Several days ago, John Gulliver and Hal Scott announced on this Forum that “No, SPACs Do Not Dilute Investors” This comes after the Chancery Court has concluded otherwise in about a dozen cases, and after the SEC has issued regulations requiring disclosure of the extent to which value has been diluted or otherwise extracted from SPAC shares as of the time of a de-SPAC merger. Gulliver and Scott’s post on the Forum, and their longer version on SSRN, purport to “debunk” the analysis of two articles we published in the past three years, and in doing so, to show that the Chancery Court and the SEC are wrong as well—that the value of pre-merger SPAC shares is of no consequence and need not be disclosed.

To capture the essence of Gulliver and Scott’s story, imagine the following.  A company is planning a merger, and before closing, the company’s management, bankers and others suck out 50% of the company’s value for themselves.  They are not concerned that the company’s shareholders will lose out, however, because they believe the company’s merger partner will repay what was taken from them.  This is essentially Gulliver and Scott’s view of how SPACs and de-SPAC mergers work.  Of course, it is now well understood that SPAC shareholders have indeed lost out – losing over half of their investments on average for SPACs that merged between 2019 and 2021 – something that is contrary to Gulliver and Scott’s rosy predictions but is precisely in line with what we forecast in the fall of 2020 when we first released our initial paper (and our blog post on this Forum).

Gulliver and Scott recognize that the SPAC structure has substantial embedded costs. SPACs go public at $10 per unit, and when they merge, the SPAC attributes a value of $10 to its shares. But, in fact, as a SPAC enters into a “de-SPAC” merger, it holds far less than $10 per share in net cash.  In our first article, covering 2019 to 2020, we found that, on average, they had $5.70 per share in net cash to contribute to a merger. In a follow-up article covering a later period, we found that they had more cash but still far less than $10.

Gulliver and Scott do not disagree.  They nonetheless claim that despite contributing far less, SPAC shareholders will end up with shares worth $9.45 following the merger. The basis of their belief is a hypothetical example that they construct in which a target is eight times the value of the SPAC, and the SPAC’s costs are borne pro rata by the target and the SPAC shareholders.  That is, they assume that target shareholders bear 7/8 of SPAC costs and SPAC shareholders bear 1/8 of those costs—$0.55 in their hypo. They offer no basis for the pro rata cost sharing assumption, which drives the rest of their claims.  So, in short, they simply assume their conclusion that SPAC shareholders suffer little dilution.  Given that both our article, and that by Minmo Gahng, Jay R. Ritter, and Donghang Zhang find that SPAC costs are 2-3 times as large as IPO costs, one may naturally be skeptical that targets would agree to bear such large portions of SPACs’ costs.

In contrast, our original articles assumed, as a starting point, the proposition that SPAC shareholders can typically expect to receive about what they invest in a merger. If there is $6 in net cash underlying SPAC shares, after accounting for what SPAC sponsors and others have extracted, then SPAC shareholders can reasonably expect target shareholders to exchange about $6 in value for those shares. A Target would do so by inflating its value commensurately with the inflation of the SPAC’s value from $6 to $10 per share in the merger. The result will be that SPAC shareholders will hold post-merger shares worth about $6.  Of course, there may be situations in which SPAC managers get more from targets—especially when surplus value is created by the merger—but an even exchange is the reasonable place to start.

Gulliver and Scott’s next claim that the $0.55 cost that SPAC shareholders bear—5.5% of the $10 they forego by not redeeming their shares “is lower than [the cost of going public] for owners of private companies that go public through a traditional IPO, which is typically underpriced by 20%.” It is not worth sorting out all the apples and oranges being compared in this claim. But the two most glaring errors are the following: First, Gulliver and Scott apparently forget, that the SPAC shareholders’ cost in their hypo is just 1/8 of the total cost. The rest, they assume, will be paid by target shareholders. As stated above, the total cost of taking a company public through a SPAC is at least twice the cost of an IPO, on average.  Second, Gulliver and Scott have a math problem.  Their 5.5% cost of going public through a SPAC is based on a denominator that includes the entire value of the post-merger company.  But the 20% IPO cost applies only to the shares issued in an IPO—a much smaller denominator.  What Gulliver and Scott miss is that smaller denominators lead to larger fractions, and mixing denominators makes for an apples-to-oranges comparison.

Finally, Gulliver and Scott attempt to challenge our statistical findings that pre-merger net cash per share correlates with post-merger share price—again, the simple proposition that one tends to get what one pays for. In several regressions, using different models, we (unsurprisingly) found a strong correlation with post-merger share prices shortly after a merger and extending past one year following a merger. Gulliver and Scott challenge this finding only by cherry-picking one regression and mischaracterizing its result.

Gulliver and Scott do not question our main regressions. Instead, they pick one regression correlating pre-merger share price and post-merger stock price 18 months out using an incomplete dataset. We included this regression and some of the others as supplemental analysis to isolate the channels through which pre-merger dilution might work to reduce post-merger share prices. The coefficients were consistent with regressions covering shorter post-merger time periods and more complete datasets. But as we reported, while the results 12-months post-merger are statistically significant, the statistical significance fell below conventional levels after 18 months had elapsed post-merger.  This is not surprising, nor does it raise any doubt regarding our primary results. Over a period of 18 months, a lot of extraneous factors can influence stock prices and thus weaken a correlation through added noise.

Nonetheless, on the basis of this single regression Gulliver and Scott exclaim (in bold type!) that “the statistical significance of the net cash per share measure disappears – just the opposite of the point they were seeking to prove.” On this basis, they conclude: “[Net cash per share] is irrelevant as there is no statistically significant relationship between NCPS and post-merger performance.”  In other words, the amount of net cash a SPAC contributes to a merger is unrelated to the value SPAC shareholders ultimately will hold after the merger. The cash extracted from the SPAC before the merger reappears in shareholders’ pockets 18 months after the merger, despite it still being missing 6 months earlier.  We refer the reader to the paper we posted on SSRN with a more complete response to Gulliver and Scott for an explanation of the econometric principles involved here.

Our original analysis showed that pre-merger SPAC shares were worth far less than the $10 routinely attributed to them in de-SPAC mergers. The implication of that finding is that target shareholders would be expected to inflate their value similarly in response.  The result of this mutual inflation of value would be that post-merger SPACs would be worth far less than the $10 per share that SPAC sponsors and managers led their shareholders to expect.  And indeed, that is what happened.  As of today, the average share price of SPACs that merged between 2019 and 2021 is $4.13.  Only 14% of those SPACs are now trading above $10. If Gulliver and Scott believe that the low pre-merger value of SPAC shares was of no relevance to post-merger share value, how do they explain these post-merger share prices?  Bad luck?