2022, highlighting its industry-defining position. More importantly, Tesla represents a 
pioneer among electric vehicle manufacturers transitioning from a traditional hardware-based 
business model to a hybrid structure that incorporates recurring software revenues, such as 
Full Self-Driving (FSD), premium connectivity, and over-the-air updates. 
By the end of 2023, Tesla reported approximately $3.54 billion in deferred 
revenue—primarily attributed to software-related features—which makes it particularly 
relevant for studying the financial and strategic implications of ASC 606. While several other 
EV firms (e.g., NIO, XPeng, Rivian) are also experimenting with software-enabled services 
15 
and subscription models, Tesla remains the most mature and transparent in terms of publicly 
disclosed financial data under U.S. GAAP. Tesla’s SEC filings (10-K, 10-Q), investor 
presentations, and detailed footnotes on revenue recognition provide a rich and consistent 
longitudinal dataset necessary for conducting robust empirical analysis. 
The 10-year time frame (2013–2023) was purposefully chosen to span the transition from 
Tesla’s hardware-only revenue structure to a hybrid model that incorporates deferred revenue 
under ASC 606, which became effective in 2018. This time span allows for a longitudinal 
assessment of both pre- and post-adoption financial patterns, enabling the study to capture the 
full impact of the standard's implementation on profitability and earnings volatility. 
3.2 Data Sources 
Raw data was collected manually through Tesla’s annual reports (Form 10-K) of 2013 - 
2023. This 10-year period was purposely chosen to span from the time that transformed the 
company from a hardware-oriented business model to a software-based revenue structure, 
and to see how this would eventually affect the company in the long run after adopting ASC 
606. The year 2013 to 2017 is the pre-ASC 606 reporting period of Tesla under the legacy 
U.S. GAAP, and the year 2018 is the transition year to the ASC 606, hence also the 
benchmark year. Between the time after 2018 and 2023, I examine trends in deferred 
revenue, the growth of software-related revenue, and Tesla’s reporting behavior over time.  
There was special focus on key financial line items, such as “Automotive Sales,”  “Deferred 
Revenue,” and gross profit margins, along with related footnotes, particularly those on 
revenue recognition policies and deferred revenue disclosures.  
Data were extracted and structured into Microsoft Excel spreadsheets. The financial 
metrics for each year were put in different columns and the related line items were written in 
rows. By adopting this tabular structure, I could horizontally compare fiscal years, making it 
easy to observe trends in revenue mix, deferred revenue growth, and profitability dynamics. 
The Excel model was the basis for the visual inspection, summary statistics, and 
scenario-based explanations used in the analysis afterwards.   
Naturally, the empirical basis for the case study includes the following data: 
● Tesla’s annual SEC filings (10-K, 10-Q) from 2013 to 2023 
16 
● Tesla’s investor relations publications and earnings call transcripts 
● Official pricing information  
● Reports from investment banks and research institutions (e.g., Goldman Sachs, 2023) 
● Authoritative literature on ASC 606 and IFRS 15 (e.g., Deloitte, PwC, EY) 
3.3 Analytical Framework 
This study employs a structured analytical framework to evaluate Tesla’s accounting 
treatment of software-related revenue under ASC 606. The choice of this framework is 
motivated by the need to comprehensively understand how Tesla’s evolving revenue model 
influences financial performance and reporting transparency in the context of industry 
transformation. The framework is built around four interrelated dimensions, each chosen to 
capture a critical aspect of Tesla’s financial reporting strategy and its broader implications: 
3.3.1 Revenue Composition Shifts 
The study analyzes the evolution of Tesla’s revenue segmentation, tracking the relative 
weight of different revenue streams—automotive sales, services, and software-related 
income. This enables an evaluation of Tesla’s transition from a hardware-centric model to a 
hybrid model incorporating recurring, software-driven revenue. 
3.3.2 Deferred Revenue Development 
In the analysis section I will trace the development of Tesla’s deferred revenue from 2018 to 
2023, this starting point was chosen not only because ASC 606 took effect in 2018, but also 
because Tesla began providing disaggregated disclosures of deferred revenue specifically 
related to software components—such as Full Self-Driving and over-the-air updates—only 
from that year onward. The analysis focuses on the annual additions, recognized portions, and 
year-end balances, as well as their proportions relative to automotive sales and gross profit. 
Rather than quantifying direct financial impact, the study highlights how deferred revenue 
has accumulated over time and how its gradual recognition pattern reflects Tesla’s long-term 
software delivery strategy under ASC 606. 
3.3.3 Profitability Impact 
The analysis examines the structural role of deferred revenue from software offerings in 
Tesla’s evolving revenue model. By tracking the growth of deferred revenue balances, the 
17 
proportion of recognized income over time, and their relationship to total automotive sales 
and gross profit, this study identifies how deferred revenue contributes to smoothing earnings 
and reshaping profitability patterns. 
In short, by applying these dimensions, the study bridges accounting theory with real-world 
reporting behavior, offering a case-based understanding of how an innovation-driven EV firm 
implements revenue recognition standards in a changing business environment. 
3.4 Simulation Design 
In addition to the empirical analysis of Tesla’s financial data, this study also conducts a set of 
simulation analyses to further examine the financial implications of different revenue 
recognition strategies under ASC 606. In the simulation design, Company A recognizes all 
software-related revenue upfront at the time of vehicle delivery (i.e., no deferral), whereas 
Company B defers the software revenue evenly over a five-year service period, in compliance 
with ASC 606.  
The simulation spans a 15-year period, assuming a vehicle selling price of $400 and an 
optional software add-on priced at $50, with initial sales of 100 units in the first year. The 
cost of goods sold is set at approximately 75% of the selling price. A discount rate of 5% is 
applied to calculate the net present value of profits. Three sales growth scenarios are tested: 
0% (stable growth), +10% (expansion), and -5% (contraction), in order to provide a 
comprehensive view of how revenue timing impacts profitability trends, earnings volatility, 
and long-term value creation. The objective is to offer a structured comparison framework for 
understanding the financial consequences of deferred revenue strategies under 
software-integrated business models. 
3.5 Limitations 
While this study provides a detailed and data-driven case analysis of Tesla's application of 
ASC 606 in the context of software-based revenue, it is not without limitations. First, the 
research focuses on a single firm—Tesla—which, although representative of 
innovation-driven business models in the EV sector, may not reflect the practices or financial 
impacts experienced by other firms, especially those with different market structures or 
regulatory environments. 
18 
Secondly, the analysis does not explicitly incorporate non-financial factors such as political 
pressures, regulatory actions, or reputational risks, which may significantly influence both 
Tesla’s strategic direction and its accounting treatments in the future. For example, in recent 
years, Elon Musk’s public stance on geopolitics and governance has attracted scrutiny, 
particularly in Europe, where public opinion and government sentiment toward Tesla have 
shown signs of shifting. These broader socio-political dynamics, while beyond the scope of 
this financial analysis, could potentially impact the company’s long-term business model and 
the relevance of its current revenue strategy. 
3.6  Statements on the Use of Artificial Intelligence (AI) 
In accordance with the guidance provided by the School of Business, Economics, and Law at 
the University of Gothenburg regarding the use of generative AI in higher education, this 
thesis transparently acknowledges the use of AI tools in its development process. Generative 
AI tools, including ChatGPT by OpenAI,  and Grammarly were used for spelling and 
grammar correction to enhance the overall readability and quality of the document. 
The author takes full responsibility for the integrity and originality of this thesis, and 
confirms that all AI-assisted outputs were thoroughly evaluated and integrated in line with 
academic standards and ethical guidelines. 
 
 
 
 
 
 
 
 
 
 
19 
4. Results and Analysis 
This section summarizes the empirical finding derived from Tesla's financial data from 2013 
to 2023, including a focus upon the nature of revenue composition over the 10 years, the 
impact of deferred revenue changes, and the financial consequences of revenue recognized 
according to ASC 606. The goal is to find the effect of Tesla’s software-based revenue model 
on reported profitability and earnings volatility through time. 
To summarize, the prominent trend emerged from the result is that Tesla has a growing 
balance of deferred revenue, which is mainly driven by the access to their FSD Capability 
features and their ongoing maintenance, internet connectivity, free Supercharging programs 
and over-the-air software updates primarily on automotive sales (Tesla, Inc., 2018–2023). 
However, the level of deferred revenue actually turning into income each fiscal year remains 
comparatively low (Tesla, Inc., 2018–2023), indicating a gradual-attunement and cautious 
revenue-recognition strategy for long-standing software deliverables. Ultimately, these 
findings directly address the core research question of this thesis by illustrating how Tesla’s 
revenue recognition approach under ASC 606 — particularly the deferral of income from 
software features like FSD — affects both profitability and the volatility of reported earnings. 
To enhance the understanding of these dynamics, I will first present Tesla’s evolving revenue 
composition in the following subsections and then provide a more in-depth characteristic of 
deferred revenue development overtime. The next section discusses the impact of income 
deferral on profitability and earnings volatility, also, to address the research question and to 
be more general, followed by a discussion of the relevance of Tesla as a poster child of 
nascent hybrid business models in the EV sector. 
4.1 Overview of Tesla's Revenue Composition 
Tesla’s total revenue has grown rapidly and continuously over the last 10 years, from around 
$2.0 billion in 2013 to $96.77 billion in 2023(Tesla, Inc., 2013–2023). This extraordinary 
growth signals the evolution of the company from niche EV manufacturer to worldwide 
leader in automotive, energy, and technology-focused solutions. Beyond the headline 
numbers, the composition of Tesla’s revenue — specifically where, and how, it is earned — 
has also changed significantly. 
20 
 
Figure 1 Tesla’s Total Revenue (2013–2023) 
Note. Data source: Tesla Inc. 10-K filings (2013-2023) 
 
 
Figure 2 Tesla’s Revenue Breakdown by Segment (2017–2023) 
Note. Data source: Tesla Inc. 10-K filings (2017-2023) 
As Teala’s  total revenue growth, from figure 1 and 2, we can see that while automotive sales 
remain the company's critical revenue driver, the nature of these revenues has grown 
21 
significantly heavier and more diverse. And Tesla does not only behave as a product 
company anymore — its revenue streams have diversified into embedded software features, 
recurring service income and energy generation & storage. Most of these newer streams 
bundled into hardware sales or provided as post-sale add-ons, creating new ASC 606 
accounting continuing complexity (Tesla, Inc., 2023; Deloitte, 2016). 
To better characterize the nature of Tesla’s evolving business model, this section 
disaggregates Tesla’s revenue into four major buckets: (1) automotive sales, (2) services & 
other revenue, (3) electrical power generation and storage. Because each offers a distinct 
interpretation, by dissecting each category individually, my goal is to lay down a solid 
foundation of what to expect in the upcoming examination of deferred revenue, and the 
timing and volatility of Tesla’s reported results as a consequence of software-integrated 
offerings. 
4.1.1 Automotive Sales Revenue 
 
Figure 3 Automotive Sales and Its Percentatge of Total Revenue (2013–2023) 
Note: Automotive Sales includes cash and financed vehicle deliveries, as well as bundled 
software features such as Full Self-Driving (FSD), premium connectivity, and OTA updates. 
Data from Tesla, Inc. 10-K filings. Data source: Tesla Inc. 10-K filings (2017-2023) 
22 
As figure 3 illustrates, Tesla's automotive sales revenue has consistently been the most of the 
company's total sales for the past decade. And 2013 was nearly all vehicle delivery revenue 
(Tesla, Inc., 2013, Form 10-K). In 2023, automotive sales totaled $78.51 billion, accounting 
for about 81% of Tesla’s overall revenues (Tesla, Inc., 2023).  
Automotive sales revenue section is not limited to simply delivering vehicles. From Tesla’s 
10-K disclosures: “Automotive sales revenue includes revenues related to cash and financing 
deliveries of new vehicles, including access to our FSD Capability features and their ongoing 
maintenance, internet connectivity, free Supercharging programs and over-the-air software 
updates.” (Tesla, Inc., 2023, p.39). This means that software-related income — like Full 
Self-Driving and premium connectivity and OTA updates—are included with automotive 
sales, not reported separately.  
However, this practice of embedding software-related revenue within the automotive sales 
category, although compliant with ASC 606’s revenue deferral guidelines, somewhat 
obscures the actual contribution of software operations to the company’s profitability. 
High-margin digital services such as Full Self-Driving and premium connectivity are 
valuable components, yet they are not reported separately in the financial statements, making 
it difficult for external investors to accurately assess the growth and profitability of Tesla’s 
software segment. 
4.1.2 Services and Other Revenue 
Service and Other Revenue section primarily consists of non-warranty after-sales services, 
used vehicle sales, and retail merchandise, it also encompasses paid Supercharging and 
vehicle insurance. From Tesla's 10-k filings: Free or paid Supercharging, if included with 
vehicle purchases, is sometimes included as part of software packages such as Full 
Self-Driving and is subject to the deferral accounting guidance under ASC 606 (Tesla, Inc., 
2023). And so while most of the segment isn’t software income related, its partial overlap 
with digital services makes it pretty relevant to both the broader Tesla hybrid revenue thesis 
and the story around deferred revenue recognition. 
4.1.3 Energy Generation and Storage Revenue 
While energy generation and storage remains a minor segment in terms of total revenue, its 
growth reflects Tesla’s broader diversification strategy (Tesla, Inc., 2023). However, this 
23 
category does not involve deferred revenue or ASC 606 complexities, and thus falls outside 
the scope of this study. 
4.2 Development of Deferred Revenue  
From 2013 to 2023, Tesla’s deferred revenue has seen a steadily increasing trajectory, with a 
more noticeable acceleration after the adoption of ASC 606 in 2018. This trend coincides 
with Tesla’s growing emphasis on software-based features such as Full Self-Driving and 
premium connectivity, which are often sold as part of bundled packages with its vehicles. 
Although deferred revenue appears on the balance sheet as a liability, its long-term 
accumulation reveals key insights into the company’s revenue recognition strategy and the 
evolving structure of its hybrid business model  (Tesla, Inc., 2018- 2023).  
Starting in 2018, Tesla began separately disclosing deferred revenue (Tesla, Inc., 2018–2023). 
Features such as FSD are not delivered at a single point in time, but rather through 
incremental over-the-air updates, feature releases, and maintenance services provided over an 
extended time horizon. As stated in Tesla’s annual financial disclosure about deferred 
revenue account, deferred revenue in their annual reports includes the access to the FSD 
Capability features and their ongoing maintenance, internet connectivity, free Supercharging 
programs and over-the-air software updates primarily on automotive sales (Tesla, Inc., 
2018–2023). This disclosure highlights that these components are treated as distinct 
performance obligations and must be recognized over time in accordance with ASC 606, 
rather than immediately upon vehicle delivery. 
24 
 
Figure 4 Tesla Deferred Revenue Overview — Inflows, Outflows, and Year-End Balances 
(2018–2023) 
Note. Data source: Tesla Inc. 10-K filings (2018-2023) 
Regarding the figure 4 of Tesla’s deferred revenue over the six-year period from 2018 to 
2023, which includes the annual additions (inflows), recognized revenue (outflows), and 
year-end balances. As it shows in figure 4, the deferred revenue balance increased 
consistently year over year, from under $1 billion in 2018 to over $3.5 billion by the end of 
2023. Meanwhile, the recognized revenue from the deferred revenue account remained 
relatively low compared to the inflows (additions), resulting in a steady accumulation of 
deferred revenue. This widening gap between additions and recognized amounts suggests a 
slow realization pace of software-related revenues under ASC 606, reflecting Tesla’s 
long-term performance obligations such as FSD and premium connectivity. 
 
25 
 
Figure 5 Recognized Deferred Revenue as % of Total Automotive Sales (2018–2023) 
Note. Data source: Tesla Inc. 10-K filings (2018-2023) 
From figure 5, despite the increasing absolute balance of deferred revenue, the recognized 
deferred revenue consistently accounted for less than 1.5% of total automotive sales each 
year. After peaking at around 1.35% in 2019, the percentage gradually declined and stabilized 
below 0.8% from 2021 onward. This trend presents the fact that although Tesla collects 
substantial upfront cash from software-enabled sales, the actual income recognition is slow 
and incremental, resulting in only a modest short-term contribution to reported sales figures  
(Tesla, Inc., 2018–2023).  
26 
 
 
Figure 6 Price Evolution of Tesla’s Full Self-Driving (FSD) Software (2017–2024) 
 
Note Prices are in USD. Data compiled from Tesla’s official announcements, media reports, 
and industry analyses. Sources include Tesla, Inc. (2016–2024), InsideEVs, Electrek, and Not 
a Tesla App. 
 
 
 
Figure 7 Tesla Annual Vehicle Deliveries (2013–2023) 
Note. Data source: Tesla Inc. 10-K filings (2013-2023). 
27 
Moreover, as figure 6 and 7 described, while Tesla’s car deliveries and software subscription 
pricing have both clearly trended higher over the last 10 years, deferred revenue recognized 
has not risen in tandem. From 2018 to 2023, the annual number of Tesla vehicle deliveries 
increased significantly — from just over 245,000 in 2018 to over 1.8 million in 2023 (Tesla, 
Inc., 2018, 2023). At the same time, the cost of Full Self-Driving has risen by a factor of 
several (e.g., from $5,000 in 2019 to $15,000 in 2023, or $199/month on subscription) 
(Tesla, Inc., 2023; Not a Tesla App, n.d.). That, in turn, would theoretically result in higher 
recognized revenue over time.  
However, empirical evidence indicates an opposite practice. Annually only a small portion is 
recognised, as we discussed before, usually less than 1% of total automotive sales —and this 
proportion has not shown any significant increase over time (Tesla, Inc., 2018–2023).  
This flatlining of revenue is evidence that, while Tesla may be selling more of its vehicles as 
carriers for high-margin software, the revenue from software itself is effectively being pushed 
further out over time, not revving up in lockstep with increased vehicle sales. The widening 
rift between inflows and outflows of deferred revenue is evidence of a conscious revenue 
recognition decision under ASC 606 which serves to emphasize income being smoothed out 
in an effort to create predictable revenue once more into long-term predictability while 
sacrificing short-term profit levels. 
This gap further highlights a key tension in Tesla’s financial reporting: While the company’s 
business is increasingly oriented around software-related revenue, its accounting approach 
could mask the full financial effect of that shift. For investors and analysts, this means 
getting a better grasp of how deferred revenue accounting not only skews profitability metrics 
but also skews how efficiently and how much it looks like Tesla is growing. 
In conclusion,this form of accounting helps ensure that Tesla’s financial statements do not 
reflect too much revenue in early periods. It also sends a message to investors and analysts 
that the company is focused on transparency and regulatory compliance — rather than 
aggressive revenue reporting. This conservative approach to timing could help to create more 
accurate and realistic forecasting models as well as increase investor confidence in long-term 
profitability — all the more important for a company whose value increasingly rests on 
software performance and digital service delivery (FASB, 2016; Healy & Palepu, 2001; 
Wagenhofer, 2014; PwC, 2022). 
28 
4.3 Tesla’s Real and Simulated Financial Performance 
Building on the above analysis and discussion of Tesla’s increasing deferred revenue, I take a 
closer look at the financial impact of deferred revenue on profitability and the variability of 
earnings in this section. 
 
 
Figure 8 Deferred Revenue Balance vs. Gross Margin (2018–2023) 
Note. Data source: Tesla Inc. 10-K filings (2018-2023) 
 
As shown in figure 8, the balance of deferred revenue for Tesla increased from $2.44 billion 
in December 2018 to $3.54 billion in December 2023, and the amount recognized as income 
annually was less than 1% of total revenue during that period (Tesla, Inc., 2023). Tesla’s 
gross margin during the same period, however, saw slight variation, hitting between 17% to 
25%, with significant growth in vehicle deliveries - from approximately 245,000 vehicles in 
2018 to over 1.8 million in 2023.  
This practice, though technically compliant and strategically sound, results in a temporary 
understatement of financial performance. High-margin software offerings such as Full 
Self-Driving are effectively "parked" on the balance sheet as liabilities, even though they 
generate immediate cash inflows. As a result, financial analysts and investors relying on 
29 
period-specific earnings metrics may overlook the long-term profitability potential embedded 
within Tesla’s business model. 
However, this conservative approach also provides Tesla with strategic flexibility.  Since the 
total contract revenue remains unchanged and will ultimately be recognized over the life of 
the performance obligation. By deferring recognition, Tesla smooths out income fluctuations 
and avoids artificial earnings spikes tied to lump-sum sales. This is particularly valuable in an 
industry as capital-intensive and cyclical as automotive manufacturing, where cost shocks, 
delivery delays, or raw material shortages can significantly distort quarterly results. 
Ultimately, this figure underscores how Tesla's accounting strategy aligns with its broader 
financial narrative: sacrificing short-term earnings for long-term credibility and predictability. 
It also suggests that traditional profitability metrics must be interpreted carefully when 
applied to hybrid, software-integrated business models like Tesla's. 
 
 
Figure 9 Cumulative Deferred Revenue Additions vs. Recognized Revenue (2018–2023) 
Note. Data source: Tesla Inc. 10-K filings (2018-2023) 
Figure 9 illustrates the cumulative additions to Tesla’s deferred revenue (blue line) compared 
with the cumulative revenue recognized from deferred income (orange line) over the period 
2018 – 2023. The widening gap between the two lines indicates that while Tesla continues to 
30 
accumulate deferred revenue — largely from software-based features like Full Self-Driving 
— only a small portion is actually recognized as income each year. 
From an investor and analyst perspective, this accounting behavior introduces complexity. 
While Tesla’s reported profits may appear muted in the short term, the company is quietly 
accumulating value that will be realized in future periods. This raises important questions 
about how deferred revenue should be interpreted in assessing the true performance of 
technology-integrated automotive firms. Furthermore, this figure underscores how the rise of 
recurring and software-linked income streams is transforming traditional business models. 
Tesla is no longer just selling vehicles—it is creating a monetizable service platform, and its 
accounting strategy is adapting accordingly. For stakeholders, this reinforces the importance 
of looking beyond current-period earnings and considering the cumulative impact of deferred 
revenue on long-term financial health. 
 
 
 
Figure 10 Deferred Revenue as a Percentage of Total  Automotive Sales (2018–2023) 
Note. Data source: Tesla Inc. 10-K filings (2018-2023) 
Chart 10  summarises the relationship between deferred revenue (balance, additions, and 
recognized amounts) and Tesla’s total automotive sales, expressed as percentages. The 
31 
deferred revenue balance consistently represented around 4–6% of automotive sales from 
2018 to 2023, while recognized deferred revenue remained below 1% throughout the period. 
This indicates that most of the revenue from bundled software and service features is not 
immediately recognized but is spread over future periods in line with ASC 606. This aligns 
with the strategic use of deferred revenue to smooth earnings and reduce short-term volatility, 
as observed in firms leveraging discretionary revenue recognition (Stubben, 2010) and hybrid 
business models blending hardware with service-based offerings (Brynjolfsson & McAfee, 
2014; Deloitte, 2016). 
This figure has two key implications. First, it offers Tesla a mechanism to smooth its earnings 
over time, shielding the company from the full impact of short-term market fluctuations or 
delivery seasonality. Second, it indicates that Tesla’s profitability is likely understated in 
periods of high vehicle sales, as software revenue is largely deferred. Analysts and investors 
who rely solely on immediate profitability metrics may therefore underestimate the long-term 
value being created. 
In short, Figure 10 illustrates the accounting tension between revenue collection and 
recognition in software-integrated business models. It reaffirms the argument that ASC 606 
plays a crucial role in reshaping how financial performance is reported and perceived in 
hybrid product-service industries like electric vehicles. 
 
 
 
32 
 
Figure 11 Deferred Revenue as a Percentage of Automotive Gross Profit (2018–2023) 
Note. Data source: Tesla Inc. 10-K filings (2018-2023) 
Figure 11 demonstrates how deferred revenue components compare to Tesla’s automotive 
gross profit. In 2019, the deferred revenue balance accounted for over 40% of automotive 
gross profit, and even in 2023, the figure remained above 25%. This highlights the substantial 
influence of deferred revenue on Tesla’s profitability metrics. High-margin software income 
being deferred suggests that Tesla’s reported gross margins are temporarily understated, 
helping to create more stable earnings over time. 
This has important implications for both financial analysis and strategic accounting. On one 
hand, the deferral of high-margin revenue creates a conservative portrayal of profitability, 
possibly protecting the company from accusations of aggressive earnings management. On 
the other hand, it can distort short-term profitability assessments, leading analysts and 
investors to undervalue Tesla’s true operating performance, especially when comparing gross 
margins to peers that may recognize software revenue more aggressively or upfront. 
From an accounting standpoint, this phenomenon is a textbook example of ASC 606’s impact 
on modern business models. By requiring revenue to be recognized over time in alignment 
with performance obligations, the standard effectively flattens profit spikes and promotes 
long-term earnings consistency—at the cost of short-term profit visibility. In Tesla’s case, this 
33 
approach helps build a recurring revenue base while maintaining financial discipline and 
investor confidence. 
Though Tesla’s ASC 606 based revenue deferral approach is regulatory driven, it might be 
indicative of an intentional effort to manage reported earnings (smoothing) too. As Stubben 
(2010) shows, the discretion in revenue recognition is employed as an earnings management 
device, especially in high-growth or high-visibility firms. In Tesla’s instance, this will mean 
that not only do deferred revenues represent a non-performance obligation under a contract 
but that they also play a more strategic role in the company’s intention to smoothen earnings 
volatility and manage investor sentiments. 
4.4 Results of the Simulation Cases 
 
Figure 12 Profit Margin Over 15 Years - 0% Growth Scenario 
 
Figure13 Profit Margin Over 15 Years  - 10% Growth Scenario 
34 
In both the 0% and 10% growth scenario, we can see from figure 12 and 13,  Company A 
realizes more profit in first 5 year because of the immediate recognition of i's software 
revenue, with profit margins beginning at 31.6%. Earnings however decrease quite 
consistently thereafter, shrinking to about 25% by year five – indicating the risk of more 
aggressive revenue recognition. In contrast, Company B exhibits a stronger performance 
stability with a constant profit margin of 25% during the whole simulation, since income 
from deferred revenue is being released. 
From a cumulative perspective, Company A generates total profits of $232,500 in the 0% 
growth scenario and $686,229 under 10% growth. Meanwhile, Company B records slightly 
lower total profits—$222,500 and $630,382 respectively—but demonstrates far greater 
consistency in annual earnings. In terms of net present value, Company A reaches $370,255 
under 10% growth, while Company B achieves $337,239, suggesting that deferred revenue 
leads to only a modest trade-off in short-term profitability while delivering stronger earnings 
predictability and lower volatility. 
Note as well that despite its strong performance, company A does not exhibit the earned 
profit-margins stabilize by year five.The fact that one is tempted to think that earnings 
volatility could be shortlived (and that it is captured as such in the simulation) is biased by 
our static model assumptions; specifically the one-off software sale with no associated 
recurring updates or new features. These findings suggest that in real-world settings, where 
companies regularly roll out new software features, adopt subscription pricing, and operate 
with staggered delivery schedules, deferred revenue is likely to be generated and recognized 
in a continuous cycle. This creates a rolling effect in income recognition, helping to smooth 
earnings across both short and long time frames. From this perspective, deferred revenue 
under ASC 606 functions not only as a compliance mechanism, but as a structural strategy 
that contributes to earnings stability and supports investor confidence—particularly for firms 
built around software-integrated business models. 
This simulation outcome closely aligns with Tesla’s actual financial reporting practice. As 
shown in Figures 10 and 11, deferred software revenue at Tesla accounts for approximately 
4–6% of total automotive sales, yet less than 1% is recognized annually—demonstrating a 
conservative approach to revenue realization. This conservative approach is consistent with 
observations by PwC (2022) and Wagenhofer (2014), who highlight how deferred revenue 
recognition smooths out income volatility and offers greater predictability for long-term 
35 
planning. The widening gap between the inflow and outflow of deferred revenue indicates 
Tesla’s strategic use of ASC 606 to smooth earnings volatility. This approach, as emphasized 
by Deloitte (2016), is particularly valuable in capital-intensive and cyclical industries, where 
deferred revenue can buffer short-term financial volatility and present a stable outlook to 
investors. However, Healy and Palepu (2001) caution that such strategies may also lead to a 
temporary understatement of short-term profitability by deferring the recognition of 
high-margin software features, underscoring the tension between regulatory compliance and 
transparent financial reporting. 
 
 
 
Figure 14 Total Profit Over 15 Years -5% Growth Scenario 
In contrast, the -5% growth scenario more clearly highlights the risks associated with 
aggressive upfront recognition. From figure 14, Although software revenue is booked early 
on, declining vehicle sales lead to a sharp drop in profits over time. Company A’s total profit 
falls to $147,324, with an NPV of $101,268. Company B, however, maintains its stable 25% 
profit margin and generates a comparable total profit of $143,346, with an NPV of 
$95,978—demonstrating greater financial resilience during economic downturns. 
In conclusion, both the simulation results and Tesla’s real-world practice underscore that in 
software-integrated business models, the deferred revenue strategy mandated by ASC 606 not 
only contributes to earnings smoothing but also strengthens financial stability and investor 
36 
confidence. This is especially relevant when software sales are recurring or consistently 
renewed. For electric vehicle companies undergoing business model transformation, this 
strategy offers substantial long-term relevance. Brynjolfsson and McAfee (2014) have 
highlighted that the industry is increasingly shifting towards service-based, 
subscription-driven revenue models, a trend that Tesla’s approach exemplifies. 
4.4 Tesla as a Representative of Emerging Hybrid Models 
Tesla’s evolution over the last decade is part of a broader shift in the electric vehicle sector, 
away from a hardware-centric business model and toward one that is more hybrid, combining 
physical products with subscription-like services. As Brynjolfsson and McAfee (2014) assert, 
digital transformation has led firms in all sectors to reconsider value creation, something that 
can particularly be observed in the auto industry. For example companies like Tesla and NIO 
(NIO Inc., 2023) have moved beyond depending only on one-off vehicle sales. No, the next 
big aspect of the connected car race is vendors capitalising on the ongoing relationship, with 
software-driven capabilities around services like connectivity and driverless cars. "Trend 
makes the Rule" This trend demonstrates Goldman Sachs (2023) view that subscription 
revenue sources are key to hybrid vehicle valuations. 
The difference, however, is that while Tesla was a pioneer in the development of electric 
vehicles, its true innovation lies elsewhere. As Wagenhofer (2014) notes, value creation in 
modern firms increasingly hinges on intangible, service-based offerings. Tesla exemplifies 
this by leveraging features such as Full Self-Driving, premium connectivity, and over-the-air 
software upgrades—not just as technical tools, but as core components of a recurring revenue 
model. According to Tesla, Inc. (2023), these features are designed to extend the customer 
relationship beyond the point of sale, effectively turning each vehicle into a platform for 
continuous monetization. 
Financially, this appears in Tesla’s steadily surging deferred revenue balances, modest 
yet growing subscriptions income, and greater dependence on ASC 606’s rules around 
revenue deferral given performance obligations satisfied over time. Although 
software-derived revenues represent a small sliver of total income, their strategic importance 
is disproportionately great—both as a driver of margin expansion and a mechanism for 
stabilizing earnings. 
37 
Additionally, Tesla's model has since influenced other EV makers as well. Other 
manufacturers like NIO, XPeng, and Rivian are following a similar approach, combining 
hardware with digital services — be it Battery-as-a-Service (BaaS) or autonomous driving 
packs. This means that Tesla’s business model is no exception, but rather a leading 
example of an emerging industry standard. Such hybrid models are likely to become more 
common as the EV sector matures, bringing important challenges and expectations regarding 
transparency of financial reporting. 
To the end, Tesla provides a useful case study for how revenue recognition standards, and 
ASC 606 specifically, respond to changing commercial realities. The company’s disclosure 
practices highlight the challenges of allocating performance-based revenue using 
multi-component sales, as Healy and Palepu (2001) and Wagenhofer (2014) argue, financial 
reporting not only conveys performance but also serves as a strategic tool for signaling a 
firm’s market positioning and long-term vision. 
 
 
 
 
 
 
 
 
 
 
 
38 
5. Conclusions 
This study set out to examine how deferred revenue under ASC 606 influences profitability 
and earnings volatility in software-integrated electric vehicle business models, using Tesla as 
a case study. While existing literature provides technical discussions on revenue recognition 
standards and their implementation, it largely overlooks how Tesla strategically uses deferred 
revenue to manage earnings, smooth volatility, and shape investor perceptions. Few empirical 
studies have addressed the specific financial impact of deferred software income in hybrid 
product-service industries like EVs, where traditional accounting treatments struggle to 
capture the complexity of staggered service delivery and digital monetization. 
Additionally, the literature lacks detailed, data-driven analysis of how deferred revenue 
affects core profitability metrics—such as gross margin stability and reported earnings—in 
fast-growing firms that bundle high-margin software with physical products. This thesis 
contributes by filling these gaps through both empirical analysis and simulation modeling, 
offering evidence of how ASC 606 is applied not only as a compliance tool, but as a strategic 
mechanism in shaping financial narratives and investor confidence. 
One of the central insights of this paper is that deferred revenue, rather than a narrow 
accounting category, plays strategic role in the financial orientation of how Tesla approaches 
managing finances. When Tesla began to offer its customers various subscription-based 
software services, including Full Self-Driving, premium connectivity, and over-the-air 
updates, the company started to use more and more of deferred revenue to book the 
undelivered performance obligations under ASC 606. This practice of deferring income 
(actual return is deferred when revenue is deferred) has the effect of reducing income 
variability across the years. In a way, what it seems as though Tesla is doing is intentionally 
“hiding” profits in the short term, with deferred revenue functioning as a buffer against 
operational bumps or variations in vehicle delivery numbers. 
This is a particularly important smoothing effect for a fast-growing, 
capital-intensive corporation like Tesla. In quarters when deliveries were down or costs were 
up — because of supply chain disruptions, factory closures, or inflationary pressures — the 
release of previously deferred revenue worked to establish a “revenue floor” and blunted 
quarter-to-quarter volatility. Over time, predictability of earnings is improved and that can 
39 
make investors even happier, especially in a market that is paying more and more attention to 
forward looking growth and stability. 
There are also implications for gross profit margins based on Tesla’s reporting structure and 
the transparency of its financial reporting. ASC 606 places a focus on the identification of 
separate performance obligations and on the allocation of the transaction price of a contract 
to those performance obligations. Tesla's disclosures show a relatively consistent application 
of these rules, with packages—like cars sold with the right to access software that doesn't 
exist yet—broken down into revenue recognized at the point of delivery and revenue spread 
over the period with which they're associated. But according to real-world financial reporting, 
software income is typically financially blended into larger revenue categories (i.e., 
“Automotive Sales”), such that the true scale and impact of Tesla’s software-led margin 
expansion is somewhat of a black box. This calls into question whether these reporting 
practices necessarily align with the transparency objectives of ASC 606. 
Academically, the thesis adds to current discussions in accounting research about the 
implications of new revenue recognition standards on financial performance, investor 
perceptions, and managerial actions. By tying together theoretical lenses and empirical 
evidence, this paper underlines that accounting choices are not only compliance decisions 
but also means of strategic signaling and investor relations. The findings also contribute to 
the literature by suggesting that deferral of revenue mechanisms offer opportunity to further 
align economic activity with accounting results within service dominant, technology enabled 
industries. 
In conclusion, Tesla’s use of ASC 606 highlights not only the dynamic nature between 
business models and accounting, but also the conflict between innovation and static 
standards. The deferred revenue financial meme employed by Tesla is more than an 
accounting artifact, it’s a symptom of a more profound strategic kind—to manage earnings 
visibility (reducing volatility), to smooth out the buffeting of operating cash flows to a more 
constant dollar amount choice, and to make it easier to pitch the idea that long-term investors 
should like the stock. As more of these EV companies emerge, the mechanics of these 
practices — and how they influence perceptions of profit — is only going to become more 
essential for financial analysts, regulators and academic researchers alike to understand. For 
future research, this study's findings provide a useful basis for formulating hypotheses about 
the long-term impact of deferred revenue strategies on firm performance and valuation, as 
40 
well as potential interactions with emerging financial reporting frameworks in hybrid 
business models. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
41 
Reference 
 
Brynjolfsson, E., & McAfee, A. (2014). The second machine age: Work, progress, and 
prosperity in a time of brilliant technologies (Chapter 3 and 7). W. W. Norton & Company 
Caylor, M. L. (2010). Strategic revenue recognition to achieve earnings benchmarks: 
Evidence from the software industry. The Accounting Review, 85(2), 417–445. 
https://gu-se-primo.hosted.exlibrisgroup.com/primo-explore/fulldisplay?docid=TN_cdi_proq
uest_miscellaneous_37267380&context=PC&vid=46GUB_VU1&lang=en_US&search_scop
e=default_scope&adaptor=primo_central_multiple_fe&tab=default_tab&query=any,contains,
Strategic%20revenue%20recognition%20to%20achieve%20earnings%20expectations:%20E
vidence%20from%20the%20software%20industry.&offset=0 
 
Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1995). Detecting earnings management. The 
Accounting Review, 70(2), 193-225. 
https://www.jstor.org/stable/248303 
 
Deloitte. (2016). Revenue recognition by industry. Deloitte United States. 
https://www2.deloitte.com/content/dam/Deloitte/us/Documents/risk/us-aers-new-revenue-sta
ndard-issued-manufacturing.pdf 
 
Deloitte. (2023). Life sciences industry accounting guide: Revenue recognition. Deloitte 
United States.  
https://www2.deloitte.com/content/dam/Deloitte/us/Documents/risk/us-aers-new-revenue-sta
ndard-issued-life-sciences.pdf 
 
Eisenhardt, K. M. (1989). Building theories from case study research. Academy of 
Management Review, 14(4), 532–550.  
https://doi.org/10.5465/amr.1989.4308385 
 
Electrek. (2024, April 21). Tesla lowers price of 'Full Self-Driving' to $8,000, down from 
$12,000. 
42 
https://electrek.co/2024/04/21/tesla-lowers-price-of-full-self-driving-to-8000-down-from-120
00/ 
 
Ernst & Young. (2020). Applying IFRS: The revenue recognition standard – Automotive 
industry. EY Global.  
https://www.ey.com/en_gl/technical/ifrs-technical-resources/applying-ifrs-the-revenue-recogn
ition-standard-automotive-industry 
  
Financial Accounting Standards Board. (2016). ASU 2016-10: Revenue from contracts with 
customers (ASC 606). 
https://storage.fasb.org/ASU%202016-10.pdf?utm_source 
 
Goldman Sachs. (2023, July 6). Software is taking over the auto industry. Goldman Sachs. 
https://www.goldmansachs.com/insights/articles/software-is-taking-over-the-auto-industry 
 
Guiltinan, J. P. (1987). The price bundling of services: A normative framework. Journal of 
Marketing, 51(2), 74–85. 
https://www-jstor-org.ezproxy.ub.gu.se/stable/1251130?sid=primo&origin=crossref 
  
Gupta, S., & Lehmann, D. R. (2003). Customers as assets. Journal of Interactive Marketing, 
17(1), 9-24. 
https://gu-se-primo.hosted.exlibrisgroup.com/primo-explore/fulldisplay?docid=TN_cdi_scop
us_primary_2_s2_0_0347308160&context=PC&vid=46GUB_VU1&lang=en_US&search_sc
ope=default_scope&adaptor=primo_central_multiple_fe&tab=default_tab&query=any,contai
ns,Gupta,%20S.,%20%26%20Lehmann,%20D.%20R.%20(2003).%20Customers%20as%20a
ssets&sortby=rank 
 
Healy, P. M., & Palepu, K. G. (2001). Information asymmetry, corporate disclosure, and the 
capital markets: A review of the empirical disclosure literature. Journal of Accounting and 
Economics, 31(1-3), 405-440. 
https://www-sciencedirect-com.ezproxy.ub.gu.se/science/article/pii/S0165410101000180?via
%3Dihub 
 
43 
International Accounting Standards Board (IASB). (2022). IFRS 15: Revenue from Contracts 
with Customers. IFRS Foundation. 
https://www.ifrs.org/content/dam/ifrs/publications/pdf-standards/english/2022/issued/part-a/if
rs-15-revenue-from-contracts-with-customers.pdf?bypass=on&utm_source 
International Energy Agency. (2023). Global EV Outlook 2023. 
https://www.iea.org/reports/global-ev-outlook-2023  
Kaplan, R. S., & Atkinson, A. A. (2015). Advanced management accounting (Chapter 12: 
Pricing and Profitability Analysis). Pearson.  
 
Lev, B., & Thiagarajan, S. R. (1993). Fundamental information analysis. Journal of 
Accounting Research, 31(2), 190-215. 
https://www-jstor-org.ezproxy.ub.gu.se/tc/accept?origin=%2Fstable%2Fpdf%2F2491270&is
_image=False 
 
Lev, B., & Zarowin, P. (1999). The boundaries of financial reporting and how to extend them. 
Journal of Accounting Research, 37(2), 353–385.  
https://doi.org/10.2307/2491413 
 
NIO Inc. (2023). Annual Report 2023 
https://ir.nio.com/financials/annual-reports/ 
 
Not a Tesla App. (n.d.). Tesla FSD Cost and Price Increase History. 
https://www.notateslaapp.com/tesla-reference/958/tesla-fsd-price-increase-history 
 
Piosik, A. (2023). Revenue recognition in achieving consensus on analysts’ forecasts for 
revenue, operating income and net earnings: The role of implementing IFRS 15. Procedia 
Computer Science, 192, 4955–4964. 
https://doi.org/10.1016/j.procs.2021.08.160 
 
Polestar Automotive Holding UK PLC. (2023). Annual Report 2023.  
https://investors.polestar.com/static-files/b83ca36d-41b6-49f9-8735-9856ab21251a 
 
44 
PwC. (2022). Revenue from contracts with customers - 2022. PwC Viewpoint. 
https://viewpoint.pwc.com 
 
RoseRyan. (2014). Bundled products and revenue recognition. RoseRyan. 
https://roseryan.com 
 
Siggelkow, N. (2007). Persuasion with case studies. Academy of Management Journal, 50(1), 
20–24. https://doi.org/10.5465/amj.2007.24160882 
 
Stubben, S. R. (2010). Discretionary revenues as a measure of earnings management. The 
Accounting Review, 85(2), 695–717 
https://doi.org/10.2308/accr.2010.85.2.695 
 
Tesla, Inc. (2019). Tesla Q1 2019 Update. 
https://ir.tesla.com/#quarterly-disclosure 
 
Tesla, Inc. (2023). 10-K for the fiscal year ended December 31, 2023 
 https://ir.tesla.com/_flysystem/s3/sec/000162828024002390/tsla-20231231-gen.pdf 
 
Tesla, Inc. (2024a). Design your Model 3. Tesla.  
https://www.tesla.com/model3/design 
 
Tesla, Inc. (2024b). Full Self-Driving (FSD) subscription. Tesla. 
https://www.tesla.com/support/full-self-driving-subscriptions 
 
U.S. Bureau of Labor Statistics. (2024). Consumer Price Index: All urban consumers 
https://www.bls.gov/cpi/ 
 
 Wagenhofer, A. (2014). The role of revenue recognition in performance reporting. 
Accounting and Business Research, 44(4), 349–379.  
https://www.tandfonline.com/doi/abs/10.1080/00014788.2014.897867 
 
XPeng Inc. (2023). Annual Report.  
45 
https://ir.xiaopeng.com/news-releases/news-release-details/xpeng-files-2023-annual-report-fo
rm-20-f 
 
Yin, R. K. (2018). Case study research and applications: Design and methods (6th ed., 
Chapter 2: Designing Case Studies). Thousand Oaks, CA: Sage Publications. 
 
 
 
 
 
 
  
  
  
  
   
46