• Retail sales in the United States are expected to contract slightly in May.
• Correlation analysis does not reveal a strong relationship between the initial market reaction and the deviation.
• The USD / JPY has a strong negative correlation with the surprises in the sales data four hours after the release.

The US Census Bureau will release the May Retail Sales Report on Tuesday, June 15. After remaining unchanged at $ 619.9 billion in April, retail sales are expected to decline 0.4% on a monthly basis.

The general market opinion is that a positive surprise in the retail sales reading is likely to give the greenback a boost and vice versa. In order to understand the impact of this data on the performance of the USD against its rivals, we analyzed the market reaction of four major pairs, EUR / USD, GBP / USD, USD / JPY and AUD / USD, to the previous 11 retail sales impressions. .

Methodology

The FXStreet Economic Calendar assigns a deviation point to each macroeconomic data release to show the extent of the divergence between the actual impression and the market consensus. Although retail sales were unchanged in April, the spread was -0.37 as analysts expected a 1% increase.

---

---

On the other hand, the 7.6% increase recorded in January from the analysts’ estimate of 1.1% was a relatively large positive surprise, with the spread showing 5.82 for this particular publication.

“Deviation: It measures the surprise caused when the actual data differs from the consensus. It can oscillate in an open scale, usually between -7 and +7.

Next, we plotted the reaction, in terms of pips, of the above mentioned major pairs 15 minutes, one hour and four hours after the release to see if the general market opinion held.

Finally, we calculated the correlation coefficient (r) to determine which major pair had the strongest correlation at which point in time. When r approaches -1, this suggests that there is a significant inverse correlation, while a significant positive correlation is identified when r goes to 1.

Results

There were seven negative surprises and four positive surprises in the previous 11 releases. On average, the spread was -0.52 on disappointing draws and 2.08 on bullish numbers. 15 minutes after a negative surprise, the average gains in EUR / USD and GBP / USD were 1.4, 3.5 pips respectively, while USD / JPY and AUD / USD lost 3.5 pips and 0.4 pips, respectively. On the other hand, EUR / USD and USD / JPY lost 3.2 and 1.2 pips, respectively, on average following positive surprises while GBP / USD and AUD / USD lost. gained 0.75 and 0.25 pips, respectively. This result suggests that the initial market reaction to the retail sales data is largely muted.

15 minute gap table

60 minute deviation table

240 minute deviation table

EUR / USD

EUR / USD has a correlation coefficient of -0.57, -0.24 and -0.01 15 minutes, 60 minutes and 240 minutes after the release, respectively. These figures suggest that the inverse correlation between the retail sales spread and the EUR / USD action is virtually nonexistent an hour and four hours after the release. Although the initial reaction is somewhat correlated, the average pip change is too small to consider it a trading opportunity.

GBP / USD

The GBP / USD has a correlation coefficient of -0.50, -0.34 and -0.18 15 minutes, 60 minutes and 240 minutes after the release, respectively. Similar to EUR / USD, the inverse correlation of the change in the GBP / USD pip to the retail sales spread gradually weakens and disappears four hours after the release.

USD / JPY

The USD / JPY has a correlation coefficient of 0.18, -0.52 and -0.70 15 minutes, 60 minutes and 240 minutes after the release, respectively. The pair’s initial reaction to retail sales data is largely unpredictable. However, the correlation reverses one hour after release and becomes significant by the fourth hour. This finding suggests that if the impression of retail sales is better than expected, the USD / JPY falls and vice versa.

AUD / USD

AUD / USD has a correlation coefficient of -0.2, -0.04, and 0.003 15 minutes, 60 minutes, and 240 minutes after release, respectively, making it the least correlated pair, out of the four tested , with the surprises of retail sales in the United States.

summary

When there is a negative or positive surprise in US retail sales data, the initial market reaction is insignificant. There is no clear relationship between the movements of the EUR / USD, GBP / USD and AUD / USD pairs and the deviation one hour and four hours after the release. This could be because retail sales are affected by temporary factors, such as weather conditions and holiday periods, making it an unreliable leading indicator.

The most interesting result is the strong inverse correlation with the USD / JPY and the retail sales spread four hours after the release. Nonetheless, the average pip change over this period is 5.5 and -6.7 pips on negative and positive spread, respectively, not leaving enough room for a trading opportunity.

---