Tag Archives: VAT fraud

Maximilian R. P. Gebhardt: The Emperor Has No Clothes

Maximilian R. P. Gebhardt is a former US diplomat and Economic Officer for the Department of State, now working as a consultant to private clients. From 2013 to 2015 he served as the economic officer at US Embassy Budapest responsible for covering trade and investment as well as tracking corruption.

♦ ♦ ♦

I served in Hungary from 2013 to 2015 doing counter-corruption, although a fair amount of my job was watching trade and investment, along with tracking sectoral changes. I’m proud to have contributed quite a bit to our 2014 and 2015 Investment Climate Statements. It is worth saying that I am more an analyst than an economist, and the data set that I generated is currently being looked at by folks far more experienced than I. My background is ultimately in international relations with a nice six-year stint at the State Department as a Foreign Service Officer where I was trained, though mostly I picked everything up on the job – the Foreign Service way.

Late last week, I was in discussions with a few old contacts from back when I was in Hungary. The discussion seemed pretty routine: the government was seeking to land the Olympics which would mean massive construction contracts, Hungary still showed positive growth signs and Moody’s had upgraded them out of junk status, and the national debt continued to shrink. The usual story of Hungary’s reliable but ultimately unimpressive 2-3% GDP growth. The on-the-ground picture was pretty much as I had left it as well. The rising cost of living was a constant gripe, wages were flat. Generally the complete opposite of the official inflation data that included utility price cuts.

And then Momentum succeeded in forcing a referendum, and the immediate reaction on the heels of that was for the head of the Central Bank to make a not-so-veiled claim that the United States attempted to remove FIDESZ and was behind Quaestor and all the failures of Q1 2015. I admit, it was a pretty funny story. I built the case for those visa bans, and I can say with certainty that there was no plan for a coup. They were at best a shot across the bow.

But it raised an interesting question: “What are you hiding that has you so worried?”

Back in late 2014, when I was tracking agricultural VAT fraud, we got a tip from Ferenc Biró at Ernst and Young that he had tracked some limited food oil VAT fraud for a client of theirs by looking at discrepancies in the trade data. Hungary might cook the books a bit, but it does not live in a vacuum, and Hungary is certainly trade dependent, with exports and imports combining to well over 150% of GDP. Fortunately, trade is classic double-entry bookkeeping. An export logged by one country has a corresponding import recorded on the other side. As the Hungarian Tax and Customs Authority both logs the official trade data and collects taxes like VAT, it stands to reason that anyone cheating on VAT wouldn’t want NAV to know, so there would be no record of the transaction.

I  slogged through bulk Global Trade Atlas data for a dozen countries so that I could prove a point. In the end, I did find anomalies in food oil, oilseed, and some other agricultural commodities in trade with Slovakia, Ukraine, Serbia, and Croatia. It was fun, it told us we were right to worry, but ultimately it was an internal exercise in confirming what we had already confirmed with multiple sources regarding agricultural VAT fraud.

Fast forward, and I asked myself that same question – what about now? The food oil market, after all was said and done, was supposedly cleaned up. Mission accomplished, America won. I pulled out a few million data points from UNCOMTRADE data and asked the question again: “Are there gaps in the data?”

I should first say for American readers that VAT is a type of tax that is not unlike sales tax, except it is assessed on the “value added” at each transaction along the value chain. In Hungary, VAT is 27%, the highest in the world, so there is quite an incentive to cheat. When you export from one country to another, the tax authority in the exporting country refunds the VAT to the exporter, allowing trade to happen at the actual price of the good, rather than the inflated price with VAT. In trade, there are two kinds of VAT fraud you typically catch: import-based and export-based. Import-based VAT fraud rests on a simple principle. When a good is imported into country A, you are liable to country A’s tax authority for VAT on that import. Export-based is a bit more complex. You claim you exported a good to the tax authority, pocket the VAT refund, and then sell the good domestically at or below market price, pocketing the VAT. Food oil and agricultural VAT fraud was typically of the export variety, drawing lots of criminal participants since they realized you could keep claiming exports in a loop, pocketing refund after refund. For a time, that kind of VAT fraud really tied up the oilseed market – why sell off your oilseed when you can use it to keep pulling off VAT fraud ad infinitum?

I did a few initial case studies – automotive, aviation, agriculture, electronics, and petroleum products. In the case of aviation, automotive, and electronics, I found some anomalous reported exports to Hungary that were not reported as imports. In the case of agriculture and petroleum, I found export-based issues with Hungarian exports that had no corresponding imports recorded.

The total impact of this was huge, about $3.37 billion in errors in 2015 alone, which was large enough to impact GDP.

I’m still looking at the data, and several other economists are as well, so these findings are certainly not conclusive, but as my favorite nerdy webcomic XKCD once said, “Correlation doesn’t imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing ‘look over there.’”

February 28, 2017