Global Online Fraud Losses = 140,000 Jobs by Fabrice Colas, Ubivar

Over the past 10 years, the volume of global fraud losses for online payments has increased at an average rate of 10% each year. As of 2014, the global losses due to fraud are expected to reach the astounding amount of $14bn, which is equivalent of 140,000 jobs.

In what follows, we first describe the two factors that explain the exponential growth of online fraud over the past ten years, and then we review the fraud losses faced by an average US company, hypothesizing on the factors that increase and that decrease fraud rates for companies.

1. Why do global fraud losses keep on growing exponentially?

The historical trend indicates that the proportion of retail sales made online has grown from 0.8% in 2000 to 6.4% in 2014-Q1 in the US (expected). This represents a 17% year-to-year average growth over that period. Still and as mentioned earlier, the average growth rate of global fraud losses was of 10% (worldwide) which means that, taking the US as proxy for the growth of worldwide online retailing, online sales have grown 70% faster than fraud losses.

Therefore, we explain the growth of global fraud losses by the growing share of retail sales that are made online over that period. Still, the fact that fraud losses have grown 70% slower than online sales underlines the effectiveness of the efforts to mitigate fraud risk.

colas1

2. On-line retail conversion, and international, two factors catalyzing the growth of fraud

We expect global fraud to keep on growing in the coming years for two reasons.

While the US are leaders in e-commerce, only 6% of the sales are made online and 94% still derive from classical retail channels like Walmart and Walgreens. Hence, the process of converting sales online is still ongoing and should not stop anytime soon which, in turn, will fuel the growth of global fraud losses.
Because the US are the front runners in online retailing, we expect worldwide retailing to follow the path taken by the US with some lag. At this moment, online sales should still represent under 6%, which means that there is also room to convert sales globally, which will sustain the increase of global fraud losses…

3. Better fraud mitigation leads to a 3x drop of fraud in 10Y

In average in 2001, an American company selling online lost one dollar to fraud for every twenty dollars sold. In 2012, that figure dropped to one in 60, 3x less. Yet, in the US, it still represents about 1.6% of the total online revenue, and fraud losses amount to about $4bn per year, i.e. roughly 40,000 jobs…

colas2

Risk varies per sector

Previous figures reflect the fraud rates that an average US company faces. However, exposure to fraud risk depends heavily on the market vertical that the company operates in. Is it an airline company, a travel agent, a marketplace, a video shop, a payment service provider?

If markup is very low like for online travel agents, then frauds needs to be compressed to even lower rates than those of the “average US company”. Taking the margins seen by online travel agents, one single fraud requires to sell 50 to 100 airline tickets, just to recover that loss. Comparatively, a merchant making markup with two-digits on each sale has a lower incentive to mitigate its fraud risk because it only requires a few sales to recover one loss.

Risk varies per company size

Company size also influences the risk of fraud. Is it a small business, a medium-sized company, a large corp like Amazon, Apple or Paypal?

Expectedly, large American corporations are expected to run expensive but efficient fraud mitigation departments that deploy a full range of fraud management techniques: from automated screening methods to manual review. Hence, these corporations are likely to register a burden of fraud on their balance sheet that is lower than for an average company.

In the opposite, small and medium companies (SMBs) have neither the expertise nor the resources to implement a full-range department that manages fraud, because this is not their core business. Hence, they are more likely to use ad hoc solutions whose performance depends on scarce fund availability. Even more, their small size make it difficult or impossible to keep a core of competencies in-house. Consequently, SMBs are likely to be the online retail players paying the highest toll to fraud.

Summary

We first described the factors that explain the exponential growth of online fraud over the past ten years, i.e. the inherent growth of online retailing and the development of online retailing in the rest of the world.

Then, we argued that there is no reason why the 10% average growth rate should undergo an inflection since the underlying factors catalyzing this growth are expected to remain.

Finally, we illustrated that in the past 10 years, fraud mitigation techniques improved drastically, leading to a three fold decrease of the burden of fraud on online retailing (in the US).

References

US$ are kept constant as of 2014, and data sources are

– the US Census,
– the Federal Reserve
– CyberSource.

About Ubivar

Ubivar saves time and money for e-commerce websites by taking over the hassle of screening e-payments for frauds with its risk analytics API.

Our analytics engine routes e-payments automatically by their risk for acceptance, manual verification or rejection, with the aim of boosting payment conversion, speeding-up average delivery time whilst controlling risk.

www • Email • blog blog • API docs • twitter